Research On Reactive Power Optimization Algorithm Of AC/DC Hybrid System

With the increase demand for total electricity consumption,the traditional AC transmission system is gradually deficient,emerging problems such as power quality degradation and low power utilization brought by long-distance and large-capacity transmission.Pure AC gradually fails to meet the power demand,so the DC system is introduced as a supplement to AC transmission,which are combined together to constitute modern power transmission system.Reactive power optimization is one of the main ways to reduce power loss and improve power quality in the transmission of the electrical power system.It is a nonlinear,multivariate and multiconstrained mixed programming problem,whose system includes continuous variables and discrete variables,and the optimization process is very complicated.At present,the most popular algorithm is interior point method.This method adds disturbance factors into the objective function,and introduces the relaxation variable to form a new barrier function,which transforms the inequality restrictions into the equality restrictions,so greatly simplifies the calculation.However,the traditional reactive power optimization algorithm can’t process the discrete variables in the system accurately,so it is prone to occur in non-convergence in calculation,higher network loss,and unreasonable voltage distribution.First of all,aimed at non-convergence in algorithms of traditional optimization process and based on the predictor-corrector interior point method,this paper proposes a nonlinear prediction correction interior point method on the basis of variable conversion method.This algorithm preprocesses variables and casts discrete variables and continuous variables into continuous nominal variables in the same interval,and then put the converted nominal variables into the subsequent reactive power optimization calculation;at this time,it only focuses on the optimization problem contained continuous variables and the optimal solution of nominal variables will be obtained after the calculation.It will provide the optimal solution through inverse transformation to the original variable type.The prediction-correction interior point method based on the method of variable conversion proposed in this paper will not produce non-convergence in the calculation process,the algorithm is simple,the structure of reactive power optimization will not change,and the network loss will be significantly reduced after optimization,which is valuable in practical application.Secondly,considering the safety and economy of the system operation,taking the minimum active power loss as the objective function,making use of variable relationship introduced by DC system on the basis of pure AC system,it forms mathematical model of reactive power optimization of AC/DC hybrid transmission system.Finally,the MATLAB simulation software is used to modify and simulate IEEE-14 node test system and IEEE-30 node test system,then compare and analyze the results,which can verify the feasibility and superiority of the prediction-correction interior point method based on method of variable conversion in reactive power optimization of AC/DC hybrid transmission system.