Partial Least-Square Regression And Its Application On Parameter Prediction In The Units

With the popularization of DCS system in electrical industry, there is a largenumber of data collected in power plant. The paper discuses the applied actuality of datamining in the electric power industry, and find that there is a most widely space amongthe development of data mining technology in the field of the electric power system. TheStatistics method is one of the important measures in the data mining fields.The paper studies the correlativity of the variables using the regression pattern ofthe data mining, and has conducted the thorough research to the partial least squaremethod, and applied it to DCS data analysis in the thermal power plant, and the valuableresult has been gained. The method, which has been generated and developed to fit thepractical need, The method based on the abstracting components is good at handling ofmultiple variable correlation problems. Therefore, it is more credible and available formodeling and forecasting. In the end of the paper, the results of calculation examplefrom real parameters show that this method which can be used in the aspect of parameterprediction and the examination of lacking value, which can provide reference of theoptimized control of some parameters during the operation, and will be benefit to theoptimized operation in the units.