Deep learning based energy efficient optimal timetable rescheduling model for intelligent metro transportation systems

Due to the recent advances in intelligent transportation systems (ITS), Automatic Train System (ATS) gained significant attention among the research community. An effective ATS offers the whole railway network to operate in a safe, cost-effective and proficient manner against sudden disturbances like temporary platform blockages. Numerous Train Timetable Rescheduling (TTR) models have been presented for managing unforeseen events which might disturb the timetable. The main aim of an effective TTR model is to reduce power utilization by consuming the entire benefits of reproductive braking energy under a random situation. In this view, this paper presents a new TTR model to optimize the energy of metro systems by the incorporation of improved genetic algorithm (IGA) and long short term memory (LSTM) based recurrent neural network (RNN). The proposed method incorporates three different models, namely controller, timetable, and energy models. The proposed method requires minimum time to recompute a new schedule and offers effective solutions instantly after a random disturbance happens. The performance validation of the proposed IGSA-LSTM model is simulated using Chennai Metro Train Station. The proposed method significantly reduces the energy consumption of metro train and reaches to a minimum average energy utilization of 696 kWh. (C) 2020 Elsevier B.V. All rights reserved.