Cost Estimation of Wind Farms Based on BP Neural Network System

Currently, the new energy power generation is booming.And wind energy, as a clean and renewable energy, has a relatively low investment cost compared with other new energy sources. Therefore, wind power generation has a broad market prospect and development potential. Considering the construction management and budget of conventional power plants, the difficulties on estimation facing traditional project cost include: high intensity for quota query, poor intelligent computing capacity, and low accuracy. Now estimation is conducted often by the combination use of quota and bill of quantities, but defects are: on one hand, dynamics of project cost is not taken into account, such as, current market prices of materials and inflation; on the other hand, fuzzy membership of features contrast between newly built and typical project sample should be determined, which bears large subjectivity, and the pace will be greatly reduced. Therefore, it is necessary to study proper estimation method for project cost of power plants. Compared with other methods, BP neural network enjoys higher adaptability, better fault tolerance, fastercalculation speed andbetter at solving complex nonlinear modeling problems, etc. It can be more efficient and effective when dealing with multifactorial nonlinear problems. This papergives a detailed introduction forwind power generation estimation from perspectives of cost, estimation method of traditional projects, quantitative description of engineering sample mode, and estimation models based on BP neural network system, etc.