A Comprehensive Review of Optimization Algorithms for Nonlinear Systems

Abstract

The practical problems are full of nonlinearity and uncertainty due to their dynamic behavior. The uncertainties, parameters variations and other constraints make nonlinear systems very complex. To deal with such system dynamics and uncertainties, soft computing techniques are widely used. Optimization algorithms are one of the most effective and simple soft computing techniques.In literature, several controllers and control mythologies are being discussed to evaluate the system performance. Optimization algorithms are one of the popular soft computing techniques, used with different controllers like conventional PID, fuzzy logic, ANN, and many others to handle system nonlinearities and enhance the performance. These optimization algorithms not only improve the performance effectively but also gave robust response towards the nonlinearities. The proper selection of algorithm is a very important aspect to find the best solution. Here three categories of algorithms i.e. from swarm-based algorithms particle swarm optimization (PSO), grasshopper optimization algorithm (GOA), grey wolf optimization (GWO), whale optimization algorithm (WOA), gravitational search algorithm (GSA) from physics based and teaching learning based optimization (TLBO) from human-based have been reviewed for nonlinear systems. Most of the algorithms are suffering from either of abilities i.e. exploration or exploitation so sometimes they are not able to give optimal solution. To overcome with this problem, recently hybrid approach of algorithms has been widely used in which the better sides of the individual algorithms are utilized.