重型燃气轮机动态系统仿真与控制方法研究  

Research on Dynamic System Simulation and Control Method for Heavy-duty Gas Turbine

作　　者：冯志强 朱雷 李磊 王鹏飞 王沛沛 FENG Zhiqiang;ZHU Lei;LI Lei;WANG Pengfei;WANG Peipei(Guodian Shuangwei Inner Mongolia Shanghai Miao Energy Co.,Ltd.,Ordos 016200,China;Guoneng Zhishen Control Technology Co.,Ltd.,Beijing 102209,China)

机构地区：[1]国电双维内蒙古上海庙能源有限公司,内蒙古鄂尔多斯016200 [2]国能智深控制技术有限公司,北京102209

出　　处：《电工技术》2023年第8期15-19,共5页Electric Engineering

摘　　要：随着我国电力行业的发展以及西气东输发展战略的实施,燃气能源的使用范围逐渐扩大。重型燃气轮机是一种高效的热功转换装置,具有较高的运行效率,同时重型燃气轮机也是燃气轮机发电机组的重要组成部分,得到了极其快速的发展。以GE MS6001FA型重型燃气轮机为研究对象,采用燃气轮机常用的Rowen模型进行建模。由于常规PID控制对简单的线性或非线性系统有较好的控制能力,而对实际中受扰的复杂非线性系统的控制能力较差,因此引入神经网络技术,设计了神经网络PID控制器,通过在线调整相关PID设置参数,使其具有较强的自适应性和一定的智能性,能够获得更高的控制品质。最后进行数值仿真试验,通过与传统PID控制的对比分析,验证了神经网络PID控制器的可靠性与优越性。With the development of China's electric power industry and the implementation of the development strategy of west-to-east natural gas transmission,the scope of use of gas energy is gradually expanding.Heavy-duty gas turbine is an efficient thermal power conversion device with high operating efficiency.At the same time,heavy-duty gas turbine is also an important part of gas turbine generator set,which has been developed rapidly.In this paper,GE MS6001FA heavy-duty gas turbine is taken as the research object,and the Rowen model which is commonly used in gas turbine is used for modeling.Because conventional PID control has better control ability for simple linear or nonlinear systems,and poor control ability for complex nonlinear systems disturbed in practice,a neural network PID controller is designed by introducing neural network technology,which can adjust the relevant PID setting parameters on-line,has strong self-adaptability and certain intelligence,and can obtain higher control quality.Finally,numerical simulation experiments are carried out,and the reliability and superiority of the neural network PID controller are verified by comparing with the traditional PID control.

关 键 词：燃气轮机 PID控制 神经网络PID控制 

分 类 号：TP29[自动化与计算机技术—检测技术与自动化装置]