论文部分内容阅读
分析了畸变波形下感应式电能表和全电子式电能表的计量误差,引入BP神经网络建立电能表计量误差数学模型。用VC++.net开发了谐波电能计量误差分析软件,并应用多线程技术,把神经网络模块和计算模块等放在主线程中实现,而把数据采集模块放在后台工作线程中。采用加汉宁窗和插值改进的傅里叶变换,分析各次谐波电流和电压的幅值、相位。该软件可实现手工输入数据计算,也可配合数据采集系统实时计算出各次谐波对电能计量装置的计量误差和总计量误差。该软件已投入实际运行且情况良好。
The measurement error of inductive watt-hour meter and all-electronic watt-hour meter under distorted waveform is analyzed. The BP neural network is introduced to establish the mathematic model of metering error. Using VC ++ net to develop harmonic energy measurement error analysis software, and the application of multi-threading technology, the neural network module and calculation module on the main thread to achieve, and the data acquisition module on the background work thread. The improved Fourier transform with Hanning window and interpolation is adopted to analyze the amplitude and phase of each harmonic current and voltage. The software can be manually input data calculation, but also with the data acquisition system in real time to calculate the harmonics of the energy metering device metering error and the total measurement error. This software has been put into practical operation and is in good condition.