论文部分内容阅读
电力负荷预测的实质是对电力市场需求的预测,是利用以往的历史数据资料找出电力负荷的变化规律,进而预测负荷在未来时期的变化趋势.由于经济、气候以及工业生产等诸多因素的约束和限制,电力负荷预测精度很难提高.一个好的实用的电力负荷预测模型则要求既能充分利用负荷的历史数据,又能灵活方便地综合考虑其他多种相关因素的影响.提出了回归与自回归模型相结合的时间序列混合回归预测模型,它的待估参数由BP神经网络进行修正,经实例验证,预测效果良好.
The essence of power load forecasting is to predict the demand of power market by using the historical data to find out the changing law of power load and then to forecast the change trend of load in the future.Because of many factors such as economy, climate and industrial production, And limit the power load prediction accuracy is difficult to improve.A good and practical load forecasting model requires both to make full use of historical load data, but also flexible and easy to take into account the impact of a variety of other relevant factors.Return to regression Regression model combined with time series mixed regression prediction model, the estimated parameters to be revised by BP neural network, the example validation, the prediction effect is good.