Bifurcation, quasi-periodic dynamics, chaos, and soliton waves in the van der Waals normal form for fluidized granular matter Authors: Md Nur Alam, Muhammad Ziaul Haque, Shams Forruque Ahmed, Md Nurul Islam, Kok Swee Sim Published: 2026 ABSTRACT AbstractThe fluidized granular media is a complex phenomenon with sharp regime changes, pattern formation, and coherent density-wave propagation that is usually treated by nonlinear evolution equations, which are reduced. Of these, the van der Waals normal form, a fourth-order nonlinear oscillator with viscous/frictional dissipation and cubic nonlinearity, offers a general model with which to investigate such phenomena. Although there has been a considerable amount of effort on the construction of precise travelling-wave solutions to van der Waals-type models, in past work, systematic comparisons of the auxiliary-equation choices and their effects on admissible parameter regimes, as well as combining solution construction with an extensive dynamical-systems viewpoint are commonly not made. This study closes such gaps by creating a unified travelling-wave solution, based on Bernoulli-type and Riccati-type auxiliary equations, and an explicit parameterized family of wave solutions, consisting of kink-type, bright-dark soliton, and complex-valued solitons. They are then connected to a normalized nonlinear oscillator to analyze phase-space structures, stability, bifurcations, and sensitivity to perturbations. Symmetry-driven pitchfork bifurcations, dissipation-driven transitions, and an abundance of periodic, quasi-periodic, and chaotic behavior become apparent as key results through equilibrium analysis, 2D/3D phase portraits, time series, and Poincaré diagnostics. The high-dimensional parameter space and model idealism are drawbacks. However, analytical wave solutions coupled with dynamical diagnostics offer strong benchmarks to the numerical dynamics of the propagation of density waves in fluidized granular systems and have been used in augmenting predictive insights about their propagation. Both the theoretical and practical research in the areas of granular flow, material processing, and pattern formation have important implications for these findings. In contrast to previous works, which use one framework for auxiliary equations, a systematic comparison of four ansatz schemes within one common Bernoulli–Riccati framework. The analytical solutions are additionally connected to the dynamics of nonlinear oscillators via the tools of equilibria, bifurcation, and chaos diagnostics, and a direct link is established between exact travelling waves and the structures of phase space.Keywords: Exact travelling-wave solutions, Bernoulli auxiliary equation, Riccati auxiliary equation, Nonlinear oscillator dynamics, Dynamical systems, Phase-space structures, Poincaré diagnosticsSubject terms: Mathematics and computing, Physics Introduction Granular materials are macroscopic particles (i.e., large enough to be seen with the naked eye), aggregates of particles (powders, grains, pellets, or sand, etc.), which are central to industrial processes, like storage, mixing, and granulation, and to many geophysical processes. Their predictive models are still difficult due to the fact that the granular assemblies may act as a quasi-solid, dense flowing continuum, or even dilute collisional gases, depending on confinement, agitation, and forcing. According to recent syntheses, simple models can be effective when based on combining continuum reasoning with particle-size scale information, as opposed to being based on one theoretical level. In the case of dense granular media, Kamrin et al.1 offered a contemporary outlook on constitutive modeling that has developed out of the primitive rheologies of the early 1990s to nonlocal and reduced-order modeling, capturing creeping regions, size effects, and necessitating macro predictions to be tied to grain physics. Computational fluidization-based methods like coupling computational fluid dynamics with the discrete element method (CFD-DEM) have become useful in fluidization-dominated regimes, in which particle-fluid interactions dominate. Zhao et al.2 survey recent developments and current issues in the field of CFD-DEM simulation of fluidized bed and remark significant improvements in the field and the price of the calculations that encourage the use of reduced-modeling approaches. Fluidized and vibrated granular systems are an ongoing research field since rich and pattern-forming behaviors can be induced by external driving. Watson et al.3 presented a summary of the main findings on vibrated granular matter and transport, and one of the main aspects of these phenomena is the variation of dynamics with parameters over the different regimes. The second common theme in continuum descriptions is nonlocality: the stresses and shear rates at any point are often based on the material state of the surrounding material, especially around yielding or heterogeneous shear interactions. Thus, Fazelpour et al.4 experimentally showed that nonlocal rheology could be used to model quasi-2D annular shear flows, but the nonlocal parameters used must be dependent on the shape of the grain and material, and continuum closures must be interpreted with respect to particle properties. In addition to first-order nonlocal models, Kim and Kamrin5 proposed a second-order nonlocal constitutive model that uses DEM data to model the stress anisotropies and the secondary flows. At a small scale, Dumont et al.6 used DEM simulations to obtain scaling relations of dense granular flows at inclined planes with a combination of macroscopic laws and transiently organized microstructure. Taken together, these works highlight a practical viewpoint: in cases where full multiscale prediction is intractable, the reduced models are useful; however, their interpretation should be informed by the diagnostics that elucidate the qualitative regimes that surround a particular parameter set. In this context, reduced nonlinear evolution equations are commonly used to isolate mechanism-driven behaviours that include clustering, phase-separation-like transitions, and density wave propagating waves. The van der Waals normal form (VDWNF) of fluidized granular matter is a fourth-order nonlinear dissipative nonlinear cubic nonlinear model that is an analytically treatable testbed. It captures sharp spatial fronts, local structures, and pattern-forming dynamics whilst subject to travelling-wave reduction. Between 2021 and 2025, there has again sparked interest in building specific travelling-wave solutions to the VDWNF and related van der Waals-type models using a variety of algebraic-analytic methods. Tools integrability based on, e.g., Painlevé analysis, Bäcklund transformations, tanh-function construction, and elliptic-function representation are combined by Morad et al. 7 to interpret the obtained waves in terms of bubble-like interactions and phase-portrait dynamics. Zafar et al.8 used either the simplest equation and Exp-function techniques to produce new solitary as well as periodic travelling waves with graphical demonstrations. Zhu et al.9 introduced a general auxiliary-equation approach based on the Riccati, where families of solitary waves of the VDWNF are obtained. Li et al.10 generalized the model to a time-fractional version of the van der Waals model, and developed analytic solutions in a truncated version of the M-fractional derivative formulation, where the generalization of exact-solution methods to generalized operators is emphasized. The S-expansion technique is modified by Parvin et al.11 that highlights the applicability of kink, bright, and dark wave structures to the phenomenon of granular waves propagation and clustering. Lastly, Butt et al.12 offered a comparative analysis of solitary waves of van der Waals equations, clearly putting their study in the context of method/operator choice and the morphology of the solutions. Taken together, these works highlight the ongoing development of methods of analysis of the VDWNF, which are both to show the variety of the specific modes of construction of exact waves and to show the increasing interest in relating forms of solutions to dynamical-system representations of the dynamics of a granular media. Methodologically, auxiliary-equation methods are of particular interest since they give us a systematic path between a nonlinear PDE and a system of parameter constraints that is algebraic. The usual algorithmic steps are: (i) travelling-wave reduction, (ii) pick a finite ansatz which is an auxiliary function satisfying a first-order sub-equation, (iii) balancing nonlinear and derivative terms to choose the degree of the ansatz, and (iv) solving the resultant algebraic system to give explicit parameter constraints. The modern trends focus more and more on verification based on the symbolic substitution and graphical interpretation as a part and parcel of the whole workflow. As an example, Mehdi et al.13 used Riccati-Bernoulli and generalised Bernoulli sub-ODE methods to build up precise travelling waves and complement their results with phase-portrait studies. Sirisubtawee et al.14 introduced a free-access auxiliary-equation model and noted that in modified procedures, the Bernoulli-type sub-equations occur naturally. In another sense, Wu et al.15 applied methods of extended Riccati equations to pulse propagation models and showed the ongoing relevance of Riccati-based transformations in computing analytic wave families. Taken together, these studies encourage the head-on comparison of the test of auxiliary-equations options on the same reduced ODE, and not in isolation. No less important, explicit travelling-wave solutions are not sufficient to describe the underlying dynamics. When travelling-wave reduction occurs, the Van der Pol-type Nonlinear Function (VDWNF) decreases into a nonlinear oscillator with a phase portrait that structures admissible trajectories and explains equilibrium structures as well as stability properties. Modern granular simulations focus on the fact that regime boundaries, e.g. smooth transitions between jamming and flow, are associated with qualitative stability and constitutive behavior changes. Rubin et al.16 suggested a continuum model that describes the transition between elastic jamming and fast inelastic flow, which shows that constitutive structure dominates the formation of a new regime. Nicot et al.17 discussed bifurcation modes and unjamming phenomena from the viewpoint of configurational mechanisms. In applied dynamical systems, equilibria, stability, and bifurcations are described systematically by the formalism of these phenomena; a revised edition by Kuznetsov is a classic reference of the bifurcation theory18, and Ahsan et al. showed how continuation platforms, including COCO, create a platform that allows parameter exploration of nonlinear dynamical systems in a reproducible way19. In cases where external perturbations are introduced, such as an external forcing of the environment, control inputs, or a parameter modulation, classical nonlinear-dynamics diagnostics are popular in order to make the distinction between periodic, quasi-periodic, and chaotic regimes. Here, a few complementary analysis tools have been found useful in detection of qualitative changes in dynamical behavior. As an example, Urenda-Cázares et al.20 used bifurcation diagrams, Lyapunov exponents, phase portraits, and Poincaré maps to