Research On Forecasting Model And Algorithm Of Electricity Price In The Electric Power Market

With the spread of market-oriented reform in the electric power industry around the world, electricity price issues have been the key problems in the markets, and how to formulate a reasonable price for electricity (a special commodity) is critical for the smooth operation of the electricity market. Thus the prediction of the future electricity price by using the relevant historical data has been a very meaningful work.Artificial Neural Network (ANN) is widely used in the area of prediction, owning to its highly nonlinear fitting ability. Given the disadvantages of the traditional neural network algorithm, the Particle Swarm Optimization (PSO) algorithm based on the random global optimization is inducted into the network training, a new electricity price forecasting model PSONN is presented in this paper.The development of chaos theory not only provides a new idea for electricity price predicting, but also gives a new acquaintance to the complexity of the price series. The analysis based on chaos theory can draw out the inner rules in chaotic time searies, which guranteed the objectivity of prediction by avoding to frame a subjective model.With comprehensive considerations of the fluctuation rules and the various influencing factors on the forming of price in the power market, a short-term electricity price forecasting method using PSONN based on chaos theory is proposed. Finally, the proposed method is verified by testing on the the pratical data of California electricity power market, and the predictions verified the method's efficiency.