基于实时电价和加权灰色关联投影的SVM电力负荷预测

Power Load Forecasting of SVM Based on Real-time Price and Weighted Grey Relational Projection Algorithm

精准的电力负荷预测有助于保障电力系统的安全调度和稳定运行,支持向量机作为一种良好的预测工具被广泛应用于电力负荷预测。随着智能电网的快速发展,实时电价成为电力负荷的重要影响因素,因此在应用支持向量机进行电力负荷预测时,引入实时电价这一影响因素,同时将加权灰色关联投影算法应用于节假日的历史负荷序列的选择,并采用改进的粒子群算法优化模型参数,最终得到一种实时电力负荷预测方法。以新加坡的电力数据为例进行实时电力负荷预测,并与通过反向传播神经网络得到的预测结果进行对比,结果表明所提出的方法具有较高的精确度和稳定性。

Accurate power load forecasting is helpful to guarantee safe dispatch and stable operation of power system. As a great forecasting tool, support vector machine(SVM) is widely used in power load forecasting. With rapid development of smart grid, real-time electricity price becomes an important factor affecting power load. Therefore, when applying SVM to predict power load, real-time electricity price is introduced as an influencing factor. Meanwhile, weighted grey relational projection algorithm is applied to select historical load sequence of holidays, and the improved particle swarm optimization algorithm is used to optimize model parameters. Based on this, a real-time power load forecasting method is proposed. Finally, Singapore’s power data are taken as an example to carry out real-time power load forecasting, and the predicted results is compared with those obtained with back propagating neural network. Results show that the proposed method has higher accuracy and stability.