| Summary: | Metaheuristic optimization algorithms are well-established techniques to address those problems which are difficult to solve through traditional optimization methods. Among the various kinds of optimization algorithms, Simulated Kalman Filter (SKF) is a new population-based optimization algorithm inspired by the estimation capability of Kalman Filter. Based on the mechanism of Kalman filtering, the SKF includes prediction, measurement, and estimation process to search for global optimum. The SKF has been shown to yield good performance in solving benchmark optimization problems. However, the exploration capability of SKF could be further improved. From literature, Opposition-based Learning (OBL) has been employed to increase the diversity (exploration) of search algorithm by allowing current population to be compared with an opposite population. By employing this concept, more potential agents are generated to explore more promising regions that exist in the solution domain. Therefore, this research intends to improve the exploration capability of SKF through the application of OBL. The OBL is employed after the estimation process of SKF. Two versions of OBL techniques have been considered in this research, which are original OBL and Current Optimum Opposition-based Learning (COOBL). Experimental results over the IEEE Congress on Evolutionary Computation (CEC) 2014 benchmark functions indicate that Opposition-based Simulated Kalman Filter (OBSKF) has made some improvement towards exploration capability of SKF, while the Current Optimum Opposition-based Simulated Kalman Filter (COOBSKF) improved the exploration capability of SKF significantly. The COOBSKF also has been compared with five other optimization algorithms and outperforms them all. Besides that, this thesis also presents the application of COOBSKF in a system identification problem. The overall performance is evaluated based on six case studies. According to the experimental results, COOBSKF provides average of maximum model validation up to 90%. This technique can be an alternative approach to solve system identification problem, apart from using conventional method.
|
|---|