Description

Machinery remaining useful life (RUL) prediction is an important task in condition-based maintenance (CBM). CBM is a maintenance strategy that collects and assesses real-time information, and recommends maintenance decisions based on the current condition of the system. In this process, we propose a fault knowledge transfer assisted LSTM (Long Short Term Memory Neural Network) and GRU (Gated Recurrent Neural Network) method for RUL prediction. Divergence minimization and domain adversarial adaptation techniques are combined to transfer fault knowledge from a fault dataset to the run-to-failure samples. With the diagnosed fault condition information, the RUL prediction network can learn various degradation patterns under different faults using the structure of LSTM. Then a GRU process based on predicted soft fault conditions are proposed to get RUL prediction results. Finally the algorithms are implemented and generates the result in form of metrics like, accuracy, precision, recall and f1-score.