Inter-Stop Energy Prediction and Causal Driver Quantification for Dual-Source Trolleybuses via a Time-Aware Tabular Deep Learning Architecture

2026-07-13

AI Summary

This study addresses the challenge of accurately modeling and interpreting energy consumption in dual-source trolleybuses, which is influenced by heterogeneous factors such as route characteristics, driving trajectories, and weather conditions. To this end, the authors propose a time-aware tabular deep learning architecture that integrates periodic time encoding with a parameter-efficient batch ensemble backbone, coupled with Bayesian optimization for automated hyperparameter tuning. Furthermore, they introduce a three-tier causal interpretation pipeline—combining feature attribution, linear non-Gaussian acyclic models, and a meta-learner—to establish the first comprehensive explanatory framework for trolleybus energy use, encompassing marginal effects, causal directions, and net average treatment effects. Evaluated on a Zurich dataset, the model achieves a state-of-the-art performance with 6.52% MAPE and an R² of 0.982, outperforming ten baselines. Key insights reveal regenerative braking ratio and average speed as critical energy-saving factors, while excessive coasting distance is identified as a primary cause of energy waste, offering actionable thresholds for vehicle control and operational optimization.

Abstract

Dual-source trolleybuses alternate between overhead catenary supply and on-board battery operation, creating energy-use patterns driven by route attributes, high-frequency trajectories, and hourly weather. Existing models struggle to represent these heterogeneous inputs and rarely explain the causal drivers of consumption. This paper proposes a time-aware tabular deep learning framework for inter-stop energy management. Periodic time encoding is integrated into a parameter-efficient batch-ensemble backbone to jointly learn static and sequential features, while Bayesian optimization with tree-structured density estimation tunes hyperparameters. To move beyond prediction, a three-layer causal explanation pipeline combines feature attribution for marginal effects, a linear non-Gaussian acyclic model for causal direction discovery, and a meta-learner for net average treatment effects. Experiments on the Zurich trolleybus dataset enriched with meteorological records achieve a MAPE of 6.52% and R of 0.982, outperforming ten statistical, tree-ensemble, and deep learning baselines. Ablation results show that periodic time encoding contributes most to the accuracy gain. Causal analysis identifies regenerative braking ratio and average speed as the strongest energy-saving factors, while coasting distance is the main driver of excess consumption. The findings offer actionable thresholds for vehicle technology, driving behavior, capacity allocation, and catenary network planning.

Problem

• inter-stop energy prediction
• causal driver quantification
• dual-source trolleybuses
• heterogeneous inputs
• energy-use patterns

Innovation

• time-aware tabular deep learning
• causal driver quantification
• periodic time encoding
• batch-ensemble
• inter-stop energy prediction