study regimes transitions in Duffing-type systems when stochastically perturbed. Their work brings into focus the fact that several diagnostic measures enable a concurrent description of the dynamical complexity of driven nonlinear oscillators. Simultaneously, methodological advances are still made to enhance the reliability of these diagnostics. According to Zhou et al.21, the interpretation of Poincaré sections is subjective in automated or large-scale analysis, and suggests entropy-based improvements that do not lose the geometric intuition of Poincaré mapping but enhance a quantitative distinction. On a comparable note, Wang et al.22 studied bifurcation structures and paths to chaos in a time-delayed, fractional-order Duffing system with harmonic excitation (using harmonic excitation) showing how the variation of the parameter can trigger period-doubling cascades and periodic-chaotic alternating regimes. To characterize the chaotic dynamics quantitatively, the Lyapunov exponents have been a fundamental diagnostic. Dabrowski et al.23 talked about the other computational plans of the largest Lyapunov exponent and explain how they apply to nonlinear oscillators, such as Duffing systems. Escot et al.24 also discussed the more practical issue of computing Lyapunov exponents under noise and recommends Jacobian-based indirect algorithms which enhance robustness in experimental and numerical contexts. Besides these methods, recurrence-based methods have become a formidable family of nonlinear time series analysis techniques, which can be used to identify complex dynamical structures. Recent advances in this direction are summarized in the proceedings of Hirata et al.25, which emphasizes the rising popularity of analysis of recurrence to identify state changes between regular and chaotic stages. A combination of these tools offers a generalized model of the diagnosis of perturbed nonlinear system dynamics. Recent advances have also broadened the use of nonlinear wave analysis beyond the conventional auxiliary-equation techniques. The dynamics of lump, breather, M-shaped, and optical soliton structures, as well as the stochastic and noise-driven nonlinear systems, have been the subject of growing interest26–31. The work of these studies highlights the usefulness of the hybrid modelling approach of using exact solutions, numerical simulations, stochastic modelling, and dynamical diagnostics to understand the behaviour of complex nonlinear wave phenomena. A systematic study of multiple auxiliary-equation models and their relevance to bifurcation and phase-space structures in models of granular evolution of the van der Waals type, however, is still largely underdeveloped. Within the last few years, analytical techniques using artificial intelligence have appeared as a new approach to the study of nonlinear differential equations and complex dynamical systems. The latter include the use of artificial intelligence for mathematical derivation, bilinear neural network approaches, fractional sub-equation neural networks, bilinear residual networks, neural network-based symbolic computation, and multimodal neurosymbolic reasoning frameworks. These techniques show good ability in symbolic discovery and pattern recognition, but are not widely applicable to models of granular evolution that are of the van der Waals type. The ability to use artificial intelligence-enhanced mathematical derivation (AIEMD) techniques to merge symbolic reasoning with machine learning to automate analytical derivations and mathematical discovery has been shown32. Analytical representations of nonlinear fractional differential equations have been proposed using neural network guided symbolic structures, called Fractional Sub-Equation Neural Networks (fSENNs)33. Recently, Bilinear Neural Network (BNN) methods and Bilinear Residual Network (BRN) frameworks have further extended the capability of neural architectures to capture nonlinear wave interactions and obtain exact solutions of nonlinear evolution Eqs.34,35. In parallel, multi-modal neurosymbolic reasoning algorithms, and neural-network-based symbolic calculation approaches combine data-driven learning with symbolic mathematical reasoning to afford novel possibilities for automated theorem discovery, symbolic calculation, and analysis of nonlinear models36,37. The techniques here are new examples of how artificial intelligence and nonlinear mathematical analysis are merging. While they have shown potential in symbolic discovery and construction of solution, the adaptation to the granular evolution models of van der Waals type has not received much attention so far. Therefore, precise analytical frameworks are still necessary to develop benchmark solutions, to check computation-based methods, and to give a strict dynamical interpretation, which is what this is all about. Notwithstanding these progresses, this recent literature still has a number of gaps when addressing the van der Waals normal form (VDWNF). Most of the literature has focused on the creation of large compendia of travelling-wave solutions, but has not synthesized the findings. Specifically, it is commonly not clear which solutions describe actually distinct wave structures, and which are due to re-naming of parameters or similar transformations. Moreover, parameter regimes that ensure the presence of physically meaningful, real valued profiles are not necessarily systematically discovered. A second issue of significance is the effect of the selected framework of auxiliary equations: various constructions of the ansatz often result in various sets of admissible parameter roots found by matching coefficients, and these sets are hardly ever compared within a single framework. Furthermore, precise travelling-wave solutions are frequently given without any reference to the dynamical-systems interpretation of the reduced travelling-wave oscillator. Therefore, the connection between the analytical wave structures and underlying phase-space structure, including equilibrium structures, stability, and bifurcation structures, has not been adequately understood. Likewise, in a study where the dynamics are forced or perturbed, standard diagnostic means like phase portraits or Poincaré sections are usually introduced, but not linked explicitly to the equilibrium structure of the underlying unforced system upon which the dynamics is organized. In order to overcome these shortcomings, the current research has a number of objectives. First, we form and hierarchically arrange explicit travelling-wave families of the VDWNF in terms of a single auxiliary-equation framework in the form of four intimately related classes of ansatzes which are founded on the use of Bernoulli-type and Riccati-type sub-equations. Second, we give a comparative study of the ensuing parameter root, and how varying auxiliary-equations choices can affect admissible families in solutions. Third, the travelling-wave reduction can be understood as a normalized nonlinear oscillator, in which equilibrium structures, stability features and bifurcation behavior, including the symmetry-driven pitchfork bifurcation of the cubic nonlinearity, can be studied in detail. Lastly, representative perturbations are provided to demonstrate how stroboscopic Poincaré diagnostics can be used to differentiate between periodic and multi-periodic responses of the resulting non-autonomous system. The resulting analytical-dynamical complex offers a better and more organized explanation of the VDWNF. Besides providing a standard to analyze the precise exact-solution methods and numerical approximators in nonlinear PDEs, the findings provide useful information to understand and regulate the dynamics of density waves and fronts in fluidized granular materials and associated pattern-forming systems. Novelty and main contributions The present study builds on the literature on evolutionary models of van der Waals-type granular media in the following important respects. First, we develop a unified framework for four-ansatz analytical approaches to systematically construct and compare exact traveling-wave solutions within a single analytical framework, using both Bernoulli-type and Riccati-type auxiliary equations. Second, different auxiliary-equation formulations’ impact on the types of permissible parameter constraints and wave structures is examined, offering a comparison that is not done extensively in prior works. Third, explicit traveling-wave families, such as kink-type fronts, pulse-type structures, bright-dark solitons, and complex-valued wave branches, are derived and classified. Fourth, the travelling-wave reduction is reformulated as a nonlinear oscillator, making it possible to study the equilibrium structures, stability properties, phase-space organization, bifurcation transitions, sensitivity characteristics, and chaotic dynamics within a common framework. Lastly, a direct correspondence is set up between exact analytical wave solutions and the dynamics of the underlying dynamical system, thus gaining a fuller picture of the propagation of nonlinear waves in fluidized granular matter. Problem formulation A sophisticated framework that expands on the traditional van der Waals equation to simulate the complex behavior of granular flows is the van der Waals normal form applied to fluidized granular materials. Phase transitions, density changes, and particle clustering within these systems are among the key characteristics it records. This model was not created by a single person, but rather as a result of ongoing advancements in hydrodynamic theories and granular physics. It offers significant insights into the mechanics of fluidized granular materials by equating it with liquid-gas phase behavior in typical fluids. It is also a reliable tool for theoretical research and real-world applications38. Through nonlinear partial differential equations (NLPDEs), the van der Waals normal form provides insights into crucial characteristics like phase separation, jamming, and wave propagation in granular media. It is extensively used in industrial processes, geophysical flow studies, and materials science. We examine the macroscopic energy, momentum, and density balance equations38: where, Moreover, the new constitutive equation for the energy dissipation rate flux When the velocity field is low enough that nonlinear terms in u, density and horizontal momentum in the hydrodynamic equations, Eq. (2), does conserve, these values may subsequently be vertically averaged to provide and, and these quantities do conserve the continuity Eq. Equation (3) is approximated at the dominant order by the van der Waals form where : scalar field (order parameter of the reduced model), : dissipation coefficient, : control parameter, k: travelling-wave wavenumber, : travelling-wave frequency/speed parameter (real parameter), : travelling-wave coordinate and Prime : derivative with respect to . Methodology The travelling-wave reduction, balancing procedures, ansatz construction and derivation of parameter constraints have been done analytically. Algebraic systems were symbolically verified and exact solutions substituted into them using Mathematica/Maple. Let us take the NLPDE of the form, where P is a polynomial of and its partial derivatives and is an unknown function of x and, which involves the highest degree nonlinear terms and the maximum number of derivatives. We introduce a compound variable with respect to the real variable x