Research On Reactive Power Optimization Based On Chaos Self-Adaptive Differential Evolution Algorithm

In recent years, with the rapid economic development of our country, the scale of power grid and the electricity demand of residential are constantly increasing. whether the power system is running safe, reliable and economical or not is getting more and more people’s attention. The power system reactive optimization is minimizing the active power loss or some performance indexes with the given load parameters and network structure of power system by adjusting the generator terminal voltage, the gears of the OLTC tap and the gears of the capacitors based on the premise of the quality assurance of system voltage and the satisfaction of all the constraint conditions.In the traditional reactive power optimization model is generally based on active loss minimum as the core target, but less consideration of the system voltage stability. With the continuous development of power system, the increasing of load and the proportion of the long distance power supply, the transmission capacity might be close to the limit at times of peak load that has increased the possibility of voltage collapse and led to blackout. Therefore, this paper studies the multi-objective reactive power optimization with considering the static voltage stability and its mathematical model.For the reactive power optimization problem of power system which is nonlinear, multivariable, multi-constrained conditions and the coexistence of discrete and continuous, a lot of achievements have been obtained with the traditional mathematical optimization method and the artificial intelligence algorithm widely used in recent years, but exists disadvantages more or less as long convergence time, depending on the initial values highly or dropping into local optimum easily. Therefore, this paper adopted the chaos self-adaptive differential evolution algorithm with the self-adaptive improvements on the control parameters and the chaos search, which had passed the performance test. At last, the proposed algorithm CSADE and CSACPMDE applied to single-objective and multi-objective reactive power optimization with considering the voltage stability index by MATLAB programming, analyzed and summarized the relationship between active loss, voltage stability and voltage level by simulating the IEEE14 and IEEE30 nodes standard test systems and a regional power grid system. The optimization results compared by different algorithms show that the proposed algorithms had great advantages on improving the convergence speed, stability and optimizing capacity that made the power system running economically and the voltage running steadily with considering the voltage stability index.