Reactive Power Optimization Based on Improved Gradient ParticleSwarm Optimization Algorithm

As to the low convergence speed and local optimization of gradient particle swarm optimization (GPSO) algorithm, the improved GPSO algorithm is presented and applied to reactive power optimization. The improved GPSO algorithm has such characteristics as fast convergence speed by dynamical inertia weight regulation method, global optimal resolution by comined action of negative gradient direction mutation and dimension mutation, and higher convergence accuracy. The model of reactive power optimization is established by taking consideration of minimizing network losses, and simulation is carried out on IEEE 30 and IEEE 57 system, better global optimum solution can be got by improved algorithm.