论文部分内容阅读
用人工种经网络技术，对波音747—200型飞机的JT9D发动机的故障诊断进行了研究，并构成了诊断装置。该研究使用北京飞机维修工程有限公司提供的发动机性能监控数据，在脱机后根据性能排队情况用经验推定法，对发动机的一些常见故障和突发性故障进行了诊断。在诊断过程中，首先搜集发动机的故障状态数据，并对这些数据进行归纳选择，制成了诊断用的教师信号“故障模型”，通过神经网络系统对教师信号的学习，在一定范围内，对没有经过学习的实故障数据进行了诊断。其结果表明了诊断装置的有效性。
Using artificial network technology, the fault diagnosis of JT9D engine of Boeing 747-200 aircraft was researched, and a diagnostic device was constructed. The study used the engine performance monitoring data provided by Beijing Aircraft Maintenance Engineering Co., Ltd. to diagnose some common and unexpected engine faults based on the performance queuing after going offline. In the process of diagnosis, we firstly collect the fault status data of the engine and summarize and select these data to make a “fault model” of the teacher’s signal for diagnosis. Through the learning of the teacher’s signal through the neural network system, in a certain range, The actual fault data that has not been studied is diagnosed. The result shows the effectiveness of the diagnostic device.