基于GLRM模型和MC误差修正的中长期负荷预测

Medium-and Long-Term Load Forecasting Based on GLRM and MC Error Correction

为解决传统灰色模型(GM)忽略线性因素的不足和预测随机波动性大的序列误差偏大问题,提出了一种基于灰色线性回归模型(GLRM)与马尔科夫链(MC)的中长期负荷预测模型。通过搭建GLRM预测模型,分析模型拟合误差的转移规律,提出基于MC的预测误差定量估计方法,并在此基础上建立GLRM模型预测值的修正模型,构建GLRM-MC模型。实例仿真结果表明,该模型与GM模型和GLRM模型相比,能够更好地把握实际负荷的内在变化规律,可以在提高模型预测精度的同时,提升拟合和预测效果的稳定性。

To remedy the defects of traditional grey model （GM） for ignoring linear factors and having large errors when forecasting the sequences with large random fluctuation in medium and long-term load forecasting, a model which is based on grey linear regression model （GLRM） and Markov chain （MC） is proposed. In this work, the GLRM prediction model is built. A quantitative prediction error estimation method is proposed through analyzing the transfer rule of the model fitting error, a correction model is established consequently for the predicted values of the GLRM model, and then the GLRM-MC model is created. Comparing with GM （1,1） and GLRM, simulation results demonstrate that the proposed model can better grasp the inherent regularity of the actual load, and improve the prediction accuracy of the model, meanwhile, enhance the stability of fitting and forecasting effect.