Thermal Power Indicators Model Analysis

In this paper, the monthly reports of classify units from Shuangyashan Elec-tronic Power Station Ltd have been sorted out and analyzed to find the regularity.The paper discusses the relation of all indicators which a?ect the power generation ofthis thermal power plant so as to build up regression model based on the connectionof generation and all those points above.With the least square method to estimate the model parameters, we can get alinear regression model which is made of all variables. The model's significance hasbeen tested, and the results show that the model is significant. The significance ofall the parameters in the model have also been tested, and the consequence showsthat three parameters have little e?ect on dependent variable. So they are deleted.Stepwise regression analysis is used to analyze all independent variables model forfinding the best regression model.The best regression model has been analyzed by principal component analysisfor evaluating indicators. According to the cumulative contribution of eigenvalue, wecan select the right amount of principal components so as to simplify the indicators.