Development of Multi-verse Optimizer (MVO) for LabVIEW

Abstract

LabVIEW is a versatile tool with various inbuilt toolkits to perform various measurement and control tasks. Hence, it is used in almost every field of engineering. However, it does not provide enough contribution in the field of optimization which is the major concern. It has only one optimizer based on differential evolution (DE) algorithm. Even though DE is a very effective global optimization technique, but its performance highly depends on parametric settings. DE contains high number of user-defined parameters; therefore, it becomes cumbersome for user to obtain best parametric settings for a given optimization problem. Recently, several nature-inspired algorithms are developed with reduced number of parametric settings to obtain the optimum solutions while solving complex black box optimization problems. Hence, to update the LabVIEW in the field of optimization, there exists a need of continuous development of other efficient global optimizers. Multi-verse optimizer (MVO) is considered as one of the latest but effective nature-inspired optimization algorithm with only two user-defined parameters. In this paper, MVO toolkit is developed for LabVIEW platflorm and the efficiency of the proposed toolkit is validated on a test bed of five standard benchmark functions. The statistical analysis of results shows that the MVO is far better in solving optimization problems as compared to DE.