Adaptive parameter control of search group algorithm using fuzzy logic applied to networked control systems

Search group algorithm (SGA) is one of the newest nature-inspired heuristics for solving different engineering optimization problems. Like other evolutionary algorithms, SGA suffers from the parameters tuning, which is considerably dependent on the problem. The purpose of this paper is to introduce an adaptive parameter control using fuzzy logic, namely fuzzy SGA (FSGA), for enhancing the solution quality of the basic SGA. In FSGA, a fuzzy system is incorporated to dynamically adjust the control parameter value with respect to normalized iteration and normalized error value, which are the inputs of the system. To evaluate the performance of FSGA, firstly, it is compared against those of state-of-the-art algorithms over the well-known benchmark functions. Null hypothesis significance testing is then applied to make algorithm ranking. Finally, in order to demonstrate the potential applicability of FSGA in the field of control, it is adopted to design of robust proportional-integral-derivative controller for the network-based control system dealing with time delays existed in the communication channel. The results verify the feasibility of the proposed FSGA.