A Real-Time Train Timetable Rescheduling Method Based on Deep Learning for Metro Systems Energy Optimization under Random Disturbances

Table 5

Performances with three strategies in the two-train metro system.

Strategy	Calculation time (s)	Net traction energy consumption (kWh)	Energy-saving percentage compared with no action strategy (%)
No action	0	293.95	—
MGA	8694.08	274.58	6.59
MGA-GRU	0.15	280.88	4.45