Optimal Allocation Of Reactive Power Compensation For Distribution Network Based On Improved Particle Swarm Optimization

Unreasonable distribution of reactive power will decrease voltage quality, increase network losses and reduce power system stability. The reactive power optimization can integrate existing resources, maximum increase economic benefit of the system and customer. So that the study of the problem of reactive power optimization has the great significance in the theory and practical application. After this paper analyzes the advantages and disadvantages of the existing optimization algorithm; an improved particle swarm optimization algorithm for the mathematical model corresponds to the actual calculation is optimized.The advantages of simplicity and easy implementation of particle swarm algorithm have been validated in science and engineering fields. However, the weaknesses of particle swarm algorithm are the same as other evolutionary algorithm’s, such as easy to fall into local minimum, premature convergence, etc. The reasons of disadvantages of Particle Swarm Optimization (PSO) Algorithm were analyzed, and a Chaotic Sequence-Cosine PSO (CS-CPSO) algorithm has been proposed. It makes the initial particles to traverse the entire search space to use the population-initialized of chaotic sequences in the algorithm. And that increases the diversity of initial population. It makes the particles in the early to have the ability of stronger global search that the inertia weight of the SPSO was changed because of cosine functions nonlinearity. Along with the frequency of iterations increasing, the inertia weight decreases, and then the local search capability of particle reinforced. The accuracy of the algorithm improves. The particles have strong ability of social learning at an early stage because the learning factor was changed by the cosine function of nonlinear symmetric. Other particles will draw close to the optimal rapidly. Particle itself learning ability is enhanced latterly. It speeds up the convergence of the algorithm; Bacterial chemotaxis has been introduced in the CS-CPSO algorithm, which maintains the diversity of population, and prevents the particles fall into local optimum in a certain extent.The CS-CPSO algorithm is simulated and analyzed by five test functions, compared with the original particle swarm optimization algorithm and standard particle swarm optimization algorithm, results showed that:CS-CPSO algorithm can jump out of local optimum in a certain extent, effectively avoid the premature convergence problem in SPSO algorithm, and also have a faster convergence speed. Simulation software of Matlab7.10is used to write the main program of reactivepower optimization based on CS-CPSO algorithm. The results of the classicalIEEE-30node distribution network system prove that CS-CPSO algorithm is usefulin solving power system reactive optimization, and ensure the voltage quality ofpower grid and reduce the system network loss, obtained the good economic benefit.