摘要
对软测量中辅助变量的选择过程进行研究。主要讨论了基于机理分析方法选择变量,与通过对历史数据进行建模分析方法选择辅助变量。通过神经网络软测量模型预测,对两种方法选择的辅助变量的预测结果进行比较,最终确定辅助变量。
The soft-sensing secondary variables selection process were mainly studied in this paper. Mainly discusses the way of the variables selection based on mechanism analysis and modeling and analyzing of historical data. Though the soft measurement of the neural network model predicting, the two methods of secondary variables selection predicted results are compared, and ultimately determine the auxiliary variables.
出处
《电力科学与工程》
2011年第7期37-40,共4页
Electric Power Science and Engineering
关键词
辅助变量
机理分析
主元分析
软测量
神经网络
auxiliary variable
mechanism analysis
principal component analysis
soft measurement
neural network
