Research On Reactive Power Optimization Of Power System Based On The Improved Cuckoo Search Algorithm

With the development of the power industry,the scale of the power grid is becoming more and more complex,and it is increasingly important to ensure the economic,secure and steady operation of the power grid.The distribution of reactive power flow will directly affect the safety,stability and economy of the power system.Reactive power optimization is an important measure to improve the voltage quality and reduce the active power losses.In this paper,the cuckoo search(CS)algorithm,which can overcome the shortcomings of traditional methods,is selected to solve the optimal reactive power dispatch problem.Firstly,the related concepts,mathematical models and the solving methods of power flow calculation are introduced,and on this basis,the mathematical model of reactive power optimization problem is established.Then,the update process of the original CS algorithm is summarized,and the improvement of its own evolution mechanism and the teaching mechanism are proposed.In the improvement of its own evolution mechanism,an adaptive adjustment strategy is introduced,so that the step control factor?and the probability P_a of the host bird's discovery of the cuckoo's egg change with each individual's current fitness value.At the same time,the global convergence guidance strategy is added to the search equation.The optimal solution guiding the individual to fly towards the optimal individual,it improved the efficiency of algorithm search.An adaptive adjustment strategy combined the global convergence guidance strategy forming Modified Cuckoo Search(MCS)algorithm.Based on MCS algorithm,teaching mechanism is adding,enable CS to enhance local search capabilities while maintaining the original global search capabilities.The improvement of the evolution mechanism of CS combined teaching mechanism forming Teaching Modified Cuckoo Search(TMCS)algorithm.Then the flow chart of applying TMCS algorithm to solve reactive power optimization is drawn in detail.Next,there are many shortcomings in dealing with state variable constraints by applying the traditional penalty function method.Therefore,a new constraint handling method Feasibility-prior Rule(FPR)is proposed and combined with the TMCS algorithm to form the Teaching Modified Cuckoo Search with Feasibility-prior Rule(FPR-TMCS).Meanwhile,the penalty function method is combined with the TMCS algorithm to form a Teaching Modified Cuckoo Search with Penalty Function(PF-TMCS);the FPR is combined with the CS algorithm to form a Cuckoo Search with Feasibility-prior Rule FPR-CS).Finally,in order to verify the superiority of the Feasibility-prior Rule(FPR)and the improved algorithm TMCS,the proposed FPR-TMCS?PF-TMCS?FPR-CS in this paper and the original PF-CS algorithms are respectively applied on MATLAB software for reactive power optimization simulation experiment which includes IEEE 30 buses,IEEE57 buses and IEEE 118 buses three different test systems.The test case includes 7different cases,and the 3 different objective functions including minimize transmission active power losses,minimize voltage stability index and minimize voltage deviation.The results of the comparison test show that the FPR-TMCS algorithm can improve the convergence performance of the algorithm and obtain a better solution for the objective function.It is also proved that FPR can deal with the state variable constraint problem of reactive power optimization better than the traditional penalty function method PF.Furthermore,compared with the algorithm results in other literatures published in recent years,the FPR-TMCS algorithm also has superiority,indicating that the proposed improved cuckoo search algorithm combined with the new constraint handling method in this paper has certain technological breakthroughs to some extent.