Improved Marine Predators Algorithm And Its Application In Power System

Marine predators algorithm is a new metaheuristic algorithm that imitates the foraging behavior of marine predators.It has the characteristics of simple structure,few parameters,fast convergence and high search efficiency,which provides a new idea for solving real-life optimization problems.However,the algorithm still has some problems such as low optimization accuracy and easily falling into local optimization,and its research and application are still in the initial stage.Therefore,this thesis proposes two different versions of the marine predators algorithm to improve some of the shortcomings of the marine predators algorithm,which are respectively applied to solve the function optimization problem and the optimal reactive power dispatch problem with load demand and renewable energy resources uncertainties.The main research contents of this thesis are as follows:(1)To better improve the optimization accuracy and convergence speed of the marine predators algorithm,an improved marine predators algorithm based on group learning is proposed.An opposition-based learning method is adopted to enhance the quality of the initial solutions.Then,a group learning strategy is used to diversify the population.Two subgroups are produced by fitness evaluation and employing different updating mechanisms.In addition,a new position-updating strategy based on information exchange is used to help the proposed algorithm escape from the local optima in the later stage of iteration.To evaluate the effectiveness and feasibility of the improved algorithm,it is applied to solve CEC2017 benchmark functions.The simulation results show that the proposed algorithm is superior to other comparative algorithms in solving accuracy and convergence speed.(2)To solve the optimal reactive power dispatch problem of power system with load demand and wind-solar power uncertainties,a hybrid improved marine predators algorithm is proposed.In the initialization phase,an opposition-based learning method is adopted to improve the quality of initial population.In the unit velocity ratio phase,introducing the mutation strategy of differential evolution algorithm to balance the exploration and exploitation of the algorithm.In the low velocity ratio phase,using the spherical search algorithm to improve the development ability of the algorithm.Besides,the constrained technique is adopted to deal with the constraints in the optimization problem,and the IEEE 30-bus system with basic configuration is selected for case studies of deterministic optimal reactive power dispatch.Simulation results show that the proposed algorithm has better accuracy and stability.In addition,normal distribution,Weibull distribution,and lognormal distribution probability density functions are used to model the uncertainties of load,wind speed,and solar irradiance,respectively.Finally,the Monte Carlo simulation method is used to create various scenarios,with the minimum expected power loss and expected voltage deviation as the objective function.The improved IEEE 30-bus system verifies the superiority and effectiveness of the proposed algorithm in solving the stochastic optimal reactive power dispatch problem.