and t , The travelling wave transformation (6) allow us to reduce Eq. (5) into an ordinary differential equation (ODE) for of the form: where Q is a polynomial in and its derivatives and the prime specify the derivative with respect to Ansatz-I Suppose that the solution of the Eq. (7) can be expressed as the following Ansatz equation, Where are unknowns and . A new function satisfies the following ODE, in which are constant to be determined later, and . Then Eq. (9) is the type of Bernoulli equation, and we can obtain the solution as, Ansatz-II Suppose that the solution of the Eq. (7) can be expressed as the following Ansatz equation, Here are unknowns and A new function satisfies the following ODE, where is a real constant parameter. Equation (13) has solutions for different cases of . Case 1: if Case 2: if Ansatz- III Suppose that the solution of the Eq. (7) can be expressed as the following series, Here are unknowns and . A new function satisfies the following ODE, In which are constant to be determined later, and . Then Eq. (25) is the type of Bernoulli equation, and we can obtain the solution as, Ansatz-IV Suppose that the solution of the Eq. (7) can be expressed as the following series, Here are unknowns and . A new function h= satisfies the following ODE, where is a real constant parameter. Equation (29) has solutions for different cases of . Case 1: if , Case 2: if , Advantages and limitations of the unified four-ansatz framework A large variety of analytical methods have been devised for nonlinear evolution equations, but they all have their own merits and demerits. The Hirota bilinear method is very useful for generating multi-soliton solutions, and is usually used in conjunction with a proper bilinear representation of the basic equation. Sardar sub-equation approaches offer an efficient way to construct exact solutions, but are restricted to using one auxiliary-equation structure. More recently, Bilinear Neural Network (BNN) methods, Bilinear Residual Network (BRN) frameworks, and other artificial-intelligence-aided symbolic methodologies have shown great potential to find exact solutions and complex wave structures by means of symbolically computing via a neural network26–31,39–42. Most of these approaches, however, have been developed for the purpose of solution discovery and are not typically capable of a thorough and systematic comparison of several different formulations of the auxiliary equation in a single framework. The present study, on the contrary, presents a unified four-ansatz algorithm, which at the same time includes both Bernoulli-type and Riccati-type auxiliary equations, enabling a direct comparison of the constraints on the parameters, the families of solutions, and the corresponding dynamical interpretations. This feature offers a more comprehensive analytical view and allows for the seamless combination of the exact traveling-wave construction with phase space, stability, bifurcation and chaos analysis. In this study, a unified four-ansatz framework was chosen as it allows for the systematic comparison of several different formulations of the auxiliary equation in the same analytical framework. The methodology is a combination of both Bernoulli-type sub-equations and Riccati-type sub-equations and as such offers a larger class of admissible travelling-wave structures that would be derived from either one of the auxiliary-equation frameworks alone. In addition, the solutions obtained are explicitly represented in an analytical form and can be immediately related to the diagnostics of a dynamical system such as equilibrium analysis, stability classification, bifurcation structures, phase portraits, sensitivity analysis, and Poincaré sections. The present framework offers complete analytical transparency and parameters constraints as compared to analytical approaches using neural networks. It has certain drawbacks, however. The method is suitable for travelling-wave reductions and can not necessarily resolve arbitrary initial-value dynamics or higher dimensional wave interactions. Furthermore, the contemporary neural-network assisted methods can find more complicated structures like rogue waves, lump interactions and multidimensional wave phenomena. Therefore, the proposed framework is regarded as complementary methods that are developing, not a replacement. Application of Ansatz-I, II, III, IV methods The travelling wave transformation converts (4) into the ODE form, Integrating (40) with respect to two times and letting the integrating constant zero, we obtain Taking the homogeneous balance between the highest-order derivative term and the highest order non-linear term yields n=1.. Ansatz-I Setting n=1, Eq. (8) reduces to A system of linear algebraic equations is then created by setting the coefficients of this polynomial to zero, and the set represents the solution to the system, Substituting Eqs. (43) and (44) into Eq. (42) together with Eq. (10) and Eq. (11), we obtain the following travelling-wave solutions of Eq. (4): Ansatz-II Setting n=1, Eq. (12) reduces to A system of linear algebraic equations is then created by setting the coefficients of this polynomial to zero, and the set represents the solution to the system. Substituting Eq. (50) into Eq. (49) together with Eqs. (14) through (23), we obtain the following travelling-wave solutions of Eq. (4): Case-1: If Case-2: If Ansatz-III Setting n=1, Eq. (24) reduces to A system of linear algebraic equations is then created by setting the coefficients of this polynomial to zero, and the set represents the solution to the system. Substituting Eqs. (62) and (63) into Eq. (61) together with Eqs. (26) and (27), we obtain the following travelling-wave solutions of Eq. (4): Ansatz-IV Setting n=1, Eq. (28) reduces to A system of linear algebraic equations is then created by setting the coefficients of this polynomial to zero, and the set represents the solution to the system. Substituting Eq. (67) into Eq. (66) together with Eqs. (30) through (39), we obtain the following travelling-wave solutions of Eq. (4): Case-1: If Case-2: Result and discussions The current paper investigates the van der Waals-type model of granular evolution by using four sub-ODE methodologies of the form of the development of the model, obtaining a comparative study of the solutions in a systematic manner and graphical visualization, with the help of dynamical diagnostic analysis. Comparisons The current paper presents a major extension of the four sub-ODE ansatz-based methods of dynamical diagnostics of dynamical van der Waals-style granular evolution models, finding a rich and varied group of perfectly accurate solutions. In Subsections 5.1 and 5.2, one will find comparisons and graphical representations of the studied model. With the aim of contextualizing the current work, recent works of van der Waals-type granular evolution models and exact travelling-wave analyses are considered, emphasizing models’ focus, key analysis techniques, type of solutions, and the degree of phase-space, bifurcation, and forced-response diagnostics. As an example, Morad et al.7 used Painlevé analysis and Bäcklund transformations to study bubble-type interactions in fluidized granular media, which included phase-portrait classifications but not forced-response diagnostics. In other cases, solitary and periodic travelling waves were reported by Zafar et al.8 and Zhu et al.9 using simplified equation methods, and little or no dynamical interpretation. The two studies by Li et al.10 and Butt et al.12 were also based on the idea of fractional or dark/bright/periodic solution families without focusing on dynamical diagnostics. On the base level, the four sub-ODE ansatz techniques are compared in the current research. Ansatz I and III make use of Bernoulli-type auxiliary equations, and Ansatz II and IV of Riccati-type formulations. Each of the methods defines a trial structure admissible parameter-constraint sets. Ansatz I generates real kink/front-type travelling-wave (TW) profiles, whereas Ansatz II and IV generate families of complex value functions, the real and complex parts of which are viewed independently. Ansatz III extends to other real TW families of shifted structures. Each method has representative solution labels, which are used to connect the auxiliary sub-equation with the TW branches. The paper contributes to the literature through integrating the four-ansatz framework, the use of both Bernoulli- and Riccati-type auxiliary equations whose coefficient is equal, and explicit constraints. The resulting TW solutions include front/kink and pulse-type families, and complex branches to the solutions, as viewed in terms of their real and imaginary parts. Systematic inclusions of dynamical diagnostics are also made, such as equilibrium analysis, trace-determinant stability classification, bifurcation framing, and representative periodic/localized forcing are investigated through Poincaré stroboscopic sampling. Table 1 contextualizes the current study in the recent literature7–12 on van der Waals-style granular evolution models, involving solution strategies and incorporation of dynamical interpretations, including stability, bifurcation, and forced-response diagnostics. The last row emphasizes the most important contributions of the present research in comparison with the works of other researchers. The four sub-ODE ansatz methods (Ansatz I -IV) at the root level are summarized in Table 2, giving the auxiliary equation of each method, the trial solution structure, and the sets of admissible parameters and constraints. Each ansatz method is also associated with its TW branches in this table, where a representative label of each solution is given and the most common type of waveform. Table 1Comparison with other studies on van-der-Waals-type granular evolution models and exact travelling-wave analysis, including the contribution of the present work.StudyModel focusMain analytic method(s)Main solution types reportedPhase-space/bifurcation diagnosticsForced response/Poincaré diagnosticsMorad et al.7VDWNF for fluidized granular medium; “bubble interactions” framingPainlevé analysis; Bäcklund transformation; tanh-form and elliptic-function constructionsTravelling waves; bubble-type interaction interpretationIncludes phase-portrait classificationNot emphasizedZafar et al.8VDWNF (granular materials)Simplest equation and Exp-function methodsSolitary and periodic travelling wavesLimited dynamical interpretationNot includedZhu et al.9VDWNF (granular materials)New auxiliary equation method via the general Riccati equationHyperbolic solitary-wave familiesNot centralNot includedLi et al.10Time-fractional van-der-Waals model in a granular contextAlgebraic method with elliptic/hyperbolic functions (fractional formulation)Fractional travelling wavesNot centralNot centralButt et al.12van-der-Waals equation families; comparative framingDirect algebraic strategies; fractional-operator comparisonDark/bright/singular/periodic familiesNot centralNot centralParvin et al.11VDWNF for fluidized granular matterModified S-expansion methodKink/bright/dark travelling wavesNot centralNot includedThe present studyVDWNF for fluidized granular matterUnified four-ansatz framework using Bernoulli- and Riccati-type auxiliary equations with coefficient matching and explicit constraintsFront/kink and pulse-type TW families; complex branches with real/imaginary-part interpretationEquilibria, trace–determinant stability