Equivalence Problem for Non-Linearizable Fourth-Order ODEs with Five-Dimensional Lie Symmetry subalgebra via Inductive Cartan Equivalence Method
Sondos R. Khalil, Ahmad Y. Al-Dweik, Marwan Aloqeili, F. M. Mahomed · 2026
Read Equivalence Problem for Non-Linearizable Fourth-Order ODEs with Five-Dimensional Lie Symmetry subalgebra via Inductive Cartan Equivalence Method by Sondos R. Khalil, Ahmad Y. Al-Dweik, Marwan Aloqeili, F. M. Mahomed (2026) 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-2604-13138.txt.
Introduction
Khalil, Al-Dweik, Aloqeili, and Mahomed construct four coframes of invariant 1-forms for non-linearizable fourth-order ordinary differential equations with a 5D Lie symmetry subalgebra. Using the stepwise Cartan matching method with rank zero, they match each of four distinct branches to a coframe. The coframes define the equations under point transformation, and the authors give a recipe for finding the point map itself, shown through examples.
Catalog record
| CREATOR | Sondos R. Khalil, Ahmad Y. Al-Dweik, Marwan Aloqeili, F. M. Mahomed |
|---|---|
| YEAR | 2026 |
| COLLECTION | PAPERS |
| FORMAT | TXT |
| FILE SIZE | 73979 |
| LICENSE | http://creativecommons.org/licenses/by/4.0/ |
| RIGHTS | CC BY |
| SOURCE | arXiv |
Source: https://arxiv.org/abs/2604.13138
Browse the full library · About the Tower of Babel
Full text
The opening of the work, from the library's hosted text:
Equivalence Problem for Non-Linearizable Fourth-Order ODEs with Five-Dimensional Lie Symmetry subalgebra via Inductive Cartan Equivalence Method Sondos R. Khalil, Ahmad Y. Al-Dweik, Marwan Aloqeili, F. M. Mahomed Abstract Four coframes of invariant 1-forms are explicitly constructed using the Inductive Cartan equivalence method with rank zero corresponding to four distinct branches. These coframes are employed to characterize non-linearizable fourth-order ODEs under point transformation with a five-point symmetry Lie subalgebra. Moreover, we propose a procedure for obtaining the point transformation by using the derived invariant coframes, demonstrated through examples. Keywords: Cartan’s equivalence method, Differential invariants, Fourth-order ODEs, Equivalence problem, Point transformations, Canonical forms, Lie point symmetries. 1 Introduction The Lie symmetry classification [ 1 ] over the complex domain identifies four distinct classes encompassing all scalar ordinary differential equations (ODEs) of order two, u^{\prime\prime}=f(x,u,u^{\prime}) , that admit a Lie algebra of dimension three. By adopting the direct method , Ibragimov and Meleshko [ 2 , 3 ] provided an invariant classification of these equations by determining the canonical forms for each class and analyzed them extensively. They showed that any non-linearizable second-order ODE having a three-dimensional Lie symmetry algebra can be mapped, through point transformations, into one of these canonical forms. In more recent studies, Al-Dweik et al. [ 4 ] applied the Cartan equivalence method to construct explicitly three-dimensional invariant coframes corresponding to the three branches. They characterized these ODEs geometrically and a procedure was developed for deriving the corresponding point transformations. Abuloha et al. [ 5 ] naturally extended the results in [ 4 ] to the equivalence problem of third-order ODEs which are non-linearizable by point transformation and possess four-dimensional Lie point symmetry subalgebras. A direct extension of the recent research is to study the equivalence problem for non-linearizable fourth-order ODEs that admit five-dimensional Lie symmetry subalgebras, considering transformations that are point and contact. These form two separate equivalence problems, as some ODEs that are non-linearizable by point transformation may become linearizable under contact transformation. Many studies in the literature provide a full characterization of scalar fourth-order ODEs that possess real Lie algebras L_{n} of dimension n , where n ranges from 1 to 8. Mahomed in [ 6 ] derived the realizations of three-dimensional Lie algebras in the real plane. Cerquetelli [ 7 ] and