基于PSO-SVR的燃气轮机系统建模  

Modeling of gas turbine system based on PSO-SVR

作　　者：刘延泉[1] 王如蓓 杨堃[1] LIU Yanquan;WANG Rubei;YANG Kun(Hebei Power Generation Process Simulation and Optimization Control Engineering Technology Research Center,North China Electric Power University,Baoding 071003,China)

机构地区：[1]河北省发电过程仿真与优化控制工程技术研究中心(华北电力大学),河北保定071003

出　　处：《电力科学与工程》2018年第6期60-65,共6页Electric Power Science and Engineering

摘　　要：燃气—蒸汽联合发电机组的燃气轮机部分是一个复杂的多变量系统,针对其非线性、强耦合的特点,提出了一种基于粒子群算法优化的支持向量机回归建模方法对燃气轮机功率和排气温度进行建模。支持向量机可以将非线性问题的数据采用某种非线性关系映射到高维空间,将问题转化为高维空间内的线性回归问题。而粒子群算法的引入改善了网格搜索法运算时间长、计算量大的缺点。对某201 MW燃气联合发电机组的现场数据进行预处理,通过MATLAB对采用不同优化算法的支持向量回归机模型进行仿真实验。结果分析可得,支持向量回归机经过粒子群算法优化后可以提高燃气轮机系统的建模精度。The gas turbine of gas and steam combined generating set is a complex multivariable system.In the light of its nonlinear and strong coupling characteristics,a support vector machine modeling method based on particle swarm optimization is proposed to model the gas turbine power and exhaust gas temperature.Support vector machine uses nonlinear mapping to map data to high-dimensional space,and then the nonlinear problems become linear regression in high-dimensional space.The introduction of particle swarm optimization reduces the computation time of grid search algorithm and amount of computation.After preprocessing the field data of a 201 MW gas combined generator set,the simulation of support vector machine modeling using different optimization algorithms is carried out in MATLAB.The results of the study show that the support vector machine modeling method based on particle swarm optimization can improve the modeling accuracy.

关 键 词：燃气轮机 支持向量回归机 建模 粒子群优化 

分 类 号：TP181[自动化与计算机技术—控制理论与控制工程] TP202[自动化与计算机技术—控制科学与工程]