Integrating Feature Engineering with Deep Learning to Conduct Diagnostic and Predictive Analytics for Turbofan Engines

Table 5

Forecasting errors using all sensors.

Training	RF	XGB	SVM	DNN	RNN	LSTM	GRU	CNN
RMSE	16.47	19.2	19.73	20.07	20.27	19.32	19.36	19.04
MAE	12.13	15	13.24	15.32	15.64	13.72	13.88	14.2
MAAPE	19.57	15.19	21.22	22.32	22.42	21.27	21.96	21.45
Time (s)	179	7197	710	1082	3329	10078	9646	2650
Testing	RF	XGB	SVM	DNN	RNN	LSTM	GRU	CNN
RMSE	18.05	17.47	20.27	19.32	19.12	19.07	18.98	19.35
MAE	12.92	12.45	13.91	13.36	13.19	13.2	13.13	13.24
MAAPE	22.44	20.97	22.21	20.55	20.52	20.84	20.98	20.43
Time (s)	1.42	4.41	4.34	0.45	0.63	0.78	0.69	0.49