Artificial neural network-infused polar fox algorithm for optimal design of vehicle suspension components

Optimizing real-world engineering design challenges is inherently complex and difficult, especially when optimal solutions are expected. To this end, the creation of new and efficient optimization algorithms is not an option but a necessity. This paper presents an improved version of the recently developed Polar fox optimization technique. The addition of dynamic adversarial learning improves the dynamic adversarial learning Polar fox optimization algorithm by improving the performance of the algorithm to optimize real-world optimization problems not only very quickly but also accurately. Using test problems from the field of engineering disciplines, such as car crash test, welded beam structure, three-bar truss, and cantilever beam problem, the new optimizer known as the modified Polar fox optimization algorithm (MPROA) was validated before being used to optimize an automobile suspension arm. MPROA achieved superior results in achieving the goal quickly and accurately and proved its potential to solve complex engineering problems. Moreover, the comparison will also reveal the power of the MPROA developed in this work to tackle multiple issues that constrained the reach of a globally optimal solution.