电子式三相电网电能表计量误差预测仿真

发布时间：2018-04-20 02:38

本文选题：三相电能表 + 计量　；参考：《计算机仿真》2017年10期

【摘要】：对电子式三相电网电能表计量误差的预测,能够及时的发现电能表计量设备的潜在隐患。对电能表计量误差的预测,需要拟合出电子式电能表计量非线性系统,对适应度函数进行优化,完成电能表计量误差的预测。传统方法通过灰色关联分析理论,选取电能表计量误差训练样本,但忽略了对适应度函数的优化,导致预测结果偏低。提出基于神经网络的电子式三相电网电能表计量误差预测方法。研究电子式三相电网电能表计量设备综合误差的构成,建立用于描述非线性分布的组合半梯度云模型,用于预测电网传输电压和电流互感器的运行误差,利用以获取的电能表计量误差时间序列经过神经网络进行在线学习,拟合出电子式电能表计量非线性系统来预测三相电能表计量值,对适应度函数进行优化,达到电能表计量误差动态预测的目的。实验结果表明,所提方法预测误差较小,具有较强的泛化能力。
[Abstract]:The prediction of the measurement error of the three-phase electric meter in electronic power network can discover the potential hidden danger of the measuring equipment in time. In order to predict the metering error of watt-hour meter, it is necessary to fit the nonlinear system of electronic watt-hour meter and optimize the fitness function to forecast the metering error of the watt-hour meter. The traditional method selects the training sample of measuring error of watt-hour meter through grey correlation analysis theory, but neglects the optimization of fitness function, which leads to the low prediction result. A neural network based method for predicting the metering error of three-phase electric energy meter in electronic power system is presented. In this paper, the composition of the synthetic error of the measuring equipment of the electronic three-phase power meter is studied, and the combined semi-gradient cloud model is established to describe the nonlinear distribution, which can be used to predict the operation error of the transmission voltage and current transformer in the power network. Based on the time series of measurement error of electric energy meter (Watt-hour Meter), the on-line learning is carried out through neural network, and the nonlinear measurement system of electronic watt-hour meter is fitted to predict the metering value of three-phase watt-hour meter, and the fitness function is optimized. To achieve the purpose of dynamic prediction of metering error of watt-hour meter. The experimental results show that the prediction error of the proposed method is small and the proposed method has strong generalization ability.
【作者单位】：国网上海市电力公司电力科学研究院;国网上海市电力公司市南供电公司;
【分类号】：TM933.4

【相似文献】