classification, and bifurcation-type framingrepresentative periodic/localized forcing with stroboscopic (Poincaré) sampling to diagnose periodic vs. multi-periodic responses Comparison with other studies on van-der-Waals-type granular evolution models and exact travelling-wave analysis, including the contribution of the present work. Table 2Root-level comparison of the four ansatz methods and their admissible constraint sets.MethodAuxiliary sub-equation for Trial structure for Admissible parameter-constraint setsRepresentative solution labels shown in figuresTypical waveform characterAnsatz IBernoulli-type: Sets (43)–(44) Real kink/front-type TW profilesAnsatz IIRiccati-type: Set (50) Complex-valued family; real and imaginary parts interpreted separatelyAnsatz IIIBernoulli-type: Sets (62)–(63) Additional real TW families with shifted structureAnsatz IVRiccati-type: Set (67) Complex-valued family; real and imaginary parts interpreted separately Root-level comparison of the four ansatz methods and their admissible constraint sets. Bernoulli-type: Riccati-type: Bernoulli-type: Riccati-type: Besides the classical exact-solution methods mentioned above, the present research has investigated nonlinear propagation through the use of a variety of advanced analytical, numerical, and stochastic methods of the dynamical behaviour, which are discussed in Table 3. Baber et al.26 used dynamical analysis and exact wave construction to study the nonlinear wave structures, such as lump, breather, M-shaped, and others, for electrical transmission-line models. In27, the propagation of optical waves in a generalized anti-cubic magneto-optic waveguides is studied, and various nonlinear optical structures are analyzed analytically. In an allelopathic phytoplankton competition model28, the influence of noise on exact solitary-wave propagation was investigated, in GRIN multimode optical fibers29, the propagation of optical solitons was analyzed under the influence of stochastic perturbations and noise effects. In30, reliable numerical schemes for the coupled nonlinear Schrödinger equations were constructed when the noises were multiplicative in time. The effect of Brownian motion on the dynamics of optical solitons in multi-component nonlinear Schrödinger systems was studied in31. These studies illustrate the increasing interest in the amalgamation of exact analytical methods, stochastic analysis, diagnostics of the dynamics of nonlinear evolution equations, and numerical simulation. Table 3Comparison with recent advanced analytical and computational approaches.ModelMain methodNoise/stochastic effectsExact solutionsDynamical analysisReferencesElectrical transmission linesDynamical analysisNoYesYesBaber et al.26Magneto-optic waveguidesAnalytical wave analysisNoYesLimitedBaber et al.27Phytoplankton competition modelExact solutions with noiseYesYesModerateBaber et al.28GRIN optical fiberSoliton dynamics with noiseYesYesYesBaber et al.29Coupled NLS equationNumerical schemeYesNoLimitedBaber et al.30Three-component NLSBrownian motion analysisYesYesModerateBaber et al.31VDWNF granular modelUnified four-ansatz frameworkNoYesExtensivePresent Study Comparison with recent advanced analytical and computational approaches. Graphical representations of the obtained solutions This section provides the visualization of the representative travelling-wave solutions found in Sect. 4 through the 3D surface plots, contour plots, fixed-time spatial profiles, and density/intensity maps on the domain of domain and . The fixed-time profiles confirm that the waves propagate in the x direction, and the shape and amplitude of the waves remain mostly constant as the wave propagation is of a travelling-wave nature. The van der Waals-type granular evolution model analysis using the graphical analysis of the model going through the four sub-ODE ansatz methods, demonstrates a range of structures, travelling waves. Figures 1, 2, 4, 5, and 7 show kink-type solutions, which are described by an increasing (or decreasing) transition between two different asymptotic states. These kinks are localized sharp fronts of the granular medium and are a transition between stable positions, indicating areas of phase change or density fluctuation of the system. Kink‑type travelling‑wave solution of Eq. (45) for , : (a) 3D surface plot; (b) contour plot; (c) time‑slice profiles at ; (d) density plot. Kink‑type travelling‑wave solution of Eq. (51) for , : (a) 3D surface plot; (b) contour plot; (c) time‑slice profiles at ; (d) density plot. Real part of the travelling‑wave family of Eq. (57) for and : (a) 3D surface plot; (b) contour plot; (c) time‑slice profiles at ; (d) density plot. Travelling‑wave solution of Eq. (64) for , : (a) 3D surface plot; (b) contour plot; (c) time‑slice profiles at ; (d) density plot. Real part of the travelling‑wave family of Eq. (70) for and : (a) 3D surface plot; (b) contour plot; (c) time‑slice profiles at ; (d) density plot. Figure 4 further shows a bright-dark soliton solution, in which there is a localized amplification of the amplitude (bright soliton) and a localized diminution of the amplitude (dark soliton). This birefringent structure is the manifestation of complicated nonlinear interactions within the medium, which reveal coexisting localization and depletion of energies, which are physically crucial in the description of the co-existence of the regions of high- and low-density granular structures. Figure 6 shows a typical soliton solution, which is a localized, non-dispersive wave that does not change its shape and velocity as it propagates because of a fine balance between nonlinearity and dispersion. In granular evolution, these solitons are used to represent stable wave packets or pulses that travel through the medium undistorted, which is the case with solitary density or pressure waves. All these numbers not only prove the existence of various travelling-wave solutions but also show the dynamics of the system, which are rather complex, proving the richness of the solution space of the model. The kink, bright-black soliton, and conventional soliton profile observed give invaluable understanding of the interaction between nonlinear effects, stability, as well as phase space dynamics of van der Waals-like granular media. These findings highlight the efficiency of the integrated four-sub-ODE ansatz system and the employed dynamical diagnostics, such as equilibrium analysis, bifurcation framing, and Poincaré sampling in describing the entire range of physically significant wave phenomena in the system. Imaginary part of the travelling‑wave family of Eq. (70) for and : (a) 3D surface plot; (b) contour plot; (c) time‑slice profiles at ; (d) density plot. For and , Fig. 1 shows that in Eq. (45) is a kink (travelling front) linking two far-field plateaus through a localized transition layer. Figure 1(a) displays the full spatiotemporal front profile, and Fig. 1(b) maps the front trajectory in the (x,t)-plane as a narrow transition band. The profiles in Fig. 1(c) at and 7.2 preserve the kink shape while translating in x, indicating shape-preserving propagation consistent with kink dynamics. Figure 1(d) provides the corresponding intensity map, highlighting the two plateaus and the concentrated high-gradient interface. For and , Fig. 2 shows that in Eq. (51) forms a kink (travelling front) that connects two far-field plateaus through a localized transition layer. In Fig. 2(a) the full spatiotemporal profile of the front is visible, while Fig. 2(b) maps the front trajectory in the (x,t)-plane as a narrow transition band. The profiles in Fig. 2(c) at and 7.2 preserve the kink shape while translating in x, indicating shape-preserving propagation consistent with kink dynamics. Figure 2(d) provides the corresponding density map, highlighting the two plateau regions and the concentrated high-gradient interface. For and Fig. 3 depicts from Eq. (57) as a spatially localized component that changes sign over time. Figure 3(a) shows the full spatiotemporal surface of Figure 3(b) tracks the localized positive and negative regions in the (x,t)-plane and makes the sign reversal evident through the change in contour levels. Figure 3(c) presents profiles at and confirming that the localized peak diminishes and reverses sign as t increases. Figure 3(d) provides an intensity view of the same evolution, highlighting where is concentrated and how its magnitude and sign vary across time. Imaginary part of the travelling‑wave family of Eq. (57) for , : (a) 3D surface plot; (b) contour plot; (c) time‑slice profiles at ; (d) density plot. For and Fig. 4 presents from Eq. (57) as a smooth, kink-like transition between two far-field plateaus. Figure 4(a) shows the full spatiotemporal surface, highlighting a monotone front centered near the wave core. Figure 4(b) tracks the transition region in the (x,t)-plane, where the rapid change between the two plateaus is concentrated. The profiles in Fig. 4(c) at and preserve the overall kink shape with only mild variation in steepness, indicating stable, shape-preserving evolution. Figure 4(d) provides the corresponding intensity visualization, clearly separating the two background levels and emphasizing the localized transition layer. For and Fig. 5 shows that in Eq. (64) forms a kink (travelling front) connecting two far-field plateaus through a localized transition layer. Figure 5(a) presents the full spatiotemporal surface, while Fig. 5(b) tracks the front trajectory in the (x,t)-plane as a narrow transition band. Figure 5(c) compares profiles at and 7.2, where the kink shape is preserved as the front translates in x, indicating stable, shape-preserving propagation. Figure 5(d) provides the corresponding density visualization, clearly separating the two background states and emphasizing the concentrated high-gradient interface. For and Fig. 6 shows that in Eq. (70) forms a localized pulse that propagates through the (x,t)-plane. Figure 6(a) presents the full spatiotemporal surface of the imaginary component. Figure 6(b) tracks the pulse trajectory as a narrow high-intensity band, while Fig. 6(d) provides the corresponding density visualization of the same propagating structure. Figure 6(c) compares profiles at and 7.2, demonstrating a shape-preserving pulse whose peak translates in x with time, indicating stable propagation of the imaginary component within the complex travelling-wave solution. For and , Fig. 7 shows that in Eq. (70) forms a kink-type travelling front connecting two far-field plateaus through a localized transition layer. Figure 7(a) presents the full spatiotemporal surface of the real component. Figure 7(b) tracks the front trajectory in the (x,t)-plane as a narrow transition band, while Fig. 7(d) provides the corresponding density visualization of the same moving interface. Figure 7(c) compares profiles at and where the kink shape is preserved as the front translates in x, indicating stable, shape-preserving propagation of the real part within the complex travelling-wave solution. Phase-space structure and bifurcation analysis of the reduced oscillator In this section, phase portraits, chaotic behaviour, and sensitivity analysis of the study model were graphically illustrated, with a special focus on the qualitative dynamical behaviour on the change of parameters43. Recently, bifurcation analysis, sensitivity diagnostics, and chaos detection have been proven to be the most useful tools for understanding the evolution