摘要
针对传统PID控制策略对非线性、时变、多变量耦合、大迟延和惯性的循环流化床床温控制品质不佳,提出运用模糊控制及神经网络相结合的方法对锅炉燃烧系统中的床温被调量进行控制。由模糊控制与神经网络深度融合形成的模糊神经网络控制方式结合了模糊推理与神经网络的并行处理、自学习的优点,网络通过训练可以实现控制规则的自动辨识与隶属度函数的校正,实现“智能化”的调节要求。在MATLAB仿真软件中搭建仿真模块,仿真结果表明该控制策略较传统PID控制平稳调节性能更好,达到稳态的时间更短。
The temperature control of circulating fluidized bed bed has the characteristics of nonlinearity,time-vary,multivariable coupling,large delay and inertia,and the traditional PID control strategy has a poor control effect.A combination of fuzzy control and neural network is proposed to control the bed temperature in the boiler combustion system.The fuzzy neural network control mode formed by the deep fusion of fuzzy control and neural network combines the advantages of parallel processing and self-learning of fuzzy reasoning and neural network.Through training,the network can realize the automatic identification of control rules and the correction of membership function,and realize the"intelligent"adjustment requirements.Simulation modules are built in MATLAB simulation software.The simulation results show that the control strategy has better smooth adjustment performance than the traditional PID control and the time to reach steady state is shorter.
作者
胡兴
韩磊
HU Xing;HAN Lei(Jinkong Power Tongda Thermal Power Shanxi Co.,Ltd.,Datong 037001,China;School of Computer&Information Technology,Shanxi University,Taiyuan 030006,China)
出处
《电工技术》
2023年第6期3-6,共4页
Electric Engineering
关键词
模糊神经网络
循环流化床锅炉
床温控制
智能化调节
fuzzy neural network
circulating fluidized bed boiler
bed temperature control
intelligent adjustment
作者简介
胡兴(1986-),从事热控检修工作,研究方向为循环流化床自动控制技术;韩磊(1993-),博士研究生,研究方向为火电机组节能监测与智能控制。
