摘要
整机总体性能仿真是航空发动机及燃气轮机仿真的重要组成部分,在航空发动机及燃气轮机的设计制造和使用全寿命周期内发挥着重要作用。综合70多年来航空发动机及燃气轮机总体性能仿真的发展成果,梳理了各时期总体性能仿真的发展历程。从基本方法、模型精细化、求解算法和修正方法等角度,分析了国内外以部件级模型为代表的基于物理机理的总体性能仿真方法研究现状;探讨了以人工神经网络、支持向量机和深度学习为代表的人工智能算法在总体性能仿真中的应用现状;介绍了机载模型、机理-数据混合模型和多维度模型基本方法和主要成果。基于目前的研究成果和技术发展趋势,认为航空发动机及燃气轮机总体性能仿真应向物理机理模型更精细化、人工智能技术更深入和应用模型构建更为规范化的方向发展。
Engine-level performance simulation is an integral aspect of aeroengine and gas turbine simulation,and plays a crucial role throughout the entire life cycle of design,manufacturing,and operation.This paper presents a comprehensive analysis of the development process of aeroengine and gas turbine performance simulation in each historical stage,building upon the accomplishments made over the past 70 years.The research status of physical mechanism performance simulation,primarily represented by the component-level model,was examined from various perspectives including basic methods,model refinement,solution algorithms,and correction methods.Furthermore,the application of artificial intelligence algorithms,such as the artificial neural network,support vector machines,and deep learning,in engine-level performance simulation,was discussed.The paper also provided an overview of the fundamental methods and key achievements of on-board models,mechanism-data hybrid models,and multi-dimensional models.Finally,based on current research findings and technological development trends,it is believed that the engine-level performance simulation of aircraft engines and gas turbines should develop towards a more refined physical mechanism model,deeper artificial intelligence technology,and more standardized application model construction.
作者
董威
尹家录
郑培英
程显达
DONG Wei;YIN Jia-lu;ZHENG Pei-ying;CHENG Xian-da(School of Mechanical Engineering,Shanghai Jiao Tong University,Shanghai 200240,China;AECC Shenyang Engine Research Institute,Shenyang 110015,China)
出处
《航空发动机》
北大核心
2023年第5期8-21,共14页
Aeroengine
基金
航空动力基础研究项目资助。
关键词
航空发动机
燃气轮机
总体性能
仿真
物理机理模型
人工智能
应用模型
aeroengine
gas turbine
engine-level performance
simulation
physical mechanism model
artificial intelligence
application model
