摘要
随着工业技术的快速发展,电站等大型工业企业规模越来越大、生产过程越来越复杂,被控对象通常难以精确建模,导致传统基于模型的控制方法难以获得理想的控制效果。数据驱动控制方法可直接根据系统输入输出数据进行控制器设计,有效解决了传统控制方法对对象数学模型的依赖问题。本文提出了基于智能虚拟参考反馈整定的神经元PID控制方法,通过采用人类学习优化算法,依据智能虚拟参考反馈整定目标函数,直接根据系统一组输入输出数据对神经元PID控制器进行全局优化设计,提高了控制性能。对电站过热蒸汽温度控制的仿真结果表明了提出的智能虚拟参考反馈整定神经元PID控制方法的有效性和在工程应用中的优越性。
With the rapid development of industrial technology,the scale of large-scale industrial enterprises such as power stations becomes larger and larger,and the production process becomes more and more complex.Therefore,the controlled object is usually difficult to be accurately modeled,which makes it difficult for traditional model-based control methods to obtain the ideal control effect.Data-driven control methods can directly design controllers based on the input and output data of systems,which frees controller design from the traditional model-based control mode.A new neuron PID control method based on intelligent Virtual Reference Feedback Tuning(VRFT)is proposed,in which human learning optimization(HLO)is introduced to optimize the controller for improving the control performance according to the objective function of intelligent VRFT(IVRFT)with a set of input and output data of systems.The simulation results of the temperature control of overheat steam in power plants demonstrate the effectiveness of this proposed data-driven control method and its superiority in engineering applications.
作者
黄博文
叶琪贤
胡琦
叶程微
王灵
HUANG Bowen;YE Qixian;HU Qi;YE Chengwei;WANG Ling(Shanghai Key Laboratory of Power Station Automation Technology,School of Mechatronics Engineering and Automation,Shanghai University,Shanghai 200444,China;Shanghai Power Construction Testing Institute Co.,Ltd.,Shanghai 200031,China)
出处
《流体测量与控制》
2021年第1期1-9,共9页
Fluid Measurement & Control
