摘要
为了提高涡轴发动机性能参数换算的精度,研究了相似换算公式的指数修正方法,通过修正相似换算公式的指数提高大气温度和压力影响的性能参数换算精度,在此基础上乘以湿度修正系数以修正大气湿度对性能参数的影响。由于修正系数法无法综合考虑各种环境条件之间的耦合影响,提出一种基于神经网络的涡轴发动机性能参数换算方法。仿真结果表明:基于神经网络的性能参数换算方法精度更高,更容易实现。
To improve the conversion accuracy of turboshaft engine performance parameters,the exponents correction method of similarity-conversion formula was studied.By modifying the exponents of the similarity-conversion formula,the conversion accuracy of the performance parameters affected by atmospheric temperature and pressure was improved,and on this basis,the influence of humidity on performance parameters was corrected by the correction coefficient method.Since the correction coefficient method fails in considering comprehensively the coupling effects of various environmental conditions,a performance parameter conversion method based on neural network was proposed.The simulation results show that the performance parameter conversion method based on neural network is more accurate and easier to implement.
作者
黄浩
黄向华
HUANG Hao;HUANG Xianghua(College of Energy and Power,Nanjing University of Aeronautics and Astronautics,Nanjing 210016,China)
出处
《机械制造与自动化》
2022年第3期187-190,共4页
Machine Building & Automation
基金
国家自然科学基金项目(51576097)。
关键词
涡轴发动机
参数换算
相似换算
湿度修正
神经网络
turboshaft engine
parameter correction
similarity theory
humidity correction
neural networks