摘要
发动机是飞机重要的组成部分,是飞机正常安全飞行的保障。由于其工作环境恶劣,难以进行准确的故障诊断。为此,提出了基于ABC-LSTM的飞机发动机故障诊断方法。该方法首先设计了长短时记忆神经网络(Long Short-Term Memory,LSTM)故障诊断模型;然后采用人工蜂群算法(Artificial Bee Colony Algorithm,ABC)对所设计的长短时记忆神经网络故障诊断模型中的隐含层节点数进行优化,确定最优隐含层节点个数,创建ABC-LSTM的飞机发动机故障诊断模型;最后采用发动机监测数据对所创建的ABC-LSTM故障诊断模型进行测试和验证。为表明所设计的故障诊断模型的优越性,设计了PSO-LSTM和GA-LSTM故障诊断模型。研究结果表明,所提出的ABC-LSTM方法的故障诊断率达93.14%,与LSTM、PSO-LSTM、GA-LSTM故障诊断方法相比具有更好的诊断效果,为飞机发动机故障诊断提供了一种新思路。
Engine is an important part of the aircraft which should be guaranteed for the safe and normal flight of the aircraft.Due to its harsh working environment,it is difficult to perform accurate fault diagnosis.For this reason,an aircraft engine fault diagnosis method based on ABC-LSTM was proposed.First,this method designed the Long Short-Term Memory(LSTM)fault diagnosis model.Then the Artificial Bee Colony Algorithm(ABC)to design the fault diagnosis model of the long and short-term memory neural network was used.To optimize the number of hidden layer nodes,the optimal number of hidden layer nodes and created an ABC-LSTM aircraft engine fault diagnosis model was determined.Finally,engine monitoring data to test and verify the created ABC-LSTM fault diagnosis model was used.In order to show the superiority of the designed fault diagnosis model,PSO-LSTM and GA-LSTM fault diagnosis models were designed.The research results showed that the proposed ABC-LSTM method can diagnose aircraft engine faults as high as 93.14%.Compared with LSTM,PSO-LSTM,and GA-LSTM for fault diagnosis,it shows a better diagnostic effect and provides a new idea.
作者
崔建国
宋博文
崔霄
王景霖
杜文友
于明月
刘冬
蒋丽英
CUI Jian-guo;SONG Bo-wen;CUI Xiao;WANG Jing-lin;DU Wen-you;YU Ming-yue;LIU Dong;JIANG Li-ying(College of Automation,Shenyang Aerospace University,Shenyang 110136,China;Model Balance and Wind Tunnel Equipment Department 5,AVIC Aerodynamics Research Institute,Shenyang,110034,China;Aviation Key Laboratory of Science and Technology on Fault Diagnosis and Health Management,AVIC,Shanghai 201601,China)
出处
《沈阳航空航天大学学报》
2022年第3期50-55,共6页
Journal of Shenyang Aerospace University
基金
国家自然科学基金(项目编号:51605309)
中国航空科学基金(项目编号:20163354004,201933054002)
辽宁省教育厅基金项目(项目编号:JYT2020021)。
