Technology Category
Planning, Routing, and Scheduling: Mixed Discrete/Continuous PlanningConstraint Satisfaction and Optimization: Mixed Discrete/Continuous OptimizationSearch and Optimization: Mixed Discrete/Continuous Search
Optimizing Freight Rail Electrification: A Framework for Charge Station Selection and Battery Charge/Swap Scheduling
🤖 AI Summary
This paper addresses the joint optimization of charging station location and battery charging/swapping scheduling for electrified freight rail systems. We propose an integrated decision-making framework incorporating multi-battery carriage and dual-mode (charging/swapping) operations. A mixed-integer linear programming (MILP) model is formulated, enhanced by three algorithmic innovations: rectangular piecewise-linear approximation, a heuristic fixing algorithm, and Benders decomposition. These techniques collectively improve computational efficiency for large-scale networks—demonstrated on a 25-station instance. Experimental results show that Benders decomposition achieves the optimal objective value, while the rectangular approximation yields near-optimal solutions with superior engineering practicality. The framework balances system adaptability and economic viability, offering a scalable paradigm for infrastructure planning and operational scheduling to enable zero-emission, battery-powered freight rail operations at scale.
Technology Category
Application Category
📝 Abstract
Battery electric freight trains are crucial for decarbonization by providing zero-emission transportation alternatives. The proper adoption of battery electric freight trains depends on an efficient battery electrification strategy, involving both infrastructure setup and charge scheduling. The study presents a comprehensive model for the optimal design of charging infrastructure and charge scheduling for each train. To provide more refueling flexibility, we allow batteries to be either charged or swapped in a deployed station, and each train can carry multiple batteries. This problem is formulated as a mixed integer linear programming model. To obtain real-time solutions for a large scale network, we develop three algorithms to solve the optimization problem: (1) a Rectangle Piecewise Linear Approximation technique, (2) a Fixed Algorithm heuristic, and (3) Benders Decomposition algorithm. In computational experiments, we use the three proposed algorithms to solve instances with up to 25 stations. Statistical analysis verifies that Benders Decomposition outperforms the other two algorithms with respect to the objective function value, closely followed by the Rectangle Piecewise Linear Approximation technique, and the Fixed Algorithm provides the least optimal solution.
Problem
Research questions and friction points this paper is trying to address.
Optimizing charge station selection for freight rail electrification
Scheduling battery charging and swapping operations efficiently
Developing scalable algorithms for large-scale network solutions
Innovation
Methods, ideas, or system contributions that make the work stand out.
Mixed integer linear programming model
Rectangle Piecewise Linear Approximation technique
Benders Decomposition algorithm
🔎 Similar Papers
No similar papers found.
