On the Use of Diversity Mechanisms in Dynamic Constrained Continuous Optimization

Abstract

Population diversity plays a key role in evolutionary algorithms that enables global exploration and avoids premature convergence. This is especially more crucial in dynamic optimization in which diversity can ensure that the population keeps track of the global optimum by adapting to the changing environment. Dynamic constrained optimization problems (DCOPs) have been the target for many researchers in recent years as they comprehend many of the current real-world problems. Regardless of the importance of diversity in dynamic optimization, there is not an extensive study investigating the effects of diversity promotion techniques in DCOPs so far. To address this gap, this paper aims to investigate how the use of different diversity mechanisms may influence the behavior of algorithms in DCOPs. To achieve this goal, we apply and adapt the most common diversity promotion mechanisms for dynamic environments using differential evolution (DE) as our base algorithm. The results show that applying diversity techniques to solve DCOPs in most test cases lead to significant enhancement in the baseline algorithm in terms of modified offline error values.

Using Neural Networks and Diversifying Differential Evolution for Dynamic Optimisation

“…We applied the most common diversity mechanisms. For a recent survey regarding the effect of diversity mechanisms in dynamic constrained optimisation see [6].…”

“…This method behaves particularly poorly in exp2 and exp3, even compared to noNN No. By promoting diversity unnecessarily, CwN adversely affects the convergence of the algorithm to new optimum position, which is not too distant from the CwN has been reported as one of the best methods for handling dynamic environments in a recent study considering other diversity methods [6]. Conversely, CwN is not competitive in our study, because of the following reasons: Firstly, in the evolution process of this method, we need to calculate distances at each generation, and as we are considering wall clock timing, so the algorithm is left with less number of fitness evaluations per each time.…”

“…In such cases, CwN helps to diversify solutions by avoiding similar individuals in each sub-region of the search space [20]. In addition, in [6], CwN demonstrated superior performance for problems that include features such as disconnected and small feasible areas. Last but not least, in the previous work, CwN was tested on a small problem dimension (2) [6].…”

“…In addition, in [6], CwN demonstrated superior performance for problems that include features such as disconnected and small feasible areas. Last but not least, in the previous work, CwN was tested on a small problem dimension (2) [6]. Since our problem's dimension is large, having a CR of 0.3 alter the offspring in only a number of dimensions.…”

“…Among the many approaches proposed for reacting to the changes, diversity mechanisms [4], [5] are the simplest and most popular. A recent study has revealed how common diversity mechanisms can significantly enhance the performance of a baseline differential evolution (DE) for different environmental changes [6]. Other approaches include memory-based approaches [7], multi-population approaches [8], and prediction methods [9].…”