基于数据挖掘的燃气轮机机组能耗特性建模分析  

Modeling and Analyzing of Energy Consumption Characteristics of Gas Turbine Unit Based on Data Mining

作　　者：郑迎九[1] 刘兴彦[1] ZHENG Yingjiu;LIU Xingyan(Hangzhou Huadian Banshan Power Generation Co.,Ltd.,Hangzhou 310000,China)

机构地区：[1]杭州华电半山发电有限公司,浙江杭州310000

出　　处：《微型电脑应用》2024年第12期134-138,共5页Microcomputer Applications

基　　金：中国华电集团有限公司项目(JG0120190686)。

摘　　要：针对半山公司燃气-蒸汽联合循环机组能耗特性建模问题,提出一种最小二乘支持向量机(LSSVM)模型分别对燃气轮机、余热锅炉和蒸汽轮机能耗特性进行建模和分析。考虑到确定LSSVM核参数和惩罚因子难题,利用果蝇优化算法(FOA)对其进行全局寻优,提升模型精度。基于机组实际能耗数据开展试验,结果表明,所提方法的能耗预测值与实际运行值的平均误差小于2%,能够满足实际应用需求,具有较高的应用前景。Aimed at the modeling problem of the energy consumption characteristics of the gas-steam combined cycle unit of Banshan Company.A least squares support vector machine(LSSVM)model is proposed for modeling and analyzing the energy consumption characteristics of gas turbine,waste heat boiler and steam turbine,respectively.To solve the problem of determining LSSVM kernel parameters and penalty factors,the fruit flying optimization algorithm(FOA)is used for global optimization to improve the accuracy of the model.Tests are carried out based on the actual energy consumption data of the unit.The results show that the average error between the predicted energy consumption of the proposed method and the actual operation value is less than 2%,which can meet the actual application requirements and has a higher application prospect.

关 键 词：能耗分析 燃气-蒸汽联合循环机组 最小二乘支持向量机 果蝇优化算法 

分 类 号：TM611[电气工程—电力系统及自动化]