Extreme Process Learning Machine Based on Improved Shuffled Leapfrog Algorithm and Its Application

In this paper, an extreme learning machine model with process input is constructed. The improved shuffled leapfrog algorithm based on three evolutionary strategies autonomous selection is proposed to optimize the number of hidden nodes and network weight of extreme learning machine model. The evolutionary behavior of the worst individuals is determined by calculating the immediate value, the future value and the comprehensive reward of each evolutionary strategy during each iteration. The experimental results show that, compared with five improved shuffled leapfrog algorithms, the proposed algorithm has the best optimization effect in six high-dimensional function optimization and oil well fault diagnosis. It has good practical application value.