Remaining Useful Life Interval Estimation-Based Maintenance Decision-Making for Key Equipment

To ensure the safe and reliable operation of key equipment, a maintenance strategy based on remaining useful life (RUL) interval estimation is developed. This methodology uses a long short-term memory (LSTM) network to predict the RUL of the equipment. By analyzing the statistical relationship between input and output variables of the LSTM network, the RUL interval with a certain confidence level is derived using probability estimation theory. Taking maintenance cost as an index, the cost models of preventive and corrective maintenance are constructed, respectively. By the comparison of the long-run maintenance cost rates of the two models, the optimal maintenance decision is made. Finally, the aircraft turbofan engine simulation dataset is introduced to verify the effectiveness of the proposed maintenance strategy.