基于SCADA及时域最小二乘法的抽水蓄能电站发电机多工况参数辨识  

Parameters' Identification of Generator in Pumped Storage Station Based on SCADA Data

作　　者：黄文涛[1] 邓长虹[1] 汪志强[1] 舒征宇[1] 翁毅选[1] 

机构地区：[1]武汉大学电气工程学院,武汉430072

出　　处：《高电压技术》2012年第4期1019-1024,共6页High Voltage Engineering

摘　　要：为了辨识抽水蓄能电站发电机多种运行工况下的模型参数,通过建立发电机的微分方程模型,并经过数学推导建立了发电机实测数据变量与发电机待辨识参数数学关系。基于时域最小二乘法的原理,给出了发电机各工况下适用的参数辨识方法,并通过抽水蓄能电站的SCADA(supervisory control and data acquisition)数据采集系统获得运行数据并对各工况参数进行辨识。仿真验算运用南方电网某抽水蓄能在发电和抽水两种工况下的实际运行数据,通过编制C++程序分析计算得到发电工况和抽水工况下的抽水蓄能整个发电机系统的参数值。将参数值与实际铭牌值相比,误差不超过5%。仿真结果表明,运用高采样速度的实时运行数据,能够激发抽水蓄能机组的暂态和次暂态过程。这种经过采样数据和公式推导相结合的辨识方法比通过传统做实验的辨识方法更简单、方便、有效,且适合抽水蓄能机组多工况运行的特点。To identify the parameters of generators in pumped storage power station in multiple operation states,by establishing differential equations of generators,we put forward the relationships between the variables of measured data and identification parameters,and put forward the identification method of the generator parameters in multiple states based on least squares method,by which the parameters in multiple states could be identified through the SCADA data of pumped storage power station.Using the actual operating data of generator state and pumping state,we simulated to obtain the parameters through C＋＋procedure.It is found that the error between the actual value and the identification value is less than 5%.The simulation results prove that high sampling rate data can stimulate transient and sub-transient process of pumped storage generators.The identification method by using sampling data and mathematical derivations has better effects than that by using experiment,and this method is fit for the feature of multiple states of pumped storage power station.

关 键 词：抽水蓄能电站 多工况 最小二乘法 SCADA数据 发电机 参数辨识 

分 类 号：TV734.21[水利工程—水利水电工程]