基于TSL-IPSO的燃气-蒸汽联合循环机组负荷对象模型辨识
Identification of Load Object Model of Gas-steam Combined Cycle Unit based on TSL-IPSO
随明鑫 1康英伟1
作者信息
- 1. 上海电力大学自动化工程学院,上海 200090
- 折叠
摘要
针对传统辨识方法与粒子群(PSO)算法在燃气-蒸汽联合循环机组负荷对象模型辨识方面存在寻优精度低和收敛速度慢的问题,提出一种基于维度学习策略的双群体学习改进粒子群优化算法(TSL-IPSO),来优化PSO算法的全局搜索能力和局部改良能力.采用开环阶跃实验得到的燃气-蒸汽联合循环机组257.4和436.07 MW负荷点处的数据对TSL-IPSO算法与PSO等算法辨识得到的负荷对象模型进行对比验证.结果表明:与PSO算法、差分进化算法DE和遗传算法GA相比,TSL-IPSO算法所得辨识模型的均方根误差、平均绝对百分误差均最小,适应度变化曲线收敛效果最好,具有更好的模型辨识精度与寻优性能.
Abstract
Aiming at the problem that the traditional identification methods and particle swarm optimiza-tion(PSO)algorithms had low optimization accuracy and slow convergence speed in identifying the load object model of gas-steam combined cycle units,a dimensional learning strategy-based two-swarm learn-ing improved particle swarm optimization(TSL-IPSO)algorithm was proposed to optimize the global search ability and local improvement ability of PSO algorithm.The load object models identified by TSL-IPSO algorithm and PSO algorithm were compared and validated using the load point data of 257.4 MW and 436.07 MW gas-steam combined cycle units obtained through open-loop step experiments.The re-sults show that compared with PSO,differential evolution(DE)and genetic algorithm(GA),the TSL-IPSO algorithm has the smallest root mean square error and average absolute percentage error of the iden-tification model,the best convergence effect of the fitness change curve,and better model identification accuracy and optimization performance.
关键词
燃气-蒸汽联合循环机组/负荷对象/模型辨识/PSO/TSL-IPSOKey words
gas-steam combined cycle unit/load object/model identification/PSO/TSL-IPSO引用本文复制引用
基金项目
国家自然科学基金(61573239)
上海发电过程智能管控工程技术研究中心资助项目(14DZ2251100)
出版年
2024
NETL
NSTL
万方数据