of nonlinear equations. Nawaz et al.44 investigated stochastically perturbed nonlinear Schrödinger equations driven by Brownian motion, for instance, and showed that stochastic perturbations can modify the solitary-wave structures and the dynamical stability. Hassaballa et al.45 discussed travelling-wave solutions and the associated dynamical structures of the generalized Benjamin–Ono equation with particular focus on the interplay between exact solutions and phase-space dynamics. Baber et al.46 studied the bifurcation structures, chaotic dynamics, and noise-induced phenomena in the stochastic Jaulent–Miodek hierarchy model and observed rich transitions between regular and chaotic regimes. Furthermore, Taskeen et al.47 investigated bifurcation, chaos, sensitivity analysis, and the dynamics of optical solitons of the Spin Hirota–Maxell–Bloch equation in erbium-doped fibers. These studies all show that equilibrium structures, bifurcation transitions, sensitivity to perturbations, and chaotic attractors are fundamental in organizing the nonlinear wave dynamics. Inspired by this progress, the present study examines similar dynamic mechanisms by studying the equilibrium points, stability classification, bifurcation interpretation, phase portraits, Poincaré diagnostics, and sensitivity investigations of the van der Waals normal form for fluidized granular material. Normalized oscillator form and planar reduction To connect the travelling-wave reduction with a regime-diagnostic viewpoint, we rewrite the reduced ODE in a normalized oscillator form. Starting from Eq. (41) in Sect. 4 and dividing by , we obtain: with The denominator provides an immediate parameter-sensitivity warning: when , the reduced coefficients become large, indicating a near-resonant/near-singular scaling in the reduced ODE. Hence, parameter choices near this locus should be treated with caution in numerical exploration. Introduce the planar state variables and. . Then the oscillator becomes the autonomous planar system Fixed points satisfy and , so . Therefore, the equilibria are and when . If , only exists, indicating an equilibrium-structure change at (pitchfork-style in the sense of appearance/disappearance of symmetric equilibria). To determine the nature of the equilibrium points, we compute the Jacobian matrix at is The trace and determinant of the Jacobian matrix are The stability of equilibrium points can therefore be classified according to the following conditions, which is shown in Table 4. Table 4The stability of equilibrium points of Eq. (41).ConditionStability Saddle point Asymptotically stable (node or spiral) Unstable (node or spiral) Center (neutrally stable) The stability of equilibrium points of Eq. (41). In this model of Eq. (78) with the system is conservative (Hamiltonian), and energy is conserved. For the generic dissipative case , the stability depends on α\alphaα and , and the equilibrium count depends on . At the origin, . When , the two additional equilibria exist and satisfy , so their saddle/non-saddle character is opposite that of the origin (for the same and . At , and . So, the stability of will be opposite of , which is shown in Table 5. Table 5Stability summary of equilibria for the reduced cubic oscillator using and . We will get only one fixed point [at and ]IfThenTherefore is asymptotically stable is unstable is Saddle Stability summary of equilibria for the reduced cubic oscillator using and . The qualitative dynamics of the nonlinear system of Eq. (78) are illustrated through the phase portraits shown in Fig. 8(a)–(j) for different parameter combinations. In Fig. 8(a)-(j), the phase portraits have been presented at various combinations of parameters of Eq. (78), where represents the inertia parameter, represents the damping coefficient, and denotes the nonlinear stiffness. The phase portraits obtained indicate the sensitivity of these parameters to the stability of the system, its oscillatory behavior, and nonlinear dynamics. Phase portraits of the nonlinear dynamical system , for different parameter sets illustrating the influence of inertia, damping, and nonlinear stiffness on the system dynamics. Panels (a)–(j) correspond to the parameter combinations: (a) , (b) , (c) , (d) , (e) , (f) , (g) , (h) , (i) , and (j) . The obtained phase trajectories reveal transitions among stable oscillatory motion, bistable structures, and unstable dynamics as a function of the interplay between damping and nonlinear stiffness parameters. Figure 8(a) is associated with parameter set . Here, inertia is positive, damping is positive, and the nonlinear stiffness coefficient is negative. The phase diagram shows that the trajectories approach the equilibrium points gradually, indicating dissipative dynamics. The bistable potential structure induced by the negative cubic nonlinearity gives three equilibrium points in the phase plane. The origin is a state of unstable equilibrium, and there are two symmetric equilibrium points, which are attractors. This is typical of a damped bistable oscillator where the system eventually settles down into one of two equilibrium positions. Figure 8(a) is associated with parameter set . Here, inertia is positive, damping is positive, and the nonlinear stiffness coefficient is negative. The phase diagram shows that the trajectories approach the equilibrium points gradually, indicating dissipative dynamics. The bistable potential structure induced by the negative cubic nonlinearity gives three equilibrium points in the phase plane. The origin is a state of unstable equilibrium, and there are two symmetric equilibrium points, which are attractors. This is typical of a damped bistable oscillator where the system eventually settles down into one of two equilibrium positions. In Fig. 8(b), the parameters are . The reversing negative inertia overturns the acceleration in the system, leaving the damping positive. The directions of the trajectories in the phase plane are therefore opposite to those in Fig. 8(a). Even though the presence of a positive damping still has a factor in the energy dissipation, the negative inertia actually changes the temporal character of the system considerably. These effects can occur in systems that have effective negative mass, or strange inertial responses that arise in more sophisticated models of wave propagation. In Fig. 8(b), the parameters are . The reversing negative inertia overturns the acceleration in the system, leaving the damping positive. The directions of the trajectories in the phase plane are therefore opposite to those in Fig. 8(a). Even though the presence of a positive damping still has a factor in the energy dissipation, the negative inertia actually changes the temporal character of the system considerably. These effects can occur in systems that have effective negative mass, or strange inertial responses that arise in more sophisticated models of wave propagation. Figure 8(c) represents the phase portrait of . In this case, the coefficient of damping is negative, i.e., energy is injected continuously into the system. The curves are drawn out of the equilibrium points, which implies the instability and the increase of oscillations. The stage diagram well illustrates the increase in the oscillatory movement as a result of the external energy input. Such behaviour is characteristic of active nonlinear oscillators, including driven circuits of electricity, or energy-pumped mechanical systems. Figure 8(c) represents the phase portrait of . In this case, the coefficient of damping is negative, i.e., energy is injected continuously into the system. The curves are drawn out of the equilibrium points, which implies the instability and the increase of oscillations. The stage diagram well illustrates the increase in the oscillatory movement as a result of the external energy input. Such behaviour is characteristic of active nonlinear oscillators, including driven circuits of electricity, or energy-pumped mechanical systems. In Fig. 8(d), the parameters are . The damping term of the equation is zero, thus making the system act as a conservative nonlinear oscillator where the total energy is conserved. Depending on the phase portrait, periodic motion can be seen as the closed trajectories around the equilibrium points. The negative cubic nonlinearity is a bistable potential structure, where oscillation can take place between symmetric potential wells. This is typical of nonlinear conservative systems of mechanics and fluid dynamics. In Fig. 8(d), the parameters are . The damping term of the equation is zero, thus making the system act as a conservative nonlinear oscillator where the total energy is conserved. Depending on the phase portrait, periodic motion can be seen as the closed trajectories around the equilibrium points. The negative cubic nonlinearity is a bistable potential structure, where oscillation can take place between symmetric potential wells. This is typical of nonlinear conservative systems of mechanics and fluid dynamics. Figure 8(e) corresponds to . In this set-up, the inertia is positive, and the damping is zero, hence the system once again acts like a conservative oscillator. The phase diagram has closed periodic curves in the phase plane. These paths are nonlinear oscillations whose motion continues indefinitely since there is no energy loss. The negative cubic term causes the creation of the double-well potential, which causes oscillatory motion of the points of stable equilibrium. Figure 8(e) corresponds to . In this set-up, the inertia is positive, and the damping is zero, hence the system once again acts like a conservative oscillator. The phase diagram has closed periodic curves in the phase plane. These paths are nonlinear oscillations whose motion continues indefinitely since there is no energy loss. The negative cubic term causes the creation of the double-well potential, which causes oscillatory motion of the points of stable equilibrium. In Fig. 8(f), the parameters are . At the point where the nonlinear stiffness coefficient turns positive, it becomes a single-well potential structure with a single equilibrium point at the origin. A phase diagram indicates a family of closed orbits around the origin, which are periodic oscillations of period of stability. This physical situation is a hardening nonlinear oscillator in which the restoring force is proportional to the displacement. In Fig. 8(f), the parameters are . At the point where the nonlinear stiffness coefficient turns positive, it becomes a single-well potential structure with a single equilibrium point at the origin. A phase diagram indicates a family of closed orbits around the origin, which are periodic oscillations of period of stability. This physical situation is a hardening nonlinear oscillator in which the restoring force is proportional to the displacement. Figure 8(g) illustrates the phase diagram at the parameters:. This negative inertia reverses the orientation of the paths, and at the same time, no damping leads to conservation of energy. Consequently, the periodic oscillatory motion continues to exist in the phase plane. The phase trajectories are closed, and this implies that there are nonlinear oscillations with inverted dynamical flow. Figure 8(g) illustrates the phase diagram at the parameters:. This negative inertia reverses the orientation of the paths, and at the same time, no damping leads to conservation of energy. Consequently, the periodic oscillatory motion