World Electric Vehicle Journal

With the increasing electrification and intelligence of commercial vehicles, the on-board power demand has risen significantly, rendering conventional independent power supply architecture inadequate. This study proposes a parallel power supply architecture utilizing two identical 28 V/120 A generators, where simplified droop control inherent in standard alternator regulation achieves automatic load sharing without complex active controllers, offering a more cost-effective and redundant alternative to conventional independent power supply systems. To validate the feasibility and stability of this parallel system, a simulation model was established, focusing on load sharing and current stability under dynamic conditions. Furthermore, empirical tests were conducted on a 10-m commercial bus under extreme conditions. The results demonstrate that the parallel architecture maintains a lower air conditioner current, thereby improving system efficiency and operational reliability compared to conventional independent power supply architecture.

Electric vehicle (EV) adoption is accelerating rapidly, increasing pressure on public charging infrastructure, urban energy systems, and network-expansion planning. However, many existing studies examine charging demand forecasting, station network characteristics, or infrastructure siting as separate problems, limiting their ability to translate observed charging behaviour into coordinated planning decisions. To address this gap, this study proposes a multi-domain machine learning framework that integrates spatio-temporal charging demand, station-level infrastructure characteristics, land-use information, and graph-based network relationships for EV charging prediction and infrastructure siting. The analysis uses 8,544,695 public charging transactions recorded at 8553 stations across Beijing during January and July 2025. Charging sessions are aggregated into an hourly station panel and modelled using seasonal-naïve and historical mean baselines, ridge regression, gradient-boosted trees, graph-augmented gradient boosting, and a gated recurrent unit network. Model reliability is assessed through temporal validation, feature ablation, spatial holdout testing, cross-season transfer analysis, explainability, and non-parametric statistical comparison. The results show that graph-augmented gradient boosting achieved the strongest RMSE and variance-explained performance, with an RMSE of 53.80 kWh and $R^2=0.734$, while the gated recurrent unit produced the lowest MAE of 23.82 kWh. Graph neighbour features yielded only a marginal and statistically non-significant forecasting improvement, indicating that the station network is structurally informative but predictively redundant once temporal history is available. For infrastructure expansion, NSGA-II achieved the highest Pareto front hypervolume and identified a knee-point solution of 156 additional chargers, reducing unmet demand by 43.4%. These findings demonstrate that integrated forecasting, graph analysis, and multi-objective siting can provide more defensible and operationally relevant evidence for city-scale EV charging planning.

Amid global economic uncertainty caused by the COVID-19 pandemic and geopolitical conflict, countries worldwide have prioritized urban economic resilience (UER). Using panel data for 283 Chinese cities from 2012 to 2023, this paper applies a difference-in-differences model to estimate the effect of the Pilot Battery Recycling Policy (PBRP) on UER. The results show that: (1) The estimated PBRP coefficient is positive at 0.004 units, equivalent to approximately 4.5% of the sample mean, and the result remains stable across a series of robustness and identification tests. (2) The mechanism and bootstrap mediation analyses provide evidence consistent with three potential city-level transmission channels associated with pilot designation: higher government technology spending, improved energy utilization, and greater total factor productivity. (3) The policy effect is positive and significant in upstream and downstream cities of the Yangtze River Basin, declining resource-based cities, and medium- and small-sized cities; it is negative in regenerative resource-based cities and megacities, while the remaining subgroup estimates are statistically insignificant. The subgroup patterns indicate that industrial foundations, adjustment costs, market capacity, and city scale may help shape policy effectiveness. (4) The Spatial Durbin Model identifies a positive direct effect on pilot cities and a negative indirect effect on neighboring cities under both selected spatial weight matrices. The matrix-specific admissibility and numerical checks confirm the validity of the spatial decomposition. Taken together, the findings point to a positive contribution of pilot designation to local economic resilience. The city-level results point to technological investment, energy-use efficiency, and productivity growth as potential transmission channels, while regional coordination can help manage cross-city competition for recycling resources and capacity.