Optimization Algorithm Based On The Emotional Intensity Of The Laws Of Social-emotional And Applied Research

Social emotional optimization algorithm (SEOA) is a novel swarmintelligent population-based optimization algorithm by simulating the humansocial behaviors. This algorithm has a broad sociological and biologicalbackground. It has been successfully applied to practical problems such ascluster structural optimization problem and reactive power optimization inpower system. However, the research and application on SEOA is still inprimitive stage. There are still many problems need to be solved and improved.In order to improve the performance of SEOA effectively, we use the law ofemotional intensity and apply the algorithm to actual optimization problems.Firstly, in this paper we describe SEOA in detail, and then analyze theupdate strategy of emotional value. In SEOA, each individual's emotional valuedecreases in a linear manner. But in the actual process of emotionalexperiencing, the value of emotional intensity should increase whenindividuals are close to the optimal position. According to the law of emotionalintensity, we design a novel update strategy to change emotional values. Theexperimental results show the improved SEOA effectively avoids the prematureconvergence problem.Secondly, with further study of emotional process, we find out emotionalintensity values exponentially decline over time, which would cause thealgorithm fall into local optimum. From this view, we add emotional intensityattenuation strategy to this algorithm. The results show that the improvedalgorithm has the better global and local optimization ability.Finally, we use the improved algorithm in this paper to solve the problem ofrank weights computing for judgment matrix, which can be attributed to theoptimization problem to minimum the consistency index. The global optimumvalue of consistency index can be found through the evolutionary iteration of thealgorithm. Simulation results show the improved algorithm in this paper has agood performance and strong stability compared with other methods.