Research on Short-term Power Load Time Series Forecasting model Based on BP Neural Network
Time series forecasting is an important aspect of dynamic data analysis and processing, in science,economics, engineering and many other applications there exists using the historical data to predict the problem of the future, and is one considerable practical value of applied research. Time series forecasting is an interdisciplinary study field, this paper is under the guidance of the introduction of artificial neural network and time series prediction theory, and then take artificial neural network into time series prediction in-depth theory, method and model studies. Power system load forecasting is an important component of power generation scheme, and is the basis for reasonable arrangements for scheduling operation mode, unit commitment plan, the exchange of power schemes, so the accuracy of load forecasting whether good or bad will be directly related to the industrial sectors economic nterests. In addition, the load forecasting is also conducive to the management of planning electricity, the fuel-efficient, lower cost of power generation; formulating a reasonable power construction plan to improve the economic and social benefits power system. So the forecasting load is necessary.First, we set BP neural network model, and predict the specific time load, and the predicted results are very satisfactory. We can test that BP neural network time series forecasting model has good predictive ability and better promotion of ability. And we also test that the effectiveness and universality of BP neural network time series forecasting model.
BP neural network short-term electric power load time series prediction
NTU Dongxiao SHI Hui LI Jianqing WEI Yanan
School of Business Management North China Electric Power University Beijing China
国际会议
The 2nd IEEE International Conference on Advanced Computer Control(第二届先进计算机控制国际会议 ICACC 2010)
沈阳
英文
509-512
2010-03-27（万方平台首次上网日期，不代表论文的发表时间）