Developing Deep Survival Model for Remaining Useful Life Estimation Based on Convolutional and Long Short-Term Memory Neural Networks

Table 2

The performance comparing our approach and state-of-the-art methods.

Methods	FD001		FD002		FD003		FD004
	RMSE	Score	RMSE	Score	RMSE	Score	RMSE	Score
MLP [41]	37.56		80.03		37.39		77.37
SVR [42]	20.96		42.00		21.05		45.35
RVR [43]	23.80		31.30		22.37		34.34
CNN [31]	18.45		30.29		19.82		29.16
LSTM [22]	16.14		24.49		16.18		28.17
ELM [10]	17.27		37.28		18.47		30.96
DBN [15]	15.21		27.12		14.71		29.88
MODBNE [16]	15.04		25.05		12.51		28.66
RNN [19]	13.44		24.03		13.36		24.02
DCNN [32]	12.61		22.36		12.64		23.31
BiLSTM [26]	13.65		23.18		13.74		24.86
Aug+CNN+LSTM [30]	23.57		20.45		21.17		21.03
DAG [35]	11.96		20.34		12.46		22.43
CNN+LSTM w/regression	14.04		15.15		14.62		21.92
CNN+LSTM w/Weibull	13.98		15.77		15.55		23.05