基于QAR数据的民用飞机故障预测及验证

Civil Aircraft Fault Prediction and Verification Based on QAR data

飞机飞行过程中产生成百上千种飞行参数和数量庞大的飞行数据,但目前这些数据并没有得到充分有效的利用,飞机的维修还处在以定期维修和事后维修为主的阶段;随着航空技术的不断发展,利用飞行数据进行故障预测,转变民机维修模式向视情维修发展变得越来越有必要;首先对基于QAR数据的民用飞机故障预测技术路线进行了说明;其次介绍了适用于民机QAR数据的两种故障预测方法,包括基于曲线拟合的性能预测方法和基于时间序列的趋势预测方法;再次,详细描述了基于QAR数据的故障预测系统的实现途径;通过预测关键参数变化趋势,达到提前发现故障,以制定合理的维护计划,确保飞行安全的目的;最后采用提出的方法对波音飞机的空调、滑油系统关键参数数据进行预测,预测结果验证了方法的有效性。

There are hundreds of flight parameters and a large number of flight data during aircraft flight,but these data have not been fully and effectively utilized at present,and the maintenance of aircraft is still in the phase of regular maintenance and post-repair. With the continuous development of aviation technology,it is more and more necessary to make use of flight data to predict faults,and to change the maintenance mode of civil aircraft to the development of maintenance according to the situation.First,the fault prediction technology of civil aircraft based on QAR data is described.Secondly,two fault prediction methods for civil aircraft’ s quick access recorder(QAR)data are introduced,including the performance prediction method based on curve fitting and the performance trend prediction based on time series.Again,the realization of fault prediction system based on flight data is described in detail. By predicting the change trend of key parameters,the fault can be detected in advance to make a reasonable maintenance plan and ensure flight safety.Finally,the proposed method is used to predict the key parameters of the air conditioning and lubricating system of Boeing aircraft,and the prediction results verify the effectiveness of the method.