continues to exist in the phase plane. The phase trajectories are closed, and this implies that there are nonlinear oscillations with inverted dynamical flow. In Fig. 8(h), the parameters are . The inertia and damping are negative in this case that creates a lot of instability in the system. The phase diagram illustrates that there are quickly spreading curves that head off towards the equilibrium points. The negative damping continuously feeds the system with energy, therefore leading to an unlimited increase in the amplitude of oscillation and nonlinear unstable behaviour. In Fig. 8(h), the parameters are . The inertia and damping are negative in this case that creates a lot of instability in the system. The phase diagram illustrates that there are quickly spreading curves that head off towards the equilibrium points. The negative damping continuously feeds the system with energy, therefore leading to an unlimited increase in the amplitude of oscillation and nonlinear unstable behaviour. Figure 8(i) corresponds to . During this regime of parameter values, inertia and damping are positive, and the nonlinear stiffness is positive. The phase diagram indicates curves that approach the origin, and this shows that the equilibrium is stable, i.e., it is asymptotically stable. The energy is lost slowly, and the oscillations reduce in intensity, and the system moves towards a steady state level. Figure 8(i) corresponds to . During this regime of parameter values, inertia and damping are positive, and the nonlinear stiffness is positive. The phase diagram indicates curves that approach the origin, and this shows that the equilibrium is stable, i.e., it is asymptotically stable. The energy is lost slowly, and the oscillations reduce in intensity, and the system moves towards a steady state level. Lastly, Fig. 8(j) shows the scenario of and with all the parameters being large and the inertia and damping being negative. The phase diagram is characterized by highly divergent curves with high amplitude growth. This implies that it is highly unstable in a nonlinear way, and it increases the energy in the system. Such dynamics can occur in nonlinear systems that are highly energized, in which external forcing or energy input prevails in the system evolution. Lastly, Fig. 8(j) shows the scenario of and with all the parameters being large and the inertia and damping being negative. The phase diagram is characterized by highly divergent curves with high amplitude growth. This implies that it is highly unstable in a nonlinear way, and it increases the energy in the system. Such dynamics can occur in nonlinear systems that are highly energized, in which external forcing or energy input prevails in the system evolution. The findings are a clear indication that the qualitative dynamics of the system highly rely on the interaction of nonlinear stiffness, inertia, and damping. Positive damping stabilizes and gives rise to attractors, whereas negative damping injects energy and results in instability. Besides, the sign of the cubic nonlinearity is what defines the existence of the bistable or single-well potential structures in the system. Such observations give much understanding of the nonlinear behavior of oscillatory systems that occur in mechanical vibrations, fluid dynamics, nonlinear propagation of waves, and plasma physics. The two-dimensional phase portraits, three-dimensional trajectories, time series, and Poincaré sections are investigated to study the dynamical properties of the nonlinear system using various combinations of the parameters of , , and . Figure 9 shows the time dependence of the system at the parameters , , and . Because is the damping coefficient, then the system will act as a nonlinear conservative oscillator. Figure 9(a) in the two-dimensional phase portrait indicates that the equilibrium point has close, symmetric trajectories, indicating bounded periodic motion in the phase plane. Figure 9(b) shows the periodicity of the system dynamics since the solution varies continuously with time, but it is in a closed phase space. This circular path proves the permanence of the vibratory movement. It can be observed in the time series in Fig. 9(c) that there are regular oscillations of the displacement X(t) and velocity Y(t). The profiles of oscillations are slightly out of the shape of a purely sinusoidal waveform as the cubic nonlinear term exists. The Poincaré section, as shown in Fig. 9(d), reduces to a unique point on the phase plane, which is a common attribute of periodic motion. This proves the fact that the system goes into the same state after a definite period of oscillation. This regime is associated with a stable nonlinear vibration system, a nonlinear system in which the interaction of linear restoring forces and cubic nonlinear stiffness results in sustained periodic oscillations. Dynamical behavior of the nonlinear system with the parameter set of , , and : (a) two-dimensional phase portrait, (b) three-dimensional phase trajectory, (c) time series of displacement X(t) and velocity Y(t), (d) Poincaré section of a bounded periodic motion. Under the conditions of Fig. 10, i.e., the parameters are , , and , the system behavior is as given in Fig. 10. The inverted inertial parameter greatly affects the stability characteristic of the system. The phase diagram displayed in Fig. 10(a) demonstrates the trajectories that widen outward as compared to the equilibrium point, as opposed to closed orbits. This external instability will mean that the equilibrium position is now dynamically unstable. The three-dimensional path in Fig. 10(b) also indicates that system behaviour increases with time since the path in the phase space moves away, although not bounded. Figure 10(c) displays a time series that indicates that the values of both the displacement and the velocity gradually grow with time, which confirms the existence of an unstable dynamical regime. Besides, the Poincaré section in Fig. 10(d) reflected scattered points in a part of the phase plane. Such a non-uniform distribution implies that the system is not on a simple periodic orbit and has unstable or irregular dynamics. This parameter set is that of an unstable nonlinear oscillatory system, in which minor perturbations will increase with time and cause exaggerated movement. Nonlinear system dynamics: high dynamical activity of the nonlinear system with , , and : (a) phase diagram, (b) three-dimensional phase diagram, (c) time series, and (d) Poincaré section, showing the development of unstable and divergent oscillatory behavior. The parameter configuration with , , and and system dynamics would appear as shown in Fig. 11. In this parameter regime, the system once again turns out to be a nonlinear conservative oscillator. The phase portrait of Fig. 11(a) has closed curves that encircle the equilibrium point, and these curves are periodic and stable orbits of the phase plane. Figure 11(b) and (c) confirm the persistence of limited oscillatory motion through the three-dimensional trajectory since the solution continues advancing in period in time without separation. Figure 11(c) time series illustrates constant oscillations of the system variables with minimal nonlinear deviations due to the cubic nonlinear term. The Poincaré section of Fig. 11(d) is a point in the phase plane, which confirms the presence of a stable periodic orbit with a definite oscillation period. This is characteristic of nonlinear mechanical oscillators and wave propagation systems, in which nonlinear stiffness changes the oscillation profile without changing periodic motion. Time-dependent dynamics of the nonlinear system at , , and : (a) phase diagram, (b) three-dimensional phase curve, (c) time series, and (d) Poincaré section, which prove stable periodical oscillations of the system. As it is shown in the numerical simulations in Figs. 9, 10 and 11, the qualitative dynamics of the system are highly sensitive to the values of the parameters. In the case where the inertial parameter is positive, and the damping coefficient is zero, the system can be considered as a conservative nonlinear oscillator with periodic stable movement. But inverting the sign of the is a major change in the stability properties and can cause unstable oscillations or divergent oscillations. These findings reflect the high sensitivity of nonlinear dynamical systems to changes in parameters, which is commonly manifested in nonlinear mechanical vibrations, plasma oscillations, and fluid dynamical systems. Quasi-periodicity and chaos analysis after the introduction of different perturbation terms In order to explore the dynamical nature of the system further, we impose some representative external perturbations and examine the ensuing responses of the van der Waals normal form of fluidized granular matter. Qualitative indicators of the time series, phase-portrait geometry, and geometry of the associated Poincaré sections are also used to examine the nature of the system dynamics (periodic, quasi-periodic, or chaotic) based on the time series. As these diagnostics could be affected by sampling decisions and time-dependent effects, the analysis of the results is conducted in a consistent manner with the help of the criteria summarized in Table 6 for all of the considered forced cases. In the sense of nonlinear science and dynamical systems, chaos is a very complex and apparently disordered behavior that occurs in deterministic systems when parameters lie within certain regimes. This seems to be arbitrary and unpredictable, in spite of the fact that such behavior is determined by clear deterministic equations. The property of chaotic dynamics is extreme sensitivity to initial conditions were small, nevertheless, changes in the initial state can cause very different long-term behavior. The given phenomenon is significant in the study of the structure and the stability of nonlinear systems. Chaotic dynamics give a useful understanding of the mechanisms of transition between ordered and disordered phases in the context of van der Waals normal form of fluidized granular matter. Specifically, the onset of chaos emphasizes the nonlinear interactions that the system exhibits and demonstrates the circumstances in which the dynamics is irregular and hard to predict. In order to exemplify these characteristics, we think of three representative perturbation conditions and examine the response of the system in these cases as phase portraits, time-series evolution, and in a suitable way, constructed Poincaré sections. Table 6Qualitative signatures used to classify periodic, quasi-periodic, and chaotic responses from time series, phase portraits, and Poincaré sections.BehaviorTime SeriesPhase PortraitPoincare MapPeriodicRepeats ExactlyClosed loopOne or few pointsQuasi-periodicSmooth, Layered CyclesDense loops on a torusClosed smooth curveChaoticIrregular and UnpredictableStrange attractorsScattered points Qualitative signatures used to classify periodic, quasi-periodic, and chaotic responses from time series, phase portraits, and Poincaré sections. To study the input of periodic external forcing, a perturbation term is in added to Eq. (78). Consequently, the initial system is changed to the subsequent perturbed dynamical system. Figure 12 represents the dynamical properties of the periodically forced nonlinear system with the parameter set, , , , A=0.3, and B=0.5. The two-dimensional phase plot in Fig. 12(a) demonstrates that the