基于数据驱动的发电机组软测量建模方法优化研究.pdf

文档介绍：30 1 Vol. 30 No. 1
第卷第期机电工程
2013 1 Journal of Mechanical & Electrical Engineering Jan. 2013
A-PDF年月 Split DEMO : Purchase from to remove the watermark
DOI .1001-
:
基于数据驱动的发电机组软测量建模方法优化研究
陈凯亮,陈坚红* ,盛德仁,李蔚
310027
(浙江大学热工与动力系统研究所,浙江杭州)
GRNN B
摘要:针对发电机组中部分难测准参数的软测量及已测量参数的准确性校验等应用问题,提出了融合广义神经网络( )和样
PLSR NN-PLS GRNN
条偏最小二乘回归( )各自优势的数据驱动软测量建模新方法。该方法先采用了对由机理分析初步选定的建模
B PLS
变量预建模,考察了各参变量对因变量的平均贡献率,并筛选得出了主要建模参数,然后采用样条对筛选后的变量建模,从而
NN-PLS
得出了简化、可靠的模型,最后以联合循环电厂的实测数据为样本进行了建模。研究结果表明, 方法对不同工况下测量参数
拟合准确、精度高、模型泛化能力强,同时由于该模型需要保存的参数相对较少,更适合作为解决上述在线问题的模型。
NN-PLS
关键词:发电机组;数据驱动;软测量建模;广义神经网络;偏最小二乘回归;
TK12 TM621 A 1001-4551 2013 01-0093-05
中图分类号: ; 文献标志码: 文章编号: ( )
Optimization of soft sensor modeling methods of generator
sets based on data-driven
CHEN Kai-liang CHEN Jian-hong SHENG De-ren LI Wei
, , ,
Institute of Thermal Science and Power System Zhejiang University Hangzhou 310027 China
( , , , )
In order to solve the problems existing in generating sets system such as soft sensing of some urate measured parameters
Abstract: , ,
accuracy test of sensor measurement data a new modeling method based on generalized regression work GRNN and
, ( )
partial-least squares regression PLSR was proposed. Firstly a GRNN model was built with variables after the mechanism analysis to
( ) ,
assess the average contri