BP Neural Network Application in Wind Turbine Type Selection Based on Particle Swarm Optimization

摘要

It is highly significant for the wind farm to select an appropriate type of wind turbine. Appropriate wind turbine not only can decrease the wind farm's total investment in a prior period but also can reduce the operating costs after wind farm goes into operation and increase the generating capacity. The paper mainly improves the traditional BP neural network algorithm through particle swarm optimization algorithm, i.e. PSO, and establishes the model of BP neural network based on particle swarm optimization. By using PSO algorithm to optimize the BP neural network's connection weight values and threshold values, it can give full play to the global optimization ability of the PSO and BP algorithm local search advantage as well as overcome the randomness problem of BP neural network weight values [2]. Now the instance verification results of the wind turbine type selection show that the BP neural network based upon PSO applied into choosing the type of wind turbine in the wind power plant, can achieve the purpose of selecting the suitable wind turbine in wind farm. The model has two advantages: the first is the convergence speed is very fast in the operation process and the second is the computation results have a higher precision.