基于UTCI-MIC与振幅压缩灰色模型的用户侧微电网短期负荷预测方法

Short-term Load Forecasting Method for User Side Microgrid Based on UTCI-MIC and Amplitude Compression Grey Model

短期负荷预测是保障用户侧微电网经济、安全运行的基础。现有研究表明,综合考虑气象、地理等影响因素的负荷预测模型在一定程度上提升了预测准确率。但在具有明显城市特征的用户侧微电网负荷预测中存在局限性,其预测结果呈现显著偏差,使得基于该结果的潮流计算偏离实际情况,危及系统的安全运行。针对此问题,提出了一种新型热气候指数–最大信息系数(universal thermal climate indexmaximal information coefficient,UTCI-MIC)与振幅压缩灰色模型的用户侧微电网短期负荷预测方法。首先,该模型采用经验模态分解将用电负荷分解为波动负荷和趋势负荷曲线;其次,建立了考虑相似日的MIC矩阵与涵盖多时刻气象、地理区位、城市特征因素的UTCI热环境评估方法,用于波动负荷预测;最后,将基于振幅压缩灰色模型获得的趋势预测结果与波动预测结果重构,得到用电负荷预测结果。案例验证表明,所提方法可有效预测城市特征明显的用户侧负荷变动情况,其预测准确率可达96.91%以上,为城市电网的能量管理系统和电力市场交易提供重要参考。

Short-term load forecasting is the basis to ensure economic and safe operation of user side microgrid.A lot of studies declared that the model taking meteorological and geographical factors into consideration improves prediction accuracy to some extent.However,there are limitations in the prediction of the user side microgrids with obvious city characteristics,and their prediction results show significant deviation,making power flow calculation deviate from actual situation and endangering safety of the system.Aiming at this problem,a short-term load forecasting method was proposed,combining new universal thermal climate index(UTCI)and maximal information coefficient(MIC)similar days’matrix with amplitude compression grey model.Firstly,the electrical load was decomposed into fluctuating and trend load curves using empirical mode decomposition(EMD).Secondly,an MIC history matrix of load similar days and an UTCI thermal environment assessment method covering meteorological,geographical location and urban characteristics was established for accurate prediction of fluctuating load curve.Finally,the obtained results of fluctuation and trend forecasting were reconstructed using the amplitude compression grey model to obtain final results.Practical cases were used to verify the proposed method.The model proposed in this paper can obtain good performance of accuracy about 96.91%,eminently superior to conventional methods.