Mating-Based Manta Ray Foraging Optimization for Fuzzy-Hammerstein Model of an Electric Water Heater

This paper presents an improved variant of the Manta Ray Foraging Optimization (MRFO) algorithm. The optimization method of the original MRFO is a combination of random and spiral strategies. Like other optimization algorithms, MRFO still has a limitation when applied to a complex real-world problem. In this work, mating strategy of a barnacle species is incorporated into the MRFO algorithm. It allows good features of the parent manta ray to be inherited by its offspring thus creating a high-quality population. The proposed algorithm is applied to acquire a dynamic model of an electric water heater. A fuzzy-Hammerstein model is chosen as the candidate model for the water heater considering electrical voltage and water temperature as the input and output responses respectively. The result of the modelling has shown MRFO and the proposed MRFO variant have satisfactorily acquired the dynamic model of the water heater. The improved MRFO variant has tracked the output temperature response more accurately than the original MRFO at costs 425 and 508 respectively.