基于人工神经网络的短期风电功率预测研究-电力系统及其自动化专业论文.docx

AbstractTitle：SHORTITERM Abstract Title：SHORTITERM WIND POWER FORECAST BASED ON NEURAL NETWORKS Major：Electrical power system and automation Name： Jian YANG Signature：狴细 Supervisor：Associate prof．Kaoshe ZHANG Signature： Abstract Along with the energy shortage，and the severe energy supply security situation，ne w energy such as wind power is developing rapidly,and the proportion of wind power in the西d is continuously increasing．However,compared with the hydropower and ther mal power．the wind power iS intermittent、randomness and uncontrollable，large—scale W ind power integrated into power system will bring severe challenges of power system sa fety operation and power quality．Wind power forecasting technology carl guide the disp arching of grid and the production planning of wind farm effectively．Therefore，it is ur gent to carry out in-depth study for wind power output prediction technology，and for wind power of comparatively accurate prediction． This paper studies short-term wind and power prediction．The statistical rules of parameters from the willd farm in West China are acquired after pretreatment，and the paper analyses wind speed’S chaotic characteristics．Firstly compare dynamic neural network with the static neural network model，consider the wind，temperature，pressure on the influence of the wind speed，a short-term wind speed forecasting model based on Elman regression neural network was presented to model wind speed for 1 hour and 24 hours of short—term forecasting． The model improves the prediction precision，has the ability to adapt to the time-varying behavior． Secondly，use the direct prediction method，using the collected historical data of wind power short-term forecasting．Respectively established the single hidden layer Elman neural network forecast model。double hidden Elman neural network forecast model，three hidden Elman neural network model，and compared with the BP neural network model and RBF neural network model，do the wind farm sho