Mathematical distribution coyote optimization algorithm with crossover operator to solve optimal power flow problem of power system

Abstract

The optimal power flow (OPF) of power system is to optimize the objective function such as generation cost by adjusting control variables on the premise of satisfying operation constraints and supply–demand balance. A mathematical distribution coyote optimization algorithm (MDCOA) is proposed. Crossover operation is added to the algorithm initialization process to improve the randomness and diversity of offspring. In order to improve its search efficiency and balance the process of exploration and exploitation, the step size is generated by random numbers with seven mathematical distributions, including Gaussian distribution, Uniform distribution, Exponential distribution, Beta distribution, Gamma distribution, Weibull distribution and Rayleigh distribution. Firstly, the original COA and seven improved COA were simulated under the CEC2017 functions, wCOA ranks first among these comparison algorithms and was selected to be compared with PSO, CMA-ES, MFO, HHO, HBA and WOA to verify its effectiveness. Aiming at the OPF problem, the simulation experiments are carried out on IEEE 30-bus system and IEEE 57-bus system. The generation cost, active power loss, voltage stability and bus voltage offset are selected as objective functions respectively. The proposed method can find a better solution than the standard COA and eCOA has the best comprehensive performance in all cases. Simulation are compared with ARCBBO, ECHT-DE, I-NSGA-III, MSA and other algorithms and the results show that the MDCOA can effectively solve the OPF problem.