A Smart Strategy for Power Factor Correction of Distribution Network under the Imbalance Condition

The power factor and unbalanced three-phase power are important factors affecting the power quality in the three-phase power supply system. The smart algorithm strategy used as control strategy has the high practical value in terms of implementation of integrated compensation (including the power factor correcting and the three-phase unbalance reduction). Based on the research of the basic principles of the traditional symmetrical component method, we provide a new smart strategy, namely the intelligent predictive optimization for the compensated admittances, to complete the reactive compensation in the imbalance power system. This strategy uses the genetic algorithms (GA) based artificial neural network (ANN) and gradient-particle swarm optimization (Grads-PSO) for the theoretical support. Targeted characteristics of the imbalanced circuits, the input parameters of ANN are obtained by sampling and derived from the symmetrical component method. According to output values of ANN, we construct the fitness function formula in the Grads-PSO, and using Grads-PSO algorithm to seeking for Optimal compensation admittance. Experimental results show that, the smart strategy proposed as a static var compensator (SVC) control strategy in the three-phase power network, not only the power factors can be corrected, but the three-phase unbalance degree is reduced.