Fatima [ 8 ] provided a full list of scalar fourth‐order ODEs that have 4‐dimensional Lie algebras. In [ 9 , 10 ] , Shah et. al derived a complete classification of scalar 4th-order ODEs that admit 5-, 6- and 8-dimensional Lie point symmetry algebras. In [ 9 ] , Shah et. al, showed that there are no fourth-order differential equations possessing a seven-dimensional Lie algebra as its symmetry algebra. Classification of scalar fourth-order ODEs possessing real Lie point symmetry algebras (m,n) , where m is the class of the algebra and n=5,6,8 is its dimension, is given in the Appendix I (see [ 9 ] ). As indicated in Table 5 of Appendix I, we infer that any non-linearizable fourth-order ODE u^{(4)}=f(x,u,u^{\prime},u^{\prime\prime},u^{\prime\prime\prime}) via point transformation having a five-dimensional Lie point symmetry subalgebra falls under twelve different types listed in Table 1 . Throughout this paper, we denote u^{\prime},u^{\prime\prime},u^{\prime\prime\prime} by p,~q,~r . Algebra Type Generators The corresponding fourth-order equations (24,5),\alpha=0 \partial_{x},\partial_{u},x\partial_{x}+\alpha u\partial_{u},x\partial_{u},x^{2}\partial_{u} u^{(4)}={r}^{\frac{4}{3}} (24,5),\alpha=b+1 \partial_{x},\partial_{u},x\partial_{x}+\alpha u\partial_{u},x\partial_{u},x^{2}\partial_{u} u^{(4)}=r^{\frac{b-3}{b-2}},~b\neq 1,2,3 (24,5),\alpha=2 \partial_{x},\partial_{u},x\partial_{x}+\alpha u\partial_{u},x\partial_{u},x^{2}\partial_{u} u^{(4)}=r^{2} (24,5),\alpha=1 \partial_{x},\partial_{u},x\partial_{x}+\alpha u\partial_{u},x\partial_{u},x^{2}\partial_{u} u^{(4)}=r^{\frac{3}{2}} (24,5),\alpha=\frac{b+1}{b} \partial_{x},\partial_{u},x\partial_{x}+\alpha u\partial_{u},x\partial_{u},x^{2}\partial_{u} u^{(4)}=r^{\frac{1-3b}{1-2b}},b\neq 0,\frac{1}{3},\frac{1}{2},1 (5,5) \partial_{x},\partial_{u},x\partial_{x}-u\partial_{u},u\partial_{x},x\partial_{u} u^{(4)}=\frac{5}{3}\frac{r^{2}}{q}+q^{\frac{5}{3}} (27,5),r=1 \partial_{x},\partial_{u},x\partial_{u},2x\partial_{x}+u\partial u,x^{2}\partial_{x}+xu\partial u u^{(4)}=\frac{4}{3}\frac{r^{2}}{q}+q^{\frac{7}{3}} (15,5) \partial_{x},\partial_{u},x\partial_{x},u\partial_{u},x^{2}\partial_{x} u^{(4)}=\frac{6qr}{p}-\frac{6q^{3}}{p^{2}}+K\frac{\left(3q^{2}-2pr\right)^{\frac{3}{2}}}{p^{2}} (25,5),r=3 \partial_{x},\partial_{u},x\partial_{u},x^{2}\partial_{u},x\partial_{x}+\left(3u+x^{3}\right)\partial_{u} u^{(4)}=e^{r} (26,5),r=1 \partial_{x},\partial_{u},x\partial_{u},x\partial_{x},u\partial_{u} u^{(4)}=K\frac{r^{2}}{q},~K\neq 0,\frac{4}{3},\frac{5}{3} (6,6) \partial_{x},\partial_{u},x\partial_{x},u\partial_{x},x\partial_{u},u\partial_{u} u^{(4)}=\frac{5}{3}\frac{r^{2}}{q} (28,6),r=1 \partial_{x},\partial_{u},x\partial_{x},x\partial_{u},u\partial_{u},x^{2}\partial_{x}+xu\partial u u^{(4)}=\frac{4}{3}\frac{r^{2}}{q} Table 1: Canonical forms of non-linearizable fourth-order ODEs under point transformation admitting five-dimensional Lie point symmetry subalgebras. Remark 1.1 . In Table 1, we omit all linear canonical forms listed in Table 4 of the Appendix I, as our focus is on non-linearizable fourth-order ODEs. Remark 1.2 . In Table 1, we leave out K in the following representative equations • (24,5) , \alpha=0 , listed in Table 5 of Appendix I, where it is noted that the transformation \bar{x}=x,\;\bar{u}=\frac{1}{K^{3}}u maps the canonical form \bar{u}^{(4)}=K\bar{u}^{\prime\prime\prime\frac{4}{3}} to u^{(4)}=u^{\prime\prime\prime\frac{4}{3}} . • (24,5) , \alpha=b+1 , listed in Table 5 of Appendix I, where it is observed that the transformation \bar{x}=\frac{1}{K}x,\;\bar{u}=\frac{1}{K^{3}}u maps the canonical form \bar{u}^{(4)}=K\bar{u}^{\prime\prime\prime\frac{b-3}{b-2}} to u^{(4)}=u^{\prime\prime\prime\frac{b-3}{b-2}} . • (24,5) , \alpha=2 , listed in Table 5 of Appendix I, where it is noted that the transformation \bar{x}=x,\;\bar{u}=\frac{1}{K}u maps the canonical form \bar{u}^{(4)}=K\bar{u}^{\prime\prime\prime 2} to u^{(4)}=u^{\prime\prime\prime 2} . • (24,5) , \alpha=1 , listed in Table 5 of Appendix I, where it is seen that the transformation \bar{x}=x,\;\bar{u}=\frac{1}{K^{2}}u maps the canonical form \bar{u}^{(4)}=K\bar{u}^{\prime\prime\prime\frac{3}{2}} to u^{(4)}=u^{\prime\prime\prime\frac{3}{2}} . • (24,5) , \alpha=\frac{b+1}{b} , listed in Table 5 of Appendix I, where it is noted that the transformation \bar{x}=K^{2}x,\;\bar{u}=K^{\frac{3}{b}}u maps the canonical form \bar{u}^{(4)}=K\bar{u}^{\prime\prime\prime\frac{1-3b}{1-2b}} to u^{(4)}=u^{\prime\prime\prime\frac{1-3b}{1-2b}} . • (24,5) , r=3 , listed in Table 5 of Appendix I, where it is observed that the transformation \bar{x}=\frac{-6}{K}x,\;\bar{u}=\frac{1296}{K^{3}}u maps the canonical form \bar{u}^{(4)}=Ke^{-\bar{u}^{\prime\prime\prime}/6} to u^{(4)}=e^{u^{\prime\prime\prime}} . • (5,5) , listed in Table 5 of Appendix I, where it is noticed that the transformation \bar{x}=x,\;\bar{u}=K^{-\frac{3}{2}}u maps the canonical form \bar{u}^{(4)}=\frac{5}{3}\frac{\bar{u}^{\prime\prime\prime 2}}{\bar{u}^{\prime\prime}}+K\bar{u}^{\prime\prime\frac{5}{3}} to u^{(4)}=\frac{5}{3}\frac{u^{\prime\prime\prime 2}}{u^{\prime\prime}}+u^{\prime\prime\frac{5}{3}} . • (27,5),r=1 , listed in Table 5 of Appendix I, where it is seen that the transformation \bar{x}=x,\;\bar{u}=K^{-\frac{3}{4}}u maps the canonical form \bar{u}^{(4)}=\frac{4}{3}\frac{\bar{u}^{\prime\prime\prime 2}}{\bar{u}^{\prime\prime}}+K\bar{u}^{\prime\prime\frac{7}{3}} to u^{(4)}=\frac{4}{3}\frac{u^{\prime\prime\prime 2}}{u^{\prime\prime}}+u^{\prime\prime\frac{7}{3}} . The objective of this work is to present an invariant characterization of non-linearizable fourth-order ODEs admitting a five-dimensional Lie point symmetry subalgebra, as listed in Table 1. A principal challenge in this work lies in the implementation of Cartan’s Equivalence Method, primarily due to the well-known issue of expression swell. In this context, the associated coframes become highly intricate, rendering the computational process extremely demanding. To address this challenge, we employ the Inductive Cartan Equivalence Method . This method significantly simplifies the resulting expressions. Despite this simplification, the computations still remain complicated. Consequently, we further introduce a novel framework for Cartan’s method based on a branching strategy via differential relative invariants. This is complemented by introducing a chain of auxiliary functions to systematically manage and simplify the resulting expressions. This article is arranged as follows. In the following section, the Inductive Cartan’s equivalence method is applied to scalar fourth-order ODEs via the action of the Lie group of point transformations, employed to obtain four invariant coframes associated with four distinct branches. These coframes are then utilised to characterize all fourth-order ODEs presented in Table 1 . Section 3 outlines the main theorems. Section 4 outlines a procedure for constructing corresponding point transformations from the constructed invariant coframes, illustrative examples are presented to demonstrate the principal theorems and the paper concludes with a summary. 2 Implementation of Cartan’s equivalence method We refer the interested reader to the scholarly works [ 11 , 12 ] for necessary definitions, notation and standard ideas employed in this section. Throughout this paper, the 1-forms \pi_{i},~i=1,...,12 denote the modified Mourer-Cartan forms derived from the general solution of the linear absorption system. Let \mathbf{J}^{3} denote the third-order jet space equipped with local coordinates \left(x,u,p=u^{\prime},q=u^{\prime\prime},r=u^{\prime\prime\prime}\right)\in\mathbb{R}^{5} . The following defines an adapted coframe on M=\mathbf{J}^{3} . (2.1) \left(\begin{array}[]{l}\omega^{1}\\ \omega^{2}\\ \omega^{3}\\ \omega^{4}\\ \omega^{5}\\ \end{array}\right)=\Omega\left(\begin{array}[]{c}du-pdx\\ dp-qdx\\ dq-rdx\\ dr-fdx\\ dx\\ \end{array}\right), where \Omega denotes the nonsingular transition matrix given by (2.2) \Omega=\left(\begin{array}[]{ccccc}1&0&0&0&0\\ -\frac{1}{2}I_{0}&1&0&0&0\\ \frac{3}{10}(I_{1}+I_{2})&-\frac{1}{2}I_{0}&1&0&0\\ 0&\frac{1}{10}(-3I_{1}+7I_{2})&0&1&0\\ 0&0&0&0&1\\ \end{array}\right), wherein I_{0},~I_{1},I_{2} are functions of x,u,p,q,r