摘要
详细讨论被预测参数相关变量的选择原则,偏最小二乘法中选择主成分个数的交叉有效性原则,并采用偏最小二乘法建立火电厂实时数据预测模型的回归方程。以某火电厂300 MW机组的PI实时数据库中的数据作为样本数据,建立实时数据预测模型对传感器测点测量数据进行故障检测及诊断,解决了普通多元回归模型对多重具有相关性的数据适应性差的问题,具有速度快、预测精度高的优点。
The principal on how to select related variables,cross validation on confirm the numbers of the components were discussed in detail.The forecasting regression function was deduced to forecast the realtime data based on partial least square method.The research work was based on the 24 hours data from PI server database of one 300 MW units.Partial least square method（PLS） was adopted to develop the simulating model to detect and diagnosis the sensor faults.This approach performs well when multicollinearity of variables exists and has the virtues of fast calculating speed and high precision.
出处
《能源工程》
2010年第4期62-64,68,共4页
Energy Engineering
关键词
偏最小二乘法
传感器
预测
火电厂
partial least square method（PLS）
sensor
forecasting
thermal power plant
