Department of Mathematical Sciences, College of Science, United Arab Emirates University, P. O. Box 15551, Al Ain, UAE.
Sci Rep. 2023 Jul 6;13(1):10932. doi: 10.1038/s41598-023-37466-6.
In this study, a trustworthy swarming computing procedure is demonstrated for solving the nonlinear dynamics of the Rabinovich-Fabrikant system. The nonlinear system's dynamic depends upon the three differential equations. The computational stochastic structure based on the artificial neural networks (ANNs) along with the optimization of global search swarming particle swarm optimization (PSO) and local interior point (IP) algorithms, i.e., ANNs-PSOIP is presented to solve the Rabinovich-Fabrikant system. An objective function based on the differential form of the model is optimized through the local and global search methods. The correctness of the ANNs-PSOIP scheme is observed through the performances of achieved and source solutions, while the negligible absolute error that is around 10-10 also represent the worth of the ANNs-PSOIP algorithm. Furthermore, the consistency of the ANNs-PSOIP scheme is examined by applying different statistical procedures to solve the Rabinovich-Fabrikant system.
在这项研究中，展示了一种可靠的群体计算程序，用于解决 Rabinovich-Fabrikant 系统的非线性动力学。该非线性系统的动态取决于三个微分方程。基于人工神经网络 (ANNs) 的计算随机结构以及全局搜索粒子群优化 (PSO) 和局部内点 (IP) 算法的优化，即 ANNs-PSOIP，用于解决 Rabinovich-Fabrikant 系统。通过局部和全局搜索方法优化基于模型微分形式的目标函数。通过实现和源解决方案的性能来观察 ANNs-PSOIP 方案的正确性，而接近 10-10 的可忽略的绝对误差也代表了 ANNs-PSOIP 算法的价值。此外，通过应用不同的统计程序来解决 Rabinovich-Fabrikant 系统，检查了 ANNs-PSOIP 方案的一致性。