system paths are contained within a closed area of the phase space, which means stable oscillatory behaviour. As a result of the lack of damping (), the oscillations continue indefinitely, whereas the nonlinear cubic term with changes the orbit shape and adds stiffness depending on the amplitude. As a result, the deformation of the phase trajectories is nonlinear in contrast to a purely linear oscillator, which has a circular orbit. Figure 12(b) shows the three-dimensional phase trajectory, which shows how the orbit changes with time. The trajectory shifts in a smooth direction with regard to the time axis without separation, which validates the fact that the external periodic forcing does not destabilize the system. Rather, the forcing term serves as a source of energy, keeping the nonlinear oscillator in perpetual oscillations. The continuous oscillatory motion is further proved by the temporal development of the state variables X(t) and Y(t) in Fig. 12(c). Both variables show limited and sustained oscillations, which are amplitude-modulated by the nonlinear restoring force and the periodic excitation. Interaction between intrinsic nonlinear dynamics and the external forcing results in slightly distorted waveforms relative to harmonic oscillations only. The Poincaré section in Fig. 12(d) further gives us a better understanding of the long-term behavior of the system. The dots of the Poincaré map are organized instead of being randomly distributed, which means that the system will tend towards a periodic or quasi-periodic attractor. This action affirms that the system is dynamically ordered and is not in chaos under the chosen parameter setup. The results of Fig. 12 indicate that the nonlinear oscillator with cubic stiffness that is being perturbed by a moderate periodic forcing has stable forced oscillations with a regular attractor structure in which the nonlinear stiffness has a dominant effect on the amplitude and stability of the system response. Dynamical response of the forced nonlinear system for the parameter values , , , A=0.3, and B=0.5 with initial condition . (a) 2-D phase diagram with confined oscillatory curves in the plane. (b) Phase trajectory 3D depicting the time-dependence of the system. (c) Time series of the state variables, X(t)and Y(t), which show sustained forced oscillations. (d) Poincaré section of a regular periodic or quasi-periodic attractor of the applied external forcing. In order to examine the dynamical impact of periodic external forcing, we have added the perturbation term in Eq. (78). As a result the initial model is transformed to the following disturbed dynamical system. Figure 13 represents the dynamical response of the nonlinear forced oscillator to the periodic perturbation of the oscillator in the form of a sinusoidal forcing term regardless of the Bt period. The dynamical behavior of the system will be studied based on the two-dimensional phase portrait, three-dimensional trajectory, the time series evolution, and the resulting Poincaré section. Figure 13(a) shows the two-dimensional phase portrait in the plane, and it is a description of the interaction of the system displacement variable and velocity variable. The curve is contained in a finite area in the phase space, which means that the system has stable oscillatory behaviour as a result of the external sinusoidal forcing. Due to the zero value of the damping parameter (), the system does not dissipate energy, and thus the oscillations can be maintained during the simulation. The cubic nonlinear involvement of the stiffness related to is that the restoring force of the oscillator is changed, causing a change in the classical elliptical orbit, which is usually viewed in linear oscillators. The distortion of the phase trajectory, therefore, is nonlinear, which is a result of the cubic nonlinearity. Figure 13(b) depicts the three-dimensional phase trajectory and this gives a detailed representation of the time dynamics of the system. The trajectory moves continuously on the time axis, and it is confined to the phase space. This tendency means that the sinuoidal imposed excitation is constantly attempting to inject energy into the system, making the oscillations carry on without the system diverging and becoming unstable. The periodic nature of the orbit indicates that the amplitude and phase of the oscillation are mainly modulated by the external forcing due to the external perturbation, instead of the external forcing causing chaotic behaviour. Figure 13(c) shows the temporal change of the system variables. Time series X(t) and Y(t) prove the presence of waves with limited amplitude. The oscillatory dynamics suggest that the system is in a forced oscillatory regime, in which the nonlinear dynamics governing the system interact with the periodic external excitation. The cubic nonlinearity effect adds stiffness that varies with amplitude and, thus, slightly distorts the waveform relative to harmonic motion. The oscillations are therefore nonlinearly modulated and dynamically stable. Figure 13(d) shows a Poincaré section by sampling the trajectory at fixed time intervals with the removal of transient dynamics. The Poincaré map shows the orderly distribution of points on the phase plane as opposed to the haphazardly distributed one. This organized structure shows that the system moves towards a quasi-periodic or periodic attractor due to the applied sinuoidal forcing. There are no irregular point clouds, which indicates that the effect of chaotic dynamics is not present at the chosen values of the parameters. What these results show is that forced oscillations of the nonlinear oscillator with cubic stiffness on moderate sinuoidal forcing have stable forced oscillations whose trajectories are bounded. The restoring force is nonlinear, which controls the amplitude of movement, and damping is removed, enabling the oscillator and external forcing to exchange energy indefinitely. The system therefore tends to a regular attractor with well-structured phase trajectories and an ordered Poincaré section, and therefore, the stability of the nonlinear forced dynamical system has been proven. Time-dependent response of the periodically driven nonlinear system, , , , A=0.3, and B=0.5 and with sinusoidal forcing and the initial condition . (a) Phase portrait in 2D, (b) phase trajectory 3D in time, (c) time series of state variables, and (d) Poincaré section, which displays the regular attractor structure of the system. To examine the effects of external perturbations on dynamics of the system, the forcing term is introduced in to the equation of state (78). The initial model is therefore reformulated to the subsequent perturbed dynamical system. Figure 14 is used to demonstrate the dynamical behaviour of a nonlinear oscillator when it is exposed to an external perturbation that is exponentially growing. The two-dimensional phase portrait, three-dimensional trajectory, time evolution of the state variables, along with the corresponding Poincaré section, have been used to explore the system dynamics. The phase diagram in the plane is shown in Fig. 14(a). The trajectory will start evolving in a small area of the phase space, but with time, the orbit starts to grow larger and larger since there is an exponentially growing forcing term. The value changes the sign of the linear restoring force and this considerably changes the stability properties of the system. Together with the nonlinear cubic term of , there is a destabilizing effect in the system that leads to the phase curves diverging away. Such a behavior means that the oscillator gradually acquires more and more energy from the external excitation, which results in a gradual increase in the amplitude of oscillations. Figure 14(b) is an illustration of the three-dimensional phase trajectory and this gives one a good visualization on the dynamics of the system as time progresses. The trajectory is known to have a progressive expansion on the phase space with time, due to the cumulative effect of the exponentially growing perturbation. The exponential forcing adds a continuous injection of energy on the system to push the system out of equilibrium, unlike the periodically forced systems where the motion is confined. Therefore, the phase trajectory exhibits a stretching behavior in the phase space, and it shows that the long-term dynamics are unstable. Figure 14 (c) shows the temporal evolution of the state variables X(t) and Y (t). The time series indicates that the two variables initially vary with small amplitudes, but with the progression of time, there is an amplification in the amplitude. This increase can be attributed mostly to the exponential forcing term which is essentially a time-dependent energy source that becomes more dominant. Consequently, the oscillations grow into nonlinear oscillations with large amplitude out of small and almost harmonic oscillations. The Poincaré section of the trajectory after eliminating transient behaviour is shown in Fig. 14(d), which is obtained by sampling the trajectory at the same time interval. The Poincaré points are not organized into a periodic or quasi-periodic structure as was seen in periodically driven systems. Rather, the points are distributed gradually away in the phase space. This diffusive process implies that the system is not moving to a stable attractor, but rather continuously drifting due to the exponentially growing external excitation. The above results indicate that the nonlinear oscillator, when perturbed by an exponentially growing perturbation, will show an increase in energy and instability. The fact that the damping () is zero and the sign of the linear term () is negative, which also makes the system unable to dissipate the injected energy, makes the restoring mechanism of the oscillator weaker. The interaction between nonlinear stiffness and exponential forcing then results in a progressive oscillatory motion and a widening out phase-space path showing the destabilizing nature of the effects of swiftly growing external perturbations. It can be seen that the nonlinear dynamical system forced exponentially to large values is not approaching any standard periodic attractor, but rather is growing in amplitude and is expanding its phase-space, which is the indication of the system undergoing unstable dynamical behavior. Dynamical behavior of the nonlinear system with , , , A=0.01, and B=0.05 under exponentially increasing forcing with initial condition . (a) Phase diagram in two dimensions and expansion of the trajectories, (b) phase trajectory in three dimensions and time dependence, (c) time series of the state variables’ amplitude, and (d) Poincaré section of the lack of a stable attractor because the excitation of the system is growing. The sensitivity assessment The sensitivity analysis is also significant in the study of the impact of changes in initial conditions on the dynamics of nonlinear dynamical systems. It is the sensitivity of the system dynamics to minor changes in the initial state variables. Even a minor alteration of the initial conditions can cause a drastic difference between the trajectories of the system in nonlinear wave models, which means a high sensitivity of the physical structure of the solution. To examine the sensitivity of the proposed model, the nonlinear equation that governs it is given as the equation below. The system of equations of (41) is changed to a first-order dynamical system