KPLS Based Modeling of Gas Steam Combined CyclePower Plant Under Off design Conditions
-
摘要: 以宝钢电厂燃气蒸汽联合循环发电机组为背景, 在深入分析了联合循环机组变工况性能的基础上, 为解决常规的机理建模方法主要关注机组在设计工况附近的性能而在机组运行点偏离设计工况较远时模型误差较大的问题, 将基于机组运行特性曲线的机理模型和基于KPLS的数据模型相结合, 建立了以大气温度/压力、燃料流量、冷却水温度/流量和汽轮机背压为输入, 机组出力为输出的联合循环混合模型。基于核函数的偏最小二乘方法(KPLS), 是将PLS和核函数理论相结合, 提高了PLS方法的非线性处理能力。用基于KPLS的数据模型修正机理模型预测值与机组实测值的偏差, 使得混合模型既能反映过程机理, 同时又具有较高的精度。以宝钢电厂联合循环机组的历史运行数据为例, 对单纯的机理模型与混合模型进行了比较。实验结果表明, 与单纯的机理模型相比, 混合模型的精度有较大提高。Abstract: Based on the Gas Steam combined cycle power plant (CCPP) of Baosteel power plant, a detailed analysis of the performance of CCPP under off design conditions is carried out. In order to improve the traditional modeling methods which are based on mechanism of focusing on the performance under design conditions and having the difficulty to improve the precision of models when the unit is far away from the design conditions, a hybrid mode which combines the mechanism model based on the performance curves with the data derived model based on KPLS is established. The inputs of the hybrid model are ambient temperature/pressure, fuel flow, cooling water temperature/flow and steam turbine back pressure and the output of the bybrid model is the net equipment output. KPLS which is integration of PLS and kernel function can deal with the nonlinearities. The data derived model based on KPLS with high accuracy is used to correct the error of the mechanism model which has clear physical meaning. Based on the history data of the CCPP of Baosteel power plant, the performance of the hybrid mode is compared with that of the mechanism model. Experiments indicate that the precision of the hybrid mode is higher than that of the mechanism model.
-
Keywords:
- CCPP /
- off design /
- mechanism model /
- hybrid mode /
- KPLS
-
-
焦树建.燃气蒸汽联合循环[M].北京:机械工业出版社, 2003:14. 侯国莲, 殷冲, 张建华.联合循环电厂建模综述[J].现代电力, 2006, 23(6):6872. 陈坚红.联合循环机组运行计划和负荷分配建模启发式遗传算法求解和数据处理[D].杭州:浙江大学, 2004. Kresta J, MacGregor J F, Marlin T E.Multivariate statistical monitoring of process operating performance[J].Canadian Journal of Chemical Engineering, 1991, 69(1):3547. Nomikos P, MacGregor J F.Monitoring batch processes using multi-way principal component analysis[J].Journal of American Institute of Chemical Engineers, 1994, 40(8):13611375. 葛荣彬, 刘尚明.基于人工神经网络的单轴联合循环机组变工况模拟[J].燃气轮机技术, 2008, 21(1):30-35. 刘尚明, 王纯.燃气蒸汽联合循环的神经网络变工况模拟[J].燃气轮机技术, 2006, 19(4):2024. Fantozzi F, Desideri U.Simulation of power plant transients with artificial neural networks application to an existing combined cycle[J].Proceedings of the Institution of Mechanical Engineers, Part A:Journal of Power and Energy, 1998, 212(5):299313. Wold S, Sjtrm M, Eriksson L.PLS-regression:a basic tool of chemometrics[J].Chemometrics and Intelligent Laboratory Systems, 2001(2):109130. Kim K, Lee J M, Lee I B.A novel multivariate regression approach based on kernel partial least squares with orthogonal signal correction[J].Chemometrics and Intelligent Laboratory Systems, 2005(12):2230. Jia R D, Mao Z Z, Chang Y Q, Zhang S N.Kernel partial robust M regression as a flexible robust nonlinear modeling technique[J].Chemometrics and Intelligent Laboratory Systems, 2010(2):9198. Jia M X, Chu F, Wang F L, Wang W.On line batch process monitoring using batch dynamic kernel principal component analysis[J].Chemometrics and Intelligent Laboratory Systems, 2010(2):110122.
计量
- 文章访问数: 877
- HTML全文浏览量: 16
- PDF下载量: 223
下载:
