基于机理数据协同驱动的重型燃气轮机控制系统参数解析余度构建研究  被引量:4

Research on Parameters Analysis Redundancy Construction of Heavy-Duty Gas Turbine Control System Based on Mechanism Data Cooperative Driving Method

在线阅读下载全文

作  者:孙嘉娴 谢振伟 谭湘敏 胡春艳[1,2,4] 李伟 Sun Jiaxian;Xie Zhenwei;Tan Xiangmin;Hu Chunyan;Li Wei(Key Laboratory of Light-duty Gas-turbine, Institute of Engineering, Thermo-physics, Chinese Academy of Sciences, Beijing 100190, China;University of Chinese Academy of Sciences, Beijing 100190, China;Shandong Engineering Research Center of Aeronautical Materials and Devices, Binzhou University, Shandong Binzhou 256600, China;Innovation Academy for Light-duty Gas Turbines, CAS, Beijing 100190, China)

机构地区:[1]中国科学院工程热物理研究所轻型动力重点实验室,北京100190 [2]中国科学院大学,北京100190 [3]滨州学院山东省航空材料与器件工程技术研究中心,山东滨州256600 [4]中国科学院轻型动力创新研究院,北京100190

出  处:《燃气轮机技术》2022年第2期12-18,共7页Gas Turbine Technology

基  金:国家科技重大专项(2017-V-0010-0061);基础科研项目(JCKY2020130C025)。

摘  要:针对重型燃气轮机控制系统传感器解析余度的构建问题,提出了一种机理数据协同驱动的控制系统参数解析余度构建方法。首先基于重型燃气轮机工作原理,建立参数的机理模型;然后引入网络架构搜索法得到最优前馈神经网络结构,通过前馈神经网络训练机理模型的误差,得到数据驱动模型,将其作为机理模型的补偿部分,补偿部分提高了参数解析的精度。仿真结果表明本文提出的参数解析方法的有效性,用该方法可获得精准的解析模型。Aiming at the construction of sensor analytical redundancy of heavy-duty gas turbine control system,a mechanism data cooperative driving method has been proposed,which is an analytical redundancy construction method for parameters of control system.Firstly,the mechanism model of parameters was established based on the working principle of heavy-duty gas turbine.Secondly,the optimal feed-forward neural network structure was obtained using network architecture search method.The data-driven model was established by training the error of mechanism model using feed-forward neural network.This data-driven model was used as the compensation part of the mechanism model,having improved the accuracy of parameter analysis.The simulation results show that the parameter analysis method is effective.The accurate analytical model could be obtained by this method.

关 键 词:解析余度 重型燃气轮机 机理模型 数据驱动模型 神经网络 

分 类 号:TK472[动力工程及工程热物理—动力机械及工程]

 

参考文献:

正在载入数据...

 

二级参考文献:

正在载入数据...

 

耦合文献:

正在载入数据...

 

引证文献:

正在载入数据...

 

二级引证文献:

正在载入数据...

 

同被引文献:

正在载入数据...

 

相关期刊文献:

正在载入数据...

相关的主题
相关的作者对象
相关的机构对象