of (78). Let then . Such representation makes it possible to study the development of the displacement and velocity in the conditions of various initial conditions. The parameters used in the current research are fixed, as the initial conditions are , , and and the response of the system is discussed under a number of different sets of initial conditions. In every case, there are five trajectories that are taken into account and depicted by red, blue, green, magenta, and yellow curves. Figure 15 demonstrates the nature of the sensitivity of the system. Figure 15 (set 1) is created to produce the trajectories in the case of slightly different positive initial displacements and velocities. These curves have a similar path, but as the independent variable is raised, the curves tend to diverge, and this suggests that any slight deviation of the initial state will result in some discernible difference in the profile of the solution. The first displacement is fixed, with the first velocity varying as in Fig. 15 (set 2). The resultant trajectories exhibit increased differences in amplitude and oscillatory behaviour. It means that the system dynamics are very sensitive to changes in the velocity component and changes the wave propagation properties. The cases of negative initial displacements with varying velocity values are investigated in Fig. 15 (set 3) and (set 4). The trajectories of the solutions exhibit large deviations and variations in the amplitude of oscillation. This behaviour is magnified by the nonlinear cubic of the governing equation resulting in the faster separation of trajectories as the evolution parameter grows. This kind of solution curve deviation ascertains the high sensitivity of the nonlinear system to initial perturbations. As indicated by the sensitivity graphs, slight changes in the initial displacement and velocity cause a measurable change in the amplitude and oscillatory behaviour of the wave solution. This behaviour brings out the inherent nonlinear character of the system and shows that even the physical structure of the wave dynamics is very much reliant on the starting conditions. Therefore, the system has been found to be sensitive to initial states which is a common feature of nonlinear dynamical models. Sensitivity of assessment at different initial conditions of the model of Eq. (78) with , , and Initial condition set 1: (, , , and ); Initial condition set 2: (, , , and ); Initial condition set 3: (, , , and ) and Initial condition set 4: (, , , and ). In the further investigation of the effect of system parameters on the sensitivity of the nonlinear dynamical structure, more numerical experiments are carried out by changing the coefficients , , and , with the same five sets of initial conditions. Figure 16 shows the trajectories with the parameters selected in the way of , , and . The negative cubic nonlinear coefficient in this situation changes the restoring force of the system which changes the oscillatory behaviour of the solution. The trajectories are again divergent even when the initial conditions are changed slightly suggesting that the nonlinear structure is sensitive to perturbations, but the strength and nature of the oscillations are different than in the earlier example. Sensitivity of assessment at different initial conditions of the model of Eq. (78) with , , and . Initial condition set 1: (, , , and ); Initial condition set 2: (, , , and ); Initial condition set 3: (, , , and ) and Initial condition set 4: (, , , and ). Moreover, in case the parameters are chosen as , , and , the obtained trajectories are shown in Fig. 17. The variation in the signs of the linear coefficients can have a great influence on the dynamical equilibrium of the damping and restoring forces. As a result, this system displays an alternate evolution pattern, with the appearance of greater divergence and amplitude changes in the solution curves than they had in the previous parameter configuration. Sensitivity of assessment at different initial conditions of the model of Eq. (78) with , , and . Initial condition set 1: (, , , and ); Initial condition set 2: (, , , and ); Initial condition set 3: (, , , and ) and Initial condition set 4: (, , , and ). Lastly, with the same five initial conditions, setting , , and as the value of sigma gives the resulting solution profiles in Fig. 18. In this setup, the nonlinearity of positive cubic order and the negative linear damping together changes the frequency and amplitude of the oscillations of the trajectories. The findings once again affirm that minute variations in the initial conditions result in significant changes in the wave evolution. So, these numerical experiments obviously indicate that the system parameters, as well as the initial states, are important determinants of the sensitivity and dynamical behaviour of the nonlinear model. Sensitivity of assessment at different initial conditions of the model of Eq. (78) with , , and Initial condition set 1: (, , , and ); Initial condition set 2: (, , , and ); Initial condition set 3: (, , , and ) and Initial condition set 4: (, , , and ). A few qualitative features of the dynamical behavior found in the present van der Waals normal form are similar to those of some recently reported nonlinear systems44–47. In the present system, the variation of the parameters causes significant changes in the organization and stability of the phase space similar to the stochastic resonance Schrödinger model44 and the stochastic Jaulent–Miodek hierarchy model45. Furthermore, the bifurcation transitions presented here are similar to those in the generalized Benjamin–Ono equation46 and the Spin Hirota–Maxwell–Bloch system47, where equilibrium structures are organizing centers for complex dynamical responses. In contrast with the stochastic models discussed in44–47, however, the analysis of the present work is based on a deterministic granular-matter model, where the intrinsic mechanisms of bifurcation associated with the underlying non-linear oscillator structure can be isolated and directly studied. Conclusion This work gives a detailed analytical-dynamical exploration of the van der Waals normal form of fluidized granular matter, which responds to the necessity to integrate explicit travelling-wave constructions with the perspective of a complete dynamical-systems approach. The main task was to build up precisely the travelling-wave solutions in a systematic way, and at the same time derive the interpretation of their meaning in the phase-space dynamics. We used a travelling-wave reduction to do this and came up with four closely related classes of auxiliary-equation ansatzes with Bernoulli- and Riccati-type sub-equations in them. These methods gave explicit, parameterized families of waves, both of real kink and anti-kink fronts, and more complicated complex-valued waves, whose real and complex components can be studied separately. Graphical presentations, including 3D plots, contour diagrams, and time-slice representations, were useful in a finer analysis of the spatial-temporal organization of such waves. The travelling-wave ODE was normalised to a damped cubic oscillator to relate the precise solutions to the system dynamics underlying the solution. This expression allowed an in-depth phase-portrait analysis based on equilibrium structures, the classification of linear stability, and bifurcation diagnostics. Moreover, forced perturbations in the form of representatives were also used to investigate stroboscopic Poincaré sections, which in this case emphasized periodic, quasi-periodic, and multi-periodic responses. Important results indicate how critical symmetry-based pitchfork bifurcations, dissipation-based transitions, as well as the selection of auxiliary equations, are in defining the admissible solution space and system behavior. The results have offered analytical benchmarks - in terms of explicit travelling-wave solutions with explicitly defined parameter constraints - as well as regime-sensitive diagnostics, which can be used to check numerical solvers and interpret reduced model dynamics of pattern-forming dynamics in fluidized granular and other dispersive-dissipative media. The major limitations are due to the high-dimensional parameter space and the idealization of the model, which might not be able to reflect all granular interactions and stochastic effects. However, the combined dynamical and analytical framework provides a solid basis for researching density-wave propagation and intricate oscillatory forms. Systematic parameter sweeps, particularly of chaotic quantification, and investigation of multi-scale interactions should further build on this effort in the future, enhancing predictive abilities and practical use of the granular flow engineering, material processing, and complex fluidized systems. The present framework can be extended in a number of directions in the future. First, the proposed unified four-ansatz method can be extended to fractional-order and variable-coefficient van der Waals-type models for studying memory effects and heterogeneous granular environments. Second, the introduction of stochastic perturbations and random forcing could be considered to more accurately describe realistic fluidized granular systems. Third, the analytical framework can be integrated with symbolic computation techniques that are now available using modern methods of artificial intelligence, such as neural-network-based symbolic discovery and the neurosymbolic reasoning method, to support automated construction and classification of exact travelling-wave solutions. Finally, the developed approach can be extended to apply to other types of nonlinear evolution equations that are important in fluid dynamics, plasma physics, optical systems, and complex material science. Additional future research is possible by combining the present unified auxiliary-equation framework with the recent advancements of neural-network-based analytical methods, such as enhanced bilinear neural network methods, to examine higher-dimensional (coordinate) nonlinear evolution equations, interactions between lump waves and rogue waves, rogue-wave phenomena, stochastic granular systems, and the symbolic discovery of exact solutions with assistance from AI tools39–42. Author contributions Md. Nur Alam: Project administration, Resources, Conceptualization, Writing–review & editing, Visualization, Methodology, Investigation, Validation, Formal analysis, Software, SupervisionMuhammad Ziaul Haque: Data curation, Formal analysis, Investigation, Methodology, Visualization, Writing–original draft. Shams Forruque Ahmed: Investigation, Conceptualization, Writing–review & editing, Writing–original draft, Supervision. Md Nurul Islam: Investigation, Data curation, Conceptualization, Validation Visualization. Kok Swee Sim: Writing, review & editing, Supervision. All authors have read and approved the final version of the manuscript for publication. Data availability Data will be made available on request from the first author, Md. Nur Alam (nuralam.pstu23@gmail.com). Declarations Competing interests The authors declare no competing interests. Footnotes Publisher’s note Springer Nature remains neutral with regard to jurisdictional claims in published maps and institutional affiliations. References Associated Data Data Availability Statement Data will be made available on request from the first author, Md. Nur Alam (nuralam.pstu23@gmail.com).