神经网络在电力负荷预测中的应用

Application of Neural Network in Electrical Load Forecasting

在短期负荷预测过程中,引起负荷变动的因素与负荷之间的非线性映射关系是造成预测结果与实际结果之间存在偏差的原因之一。神经网络具有很强的非线性映射能力和自学习能力。为提高短期负荷预测的精度,基于神经网络的研究方法,设计了预测网络。该网络以洋山深水港东港路10 kV开关站中沈家湾的日负荷数据为样本,对采集电量进行了预处理;然后对其隐层个数及节点个数进行了分析设计;最后对短期日负荷进行预测。对比结果表明,预测值与实际值吻合较好。

In the process of short-term load forecasting, the nonlinear mapping relationship between the factors of the load variation and the load is one of the reasons to leading deviation between the predicted results and actual results. The neural network features strong nonlinear mapping capability and self-learning ability. In order to improve the precision of the short-term load forecasting, the forecasting network is designed. The network is based on the research method of neural network, with the daily load data of 10 kV Shenjiawan Switching Station on East Road of Yangshan Deepwater Port as sample, the collected data of electricity consumption are pre-treated, the numbers of its hidden layers and nodes are analyzed and designed; short-term daily load is forecasted. The predicted result is compared with the actual measured values, the comparison shows that the predicted values are well matched with the actual values.