With the continuous development of electric vehicle intelligence, traditional battery technology faces shortcomings in sensing data and monitoring methods at the cell, module, and system levels, limiting the further enhancement of lithium-ion battery intelligence and safety. Among the “three major components” of electric vehicles-motor, electronic control unit (ECU), and battery - the first two have been or can be easily intelligentized, while the progress of battery intelligence lags significantly, becoming a key shortcoming in electric vehicle intelligence. Multi-source parameter sensors are crucial “nerves” for lithium-ion batteries to finely perceive their operational and safety status. Here, we report a prototype of a smart battery at the cell and module levels, equipped with a mechanical-thermal-electrical multi-source parameter sensor composed of fiber Bragg gratings (FBG) and advanced electrical sensors, capable of real-time monitoring of battery expansion force-displacement, temperature, voltage, current, and internal resistance. More importantly, the smart battery can achieve real-time perception of material phase transitions, state of charge (SOC), and state of health (SOH). Through the joint analysis of dV/dQ, dF/dQ, and dR/dQ, the intrinsic relationship between active-material phase-transition characteristics and the battery's mechanical-electrical response is revealed. By combining multi-source sensing with a lightweight response surface model, an SOC estimation error below 1.02% and an internal-resistance prediction error below 0.5% are achieved, and real-time SOH estimation is realized through the deviation between the predicted and measured internal resistance. Additionally, the multi-parameter and finely detailed smart battery can be easily extended into modules, enabling intelligent monitoring of each cell within the group, which is challenging to achieve in traditional battery systems. At the module level, the smart battery framework further reveals thermal non-uniformity and hotspot migration characteristics among cells, demonstrating its scalability and practical value in refined battery management. The development of smart batteries can significantly improve the quality, reliability, and lifespan of individual cells, avoid the short-board effect on battery system performance, and achieve breakthrough advancements in lithium-ion battery technology.

Cell inconsistency within a battery system can be illustrated though the “bucket effect”, where overall safety is dictated by the most vulnerable cell. However, existing unsupervised inconsistency evaluations often lack generalizability due to their single-model limitations. Therefore, this paper introduces a novel unsupervised fusion framework whose innovation stems from the synergistic fusion of a sparse autoencoder, local outlier factor, and Wasserstein distance. This combination harnesses the complementary strengths of each method, providing a comprehensive evaluation of cell inconsistency from the perspectives of pattern reconstruction, local density and distribution divergence. Two fusion strategies (hierarchical filtering and ensemble learning) were implemented and compared using real vehicle and laboratory data. In a year-long warning analysis, the framework provided 27 and 24 cumulative alarms for the thermal runaway of battery packs in #1 and #2 vehicles, respectively. This demonstrates the framework’s potent capability for early fault identification. Furthermore, the results demonstrate that the ensemble learning strategy effectively identifies the fault cells of 6 and 37 under both lithium plating and under-voltage conditions, as confirmed by the confusion matrix analysis. The framework’s unsupervised nature and validated performance establish a practical solution for intelligent battery management systems.

Lithium-rich ferrite lithium (Li5FeO4, LFO) is a promising cathode pre-lithiation reagent for high-energy-density lithium-ion batteries, but its practical application is hindered by poor air stability and excessive gassing. In this work, we adopt a solid-phase synthesis method to prepare LFO and apply it as a cathode pre-lithiation reagent in 5 Ah lithium iron phosphate (LiFePO4, LFP)/graphite (Gr) full cells. We show that LFO readily hydrolyzes to form LiOH, which further reacts with CO2 even under low-humidity conditions. Adding 2 wt% LFO to a 5 Ah LiFePO4/ graphite pouch cell increases formation gassing from 18.1 +/- 0.8 to 65.6 +/- 1.6 cm3 and shifts the gas composition from predominantly reducing gases (93.6%) to a near-equal mixture (49.5%). Despite this, the LFO-containing cell retains 83.9% capacity after 1500 cycles at 45 degrees C, whereas the control cell reaches 82.9% after only 900 cycles. These findings highlight the trade-offs associated with LFO pre-lithiation and provide practical guidance for integrating lithium supplements into long-life, high-energy batteries.

Communication delays and packet loss in large-scale Cooperative Adaptive Cruise Control (CACC) platoons degrade the accuracy of leader state propagation and threaten string stability. To address these, a novel heterogeneous networking architecture is proposed, integrating satellite direct vehicle links, cellular vehicle to everything (C-V2X), and in-vehicle passive optical networks (PON). Based on this architecture, a fault-tolerant synchronous predictive control scheme is developed. First, an analytical model for vehicle communication loop delay is established, incorporating the transmission characteristics of in-vehicle optical networks, C-V2X, and satellite links. This model captures delay uncertainty in the heterogeneous network and provides explicit stability conditions for controller design. Second, a time-window Kalman filtering mechanism is designed to overcome information asynchrony and packet loss, enabling robust multi-modal state fusion of the leading vehicle. Furthermore, a composite control strategy is formulated by integrating delay-compensated model predictive control with an H∞-optimized linear quadratic regulator, which improves longitudinal acceleration tracking accuracy. Finally, hardware-in-the-loop experiments under realistic driving scenarios verify the effectiveness and robustness of the proposed architecture and control methods.

Insulated Gate Bipolar Transistor (IGBT) is recognized as having its Remaining Useful Life (RUL) prediction constitute a crucial component in the implementation of Prognostics and Health Management (PHM). The investigation of IGBT RUL under varying operating conditions is considered to possess significant theoretical importance and engineering value. A transfer learning-based RUL prediction method is proposed in this study. Initially, corresponding current and voltage signals are extracted from aging data according to the operational characteristics of IGBTs, from which the on-state resistance is calculated and characterized. Subsequently, features from both source and target domain data are systematically selected and fused to construct a health indicator with high transferability. Finally, a Maximum Mean Discrepancy-based Domain Adversarial Neural Network (MMD-DANN) is employed to minimize cross-domain discrepancies, