A Genetic Algorithm Optimized RNN-LSTM Model for Remaining Useful Life Prediction of Turbofan Engine

Understanding the remaining useful life (RUL) of equipment is crucial for optimal predictive maintenance (PdM). This addresses issues downtime and unnecessary checks in run-to-failure preventive maintenance. Both feature extraction prediction algorithm have played roles on performance RUL models. A benchmark dataset, namely Turbofan Engine Degradation Simulation Dataset, was selected analysis evaluation. The proposal combination complete ensemble empirical mode decomposition wavelet packet transform could reduce average root-mean-square error (RMSE) by 5.14–27.15% compared with six approaches. When it comes to algorithm, results model be that needs repaired or replaced within a shorter longer period time. Incorporating this characteristic enhance model. In paper, we proposed recurrent neural network (RNN) long short-term memory (LSTM). former takes advantages whereas latter manages better long-term prediction. weights combine RNN LSTM were designed non-dominated sorting genetic II (NSGA-II). It achieved RMSE 17.2. improved 6.07–14.72% baseline models, stand-alone RNN, LSTM. Compared existing works, improvement work 12.95–39.32%.