摘要
飞机空气循环系统在飞行过程中很容易出现故障,因此在地面上模拟仿真飞机空气循环系统有着重要意义.首先建立飞机空气循环系统仿真模型,针对该系统建立神经网络模型进行训练;引入PID控制算法调节参数以应对飞机遇到的不同飞行状况;针对神经网络模型引入鲸鱼算法,并运用烟花算法、重启机制和正余弦混沌双弦算法对鲸鱼算法进行改良.结果表明:模型可以模拟飞机引气到排出过程,以及不同情况下飞机空气循环系统4个主要组件的出口温度;神经网络系统可以根据输入指标数据预测不同组件的出口温度,并分辨出飞机空气循环系统是否出现故障及哪里出现故障.改进后的神经网络训练速度加快,准确率提升,明显降低了陷入局部最优值的可能性.
The aircraft air circulation system is prone to failure due to various complex environmental factors during flight,so it is of great significance to simulate the system on the ground.Firstly,the simulation model of aircraft air circulation system is established,and the neural network model is established for training.Secondly,the PID control algorithm is introduced to adjust the parameters to cope with the different flight conditions.Finally,the whale algorithm is introduced for the neural network model,and the fireworks algorithm is used to improve the whale algorithm by restarting mechanism and positive chord chaotic double chord algorithm.The results show that the model can simulate the process of aircraft air entrainment to discharge and the outlet temperature of the four main components of the aircraft air circulation system under different conditions.The neural network system can predict the outlet temperature of different components according to the input index data,and identify whether and where the aircraft air circulation system fails.The accuracy of the improved neural network is improved,the training speed is accelerated,and the possibility of the neural network falling into the local optimal value is obviously reduced.
作者
吴会咏
靳舒春
安雪洁
WU Hui-yong;JIN Shu-chun;AN Xue-jie(College of Science,Shenyang University of Chemical Technology,Shenyang 110142,Liaoning,China;School of Business,Zhengzhou University,Zhenzhou 450001,Henan,China)
出处
《西北师范大学学报（自然科学版）》
CAS
北大核心
2023年第2期27-36,共10页
Journal of Northwest Normal University(Natural Science)
基金
国家自然科学基金资助项目(62073157)。
关键词
空气循环系统
PID控制算法
神经网络:故障分析
鲸鱼算法
烟花算法
重启机制
air circulation system
PID control algorithm
neural network
fault analysis
whale algorithm
fireworks algorithm
restart mechanism
