Energy metering device is one of the prerequisite for metering jobs in power companies. Its accuracy will directly af-fect the economic interests of the companies and users. Abnormal states that always appear to occurring to power metering devices, negative error tolerance particularly, will lead to higher of the line loss. This paper applies the GM (1,1) model to establish a user load forecasting model, then sets the load by forecasting the upper lim-its threshold and lower limits threshold. Comparing the measured load with the upper and lower limits threshold, if the measured val-ue is not between the upper and lower threshold, then sets the user as'subject for investigation', otherwise, then sets it as a regular user. The measured value will be concluded as the samples of the model parameters for re-estimating in order to improve the accura-cy of the model. Cases listed in this paper had proved that the GM (1,1) model has a decent practicability in finding unusual user.%利用灰色GM(1,1)模型建立用户负荷预测模型,并设定负荷预测阈值上限和阈值下限,将用户实际测量值与该用户负荷预测上下限阈值进行对比,若测量值不在上下限阈值之间,则将该电能计量装置列为"排查对象";反之,电能计量装置正常,并将台区实际测量值作为样本对模型的参数进行重新估计,以提高参数的准确性.实例证明,灰色GM(1,1)模型在异常电能计量装置判断中具有较好的工程实用性.
展开▼
