基于灰色模型和Kalman平滑器的多母线短期负荷预测

Short-Term Bus Load Forecasting Based on Grey Model and Kalman Smoother

提出一种基于灰色模型和Kalman平滑器的多母线短期负荷预测方法。首先利用频域分解消去母线负荷序列的周周期分量,建立序列的灰色模型;利用系统负荷预测方法得到系统负荷预测值。然后基于灰色模型,将各母线负荷的累加序列作为状态,系统负荷的累加序列作为观测,建立线性离散随机系统模型,利用Kalman平滑器计算各母线负荷累加序列的最优估计值,最后经过累减还原并加上周周期分量得到母线负荷的预测值。Kalman平滑器利用高准确率的系统负荷预测结果对母线负荷预测进行调整,降低预测误差。在实际系统中进行了仿真验证,证明了该方法的有效性。

A short-term bus load forecasting method based on grey model and Kalman smoother is proposed.The weekly component of the load of each bus is subtracted and the grey model is constructed.The system load is forecasted.Then based on the grey model,a linear discrete random model is formed.The bus load cumulative series are the states and the system load cumulative series is the observation.The Kalman smoother is utilized to compute the optimal estimate of the bus load cumulative series.The forecast of bus load is the sum of the weekly component and the result of successive subtracting of the cumulative series.The Kalman smoother uses the system load forecasting result to adjust bus load forecasts and reduces the errors.The validity of the method is verified by simulation in a practical system.