摘要
发动机试验是航空发动机开发与设计中的重要环节,空气管道系统作为发动机实验测试系统,发挥着重要作用。该文针对传统的人工调节先输出后反馈再控制的策略存在控制滞后、依赖调节经验的问题开展控制策略研究。构建压力试验系统的参数预测神经网络模型;以试验系统的目标压力和目标流量为依据,约束阀门调节方向,遍历所有的阀门开度生成策略表;将策略输入到预测模型,利用预测结果结合策略选取规则,得到最终的阀门调节策略。经过仿真与试验测试,该文提出的阀门调节策略将试验控制效率至少提高了30%,并且流量、压力控制精度优于1%。
Engine test is an important part of aeroengine development and design.As engine test system,air piping system plays an important role.This paper focuses on the problems of control lag and dependence on adjustment experience existing in the traditional manual adjustment strategy of output first,feedback second and control.The parameter prediction neural network model of the pressure test system is constructed;Based on the target pressure and target flow of the test system,the adjustment direction of the valve is constrained,and the strategy table is generated by traversing all valve openings;The strategy is input into the prediction model,and the final valve adjustment strategy is obtained from the prediction results under the strategy selection rules.According to simulation and actual test,the valve adjustment strategy proposed in this paper improves the test control efficiency by at least 30%,and the flow and pressure control accuracy is better than 1%.
作者
吴洋
张燕华
周家林
程月华
单圣强
WU Yang;ZHANG Yanhua;ZHOU Jialin;CHENG Yuehua;SHAN Shengqiang(College of Automation Engineering,Nanjing University of Aeronautics and Astronautics,Nanjing 210000,China;AECC Sichuan Gas Turbine Establishment,Mianyang 621000,China)
出处
《电子设计工程》
2024年第9期70-74,共5页
Electronic Design Engineering
基金
南京航空航天大学前瞻性布局科研专项(1003-ILA22041-1A)。
作者简介
吴洋(1999-),男,四川南充人,硕士研究生。研究方向:故障诊断与容错控制。
