Application of dynamic programming to optimal energy management of grid-independent hybrid railcars

Abstract

In this paper, a dynamic programming-based optimization analysis is performed to quantitatively assess the energy-saving potential associated to the battery hybridization of railcars, which is nowadays regarded as a high-potential solution to reduce the greenhouse-gas impact of regional trains. A computationally effective longitudinal model is first developed, starting from an available energetic macroscopic representation modeling tool. The resulting simplified railcar model was first integrated with a battery state of charge simulator and then embedded within a fast constrained optimization algorithm. The optimization outcomes, reachable within reasonable timeframe and considering cost-effective hybridization scenarios, are highly informative for train manufacturers and operators interested in substantial hybridization of railcars. Particularly, fuel savings as high as 18% can be achieved with respect to the current diesel-powered trains. As a consequence, useful design and control guidelines are made available, considering both retrofitting or replacing the existing railway systems (particularly focusing on regional trains), while coping with capital and operating costs, safety, and traffic constraints.

Li-Ion Battery-Based Hybrid Diesel-Electric Railway Vehicle: In-Depth Life Cycle Cost Analysis

“…The resulting operation is characterized by frequent switches in the power split factor, what further complicates its on-line implementation. Therefore, DP is commonly used just as baseline for benchmarking other EMSs [22]. The optimization problem is based on the following cost function:…”

“…In Global Optimization (GOP) strategies, the optimization is developed off-line based on a pre-defined or predicted drive cycle, while in Real Time Optimization (RTOP) strategies the optimization is developed on-line. Proposed GOP strategies for hybrid diesel railway vehicles include approaches based on dynamic programming [18], [22], Pontryagin's minimum principle [17], non-linear programming [23] and genetic algorithm [24]. RTOP strategies such as equivalent consumption minimization or model predictive control have been proposed for other hybrid railway topologies such as FC vehicles [25], [26].…”

Analysis of Hybrid and Plug-In Hybrid Alternative Propulsion Systems for Regional Diesel-Electric Multiple Unit Trains

“…Their aim is to minimize energy consumption by managing the power flows from different energy sources in the system. Dynamic programming (DP), as a global optimization method, is widely used in EMS optimization for hybrid railway vehicles [37][38][39]. It was also used in deriving a fuel-optimal combined driving and energy management strategy [40].…”

The development of a representative multidimensional transient duty cycle for in-service switcher locomotives

“…Given that such vehicles operate on fixed schedules, their duty cycle can be derived from vehicle and route data. [5][6][7] However, switcher locomotives do not operate on fixed schedules, and therefore require the synthesis of a driving cycle for power source sizing, [8][9][10][11][12] and energy management. 8,10 To that end, dieselelectric locomotives would typically be instrumented to record engine-generator power which would be used for power source selection and sizing.…”