Stochastic fast charging scheduling of battery electric buses with energy storage systems design

Under the background of urban green and low-carbon economic development, battery electric buses (BEBs) together with fast charging technologies are considered as an effective way in promoting carbon emissions reduction and improving energy efficiency. However, challenges arise during daily peak periods, at which BEB charging activities cause increased operation costs and substantial stress on the power grid. To fill the gaps, this work introduces energy storage systems (ESSs) into the BEB fast-charging scheduling problem. A stochastic programming model considering uncertain discharge efficiencies of ESSs is established, aiming to minimize total operation costs of fast charging stations. It is then transformed into a deterministic one with a manageable scenario sample size, using an integrated sample average approximation (SAA) and K-means approach, as well as an integrated SAA and K-means++ approach. We also present a heuristic algorithm and an improved genetic algorithm to solve the transformed model. Numerical studies based on a real-life urban transit network in Shanghai, China, are conducted to demonstrate the effectiveness of the proposed methods. Sensitivity analysis further reveals the impact of charging power on total operation cost. •We add ESSs to BEB peak charging to cut costs and grid stress.•BEB and ESS battery degradation costs from charge rates affect scheduling.•Model combines K-means (K-SAA) and K-means++ (K-SAA++) for scenarios.•We develop K-SAA based greedy and genetic algorithms to solve the problem.•Shanghai cases show our methods cut costs and grid stress in BEB charging.