Adjoint-based Hopf-bifurcation Instability Suppression via First Lyapunov Coefficient
Sicheng He, Max Howell, Daning Huang, Eirikur Jonsson, Galen W. Ng, Joaquim R. R. A. Martins · 2025
Read Adjoint-based Hopf-bifurcation Instability Suppression via First Lyapunov Coefficient by Sicheng He, Max Howell, Daning Huang, Eirikur Jonsson, Galen W. Ng, Joaquim R. R. A. Martins (2025) in the Tower of Babel, a free online library of public-domain works preserved in clean, searchable text. No sign-up. The full text is served as a plain-text file at a predictable URL, where text is present: https://antarctic-labs.com/tower-of-babel/arxiv-2511-03840.txt.
Introduction
Sicheng He and five colleagues develop a design tool for suppressing unstable limit cycle oscillations that arise through Hopf bifurcations. In aerospace work such oscillations appear as aeroelastic flutter and transonic buffet, and an unstable cycle can diverge or jump in ways that threaten structure and performance.
Stability of the cycle is captured by the first Lyapunov coefficient. The team computes its derivatives at low cost by combining the adjoint method with reverse derivative tracking. Three design problems test the approach: an algebraic Hopf model, an aeroelastic typical section, and a nonlinear Ginzburg-Landau problem. In each case the method helps steer designs away from unstable bifurcation behaviour.
Catalog record
| CREATOR | Sicheng He, Max Howell, Daning Huang, Eirikur Jonsson, Galen W. Ng, Joaquim R. R. A. Martins |
|---|---|
| YEAR | 2025 |
| COLLECTION | PAPERS |
| FORMAT | TXT |
| FILE SIZE | 89 KB |
| VERSION | 1.0 |
| LICENSE | http://creativecommons.org/licenses/by/4.0/ |
| RIGHTS | CC BY |
| SOURCE | arXiv |
Source: https://arxiv.org/abs/2511-03840
Browse the full library · About the Tower of Babel
Full text
The opening of the work, from the library's hosted text:
Adjoint-based Hopf-bifurcation Instability Suppression via First Lyapunov Coefficient Sicheng He, Max Howell, Daning Huang, Eirikur Jonsson, Galen W. Ng, Joaquim R. R. A. Martins Abstract Many physical systems exhibit limit cycle oscillations induced by Hopf bifurcations. In aerospace engineering, limit cycle oscillations arise from undesirable Hopf-bifurcation phenomena such as aeroelastic flutter and transonic buffet. In some cases, the resulting limit cycle oscillations can themselves be unstable, leading to amplitude divergence or hysteretic transitions that threaten structural integrity and performance. Avoiding such phenomena when performing gradient-based design optimization requires a constraint that quantifies the stability of the bifurcations and the derivative of that constraint with respect to the design variables. To capture the local stability of bifurcations, we leverage the first Lyapunov coefficient, which predicts whether the resulting limit cycle oscillation is stable or unstable. We develop an accurate and efficient method for computing derivatives of the first Lyapunov coefficient. We leverage the adjoint method and reverse algorithmic differentiation to efficiently compute the derivative of the first Lyapunov coefficient. We demonstrate the efficacy of the proposed adjoint method in three design optimization problems that suppress unstable bifurcation: an algebraic Hopf bifurcation model, an aeroelastic model of a typical section, and a nonlinear problem based on the complex Ginzburg–Landau partial differential equation. While the current formulation addresses only a single bifurcation mode, the proposed adjoint shows great potential for efficiently handling Hopf bifurcation constraints in large-scale nonlinear problems governed by partial differential equations. Its accuracy, versatility and scalability make it a promising tool for aeroelastic and aerodynamic design optimization as well as other engineering problems involving Hopf bifurcation instabilities. 1 Introduction Bifurcation is the shift from a linearly stable to a linearly unstable solution due to a single parameter change. This occurrence involves several mechanisms, including the saddle node, pitchfork, transcritical, and Hopf bifurcation [ 1 ] . The Hopf-bifurcation is frequently encountered in phenomena relevant to aircraft design. Examples of such phenomena include flutter, laminar-turbulent transition, Kármán vortex street, and transonic buffet. Besides the bifurcation parameter, another essential characteristic of the bifurcation is its stability. A stable Hopf-bifurcation leads to a supercritical limit cycle oscillation (LCO), which is a benign response. In contrast, an unstable Hopf-bifurcation leads to a subcritical LCO, which can be bi-stable and exhibit a hysteretic response. In addition, bifurcation can lead to semi-stable LCOs. Unstable Hopf-bifurcations should be eliminated from safe designs. Subcritical bifurcations are frequently encountered experimentally in flutter [ 2 , 3 ] . In this paper, we develop an accurate and efficient method for integrating a constraint on the stability of Hopf bifurcations into gradient-based design optimization using the first Lyapunov coefficient and an adjoint formulation. Significant progress has been achieved in tackling optimization problems constrained by ordinary differential equations (ODEs) and partial differential equations (PDEs). In the aerospace industry, various types of dynamical system solutions have been explored and applied in design optimization. Equilibrium point solutions have been explored [ 4 , 5 ] , however these problems are limited because cases involving non-stationary operating conditions frequently involve more complex dynamical system solutions. To address this issue, solutions of problems involving bifurcations [ 6 , 7 , 8 , 9 , 10 , 11 ] , LCOs [ 12 , 13 , 14 , 15 , 16 , 17 ] , and chaotic dynamics [ 18 , 19 ] have been explored in recent research. These solutions can be applied to complex cases such as aeroelastic flutter and transonic buffet, which are Hopf-bifurcations characterized by a limit-cycle oscillation (LCO) in the post-flutter response. To increase energy efficiency, next-generation aircraft wings and large floating offshore wind turbines will have larger aspect ratios and be more flexible [ 20 ] . Such designs are more likely to exhibit undesirable dynamic behavior at non-stationary operating conditions. Thus, it is critical to develop methods to efficiently and accurately consider the bifurcation constraints in a multidisciplinary design optimization (MDO) framework. Furthermore, MDO problems, such as aerodynamic and aerostructural optimization, usually involve a large number of design variables, requiring the use of gradient-based optimization [ 21 , 5 ] . Such optimizations rely on efficient gradient computation algorithms that scale well with the number of design variables [ 22 , Ch. 6] . Hence, the challenges for bifurcation-constrained optimization are two-fold: (1) how to quantify the stability of bifurcation? (2) how to compute the gradient of such a quantity? Figure 1: Stable, unstable, and semi-stable Hopf bifurcations. For each case, the motion magnitude, h , approaches the limit h\rightarrow 0 to measure the bifurcation stability at the equilibrium point. The direction of motion for solutions in the vicinity of the limit cycle oscillation (LCO) is indicated by the arrows. The stability of Hopf-bifurcation can be assessed through three primary categories of methods: (1) methods based on Floquet theory [ 23 , 24 ] , (2) methods based on the LCO curve, i.e., LCO amplitude-LCO parameter response curve [ 25 , 26 , 27 ] , and (3) center manifold-based methods [ 1 , 28 , 29 , 30 , 31 ] . The Floquet theory- and LCO curve-based method use a similar strategy because both analyze the post-bifurcation response, and bifurcation stability is determined by considering motion magnitude approaching zero, as shown in Fig. 1 , where the resulting supercritical LCO converges to an equilibrium solution. Floquet theory-based methods posit that the stability of a periodic dynamical system depends on the eigenvalue distribution of the monodromy matrix. This matrix is obtained by evaluating the fundamental solution matrix of the periodic system at times t=0 and t=T , where T represents one period of the system [ 24 , Ch. 2.4] . In practice, acquiring the monodromy matrix involves simulating the time-periodic dynamical system N times over one period, where N is the number of states of the dynamical system. However, the computational cost of this process is high for problems governed by PDEs, particularly when N is large, as seen in high-fidelity aeroelastic systems studied by Thomas et al. 2002 and He et al. 2019 . In response to this challenge, Bauchau and Nikishkov 2001 proposed using the Arnoldi algorithm to compute the dominant eigenvalue instead of calculating all eigenvalues to assess the stability of a periodic dynamical system. This approach is appealing because it avoids explicit matrix formation. Instead, the approach requires only matrix-vector products through the Arnoldi algorithm, where the response vector can be evaluated using time integration for one period with a specified initial condition. However, this approach requires an unsteady adjoint for derivative computation, which requires a substantial implementation effort and is computationally intensive [ 35 ] . Seyranian et al. 2000 introduced forward mode differentiation of Floquet analysis to bypass the need for an unsteady adjoint. However, the computational cost of this derivative computation method scales with the number of design variables, which is undesirable for large-scale MDO. The LCO curve-based methods leverage the fact that in an LCO amplitude-LCO parameter response curve, the region that has a negative slope is unstable. By studying the stability of an LCO with a small motion magnitude, this class of methods can be used to study bifurcation stability. Leveraging the earlier work of time spectral-based LCO equation