基于改进自适应卡尔曼滤波的电力大负荷预测与计费研究

Research on the prediction of power heavy load metering and charging based on improved adaptive kalman filter

摘要:

电力大负荷预测是电力公司进行高效电力系统规划和运行的重要基础。为了提高电力负荷预测精度进而更加有效地估计电力计量与计费，创新地提出了一种基于改进的自适应卡尔曼滤波（AKF，Adaptive Kalman Filter)的电力大负荷计量计费预估方法。分析了电力负荷预测研究现状，针对传统卡尔曼滤波算法不足，引入自适应遗忘因子对卡尔曼滤波算法进行改进，建立数学模型、整定因子调整模型关键参数，得到电力大负荷数据的预测值，最终通过计量计费转换公式得用电量以及电费计量预估值。仿真结果表明：基于AEKF的电力大负荷计量预测方法的负荷预测结果与实际结果误差小于1.35%，电力计费预测结果与实际结果相对误差小于1.263%。应用实例证明：基于AEKF的电力大负荷计量计费预估方法，能够提高电力公司的调度效率12%，增加电费营收5.3%—12.2%。

Abstract:

Large load forecasting is an important basis for efficient power system planning and operation of power companies. In order to improve the accuracy of power load forecasting and then estimate power metering and billing more effectively, a novel method of power bulk load metering and billing forecasting based on improved adaptive Kalman Filter (AKF) is proposed. The research status of power load forecasting is analyzed. Aiming at the shortage of traditional Kalman filtering algorithm, the adaptive forgetting factor is introduced to improve the Kalman filtering algorithm. The mathematical model and the key parameters of the setting factor adjustment model are established to obtain the predictive value of power load data. Finally, the power consumption and the predictive value of power charge metering are obtained through the metering and billing conversion formula. The simulation results show that the error between the prediction results and the actual results of the AEKF based power heavy load metering prediction method is less than 1.35%. The application example shows that the AEKF based method for estimating the metering and charging of large load of electric power can improve the dispatching efficiency of electric power companies by 12% and increase the revenue of electric charges by 5.3% - 12.2%.