Prognostics and Health Management Technology for Aircraft Assembly Production Lines

At present, it is difficult to accurately capture and transmit the status data of key links in aircraft assembly production lines, and the operating status of equipment cannot be quantified in real-time. The cost and difficulty of emergency maintenance are high. The performance level of each state, the probability of equipment failure, and the remaining useful life (RUL) are difficult to calculate precisely. Based on the actual operation of precision machining system equipment on the aircraft assembly production line, a model is built using the performance degradation state, fault conditions, and expected life during its working process. A device performance prediction method based on the CNN-LSTM neural network algorithm is proposed. The CNN-LSTM classifier fully screens long time series information through gating units and uses effective information for time series prediction. After preprocessing the samples, they are input into the CNN-LSTM prediction model to predict the failure time and probability of the equipment. Then, by adjusting the size of the time window, an optimal performance binary classification model is obtained to better adapt to the needs of predictive maintenance. This better achieves Prognostics and Health Management (PHM) of the precision machining system on the aircraft assembly production line in engineering.