An Improved NSGA-? Algorithm To The Many-Objective Optimal Power Flow Of Power System

With the increasing environmental awareness of people and the growing requirement of power supply quality,the optimization of a single objective only considering economy has been not in accord with the demand of operation and management in modern power system.The many-objective optimization of optimal power flow(OPF)considering economy,environmental protection,security,reliability and stability aiming at minimizing the fuel cost and active power loss and improving the security and reliability has become the research hot in contemporary power system.It is significantly important to optimize the many-objective optimal power flow,which can enhance the stability and reliability of power system.The many-objective OPF problem is a static non-linear,multi-constraint,large-scale,complex optimization problem.In terms of this complex problem,the following research work has been done in this article.(1)A mathematical model for many-objective optimal power flow is established.The multi-objective OPF problem only considering two or three optimized objectives can't completely reflect the security,economy,environmental protection and reliability of power system.A mathematical model for many-objective optimal power flow is established to minimize the fuel cost,pollutant emission,bus voltage deviation of system load,active power loss and maximize the stability of the voltage while satisfying equality and inequality constraints.(2)An improved NSGA-? algorithm is proposed to solve the many-objective OPF problem.NSGA-? algorithm is specially designed to solve the many-objective optimized problem.Although the effective of NSGA-? algorithm is proved to solve the standard constraint test functions,the performances of NSGA-? in solving the practical problem especially the optimization problem of power system are still rarely verified.A Newton-Ralfson's method is used to deal with the power balance constraint integrating the complex constraint characteristics of the many-objective OPF problem.The non-domination selection strategy based on constraint violation is used instead of using penalty function and a fuzzy group method is utilized to obtain the best compromise solution.The effectiveness and feasibility of the proposed improved NSGA-? approach is evaluated on different test cases of a famous standard IEEE 30-bus and IEEE 57-bus.In addition,The numerical results obtained by an improved NSGA-? method are compared with the traditional NSGA-? and other algorithms illustrated in the literature.(3)A coarse-grained NSGA-? algorithm based on MPI framework is proposed to solve the many-objective OPF problem in power system.The computation complexity of associating population member with a reference point is high and time-consuming.In order to solve this problem,a improved NSGA-? algorithm based on MPI is achieved by making full use of multi-CPU.The effectiveness and feasibility of the proposed improved NSGA-? approach is studied and evaluated on different test cases of a famous standard IEEE 30-bus.(4)A decision support system prototype for the many-objective optimization of power system is designed and developed based on the proposed mathematical model and NSGA-? algorithm.