基于偏最小二乘回归分析的供电煤耗研究  

Power Supply Coal Consumption Research Based on Partial Least Squares Regression

作　　者：刘建华[1] 纪律[1] 贺林博[1] 

机构地区：[1]华北电力大学能源动力与机械工程学院,河北保定071003

出　　处：《电力科学与工程》2011年第4期30-33,共4页Electric Power Science and Engineering

摘　　要：对偏最小二乘回归分析在供电煤耗预测中的应用进行了研究。该方法可有效地进行数据准备和样本预处理,并对影响因素提取主成分,而提取出的成分具有线性无关的特点,对供电煤耗有较好的解释能力。根据某电厂供电煤耗与锅炉效率、汽机热耗、厂用电率的对应数据资料,将供电煤耗与三者之间的关系加以分析,建立供电煤耗的预测模型。预测模型的检验方式采用交叉有效性检验,选定对模型有显著改善的PLS主成分个数。结果表明,该方法能准确地估计出变量的回归系数,建模速度快、精度高。The effectiveness of applying partial least squares regression to power supply coal consumption forecasring is investigated. This algorithm not only can effectively perform the data preparing and samples pretreatment, but also can extract principal components of influencing factors. The extracted components have the property of linearly independence, can be used to explain the power supply coal consumption better. The article is according to corresponding data of power supply coal consumption with boiler efficiency , turbine heat consumption and power utilization rate, mark the relationship between power supply coal consmnption with the three to a model, and estab- lish prediction model of power supply coal consumption. The effectiveness of cross-examination methods of the model test is used to select the number of PLS principal components. The results show that this method can estimate variables regression coefficients accurately, and with this method the high modeling speed and high forecasting accuracy can be obtained.

关 键 词：偏最小二乘回归分析 供电煤耗 主成分 预测模型 

分 类 号：TM621[电气工程—电力系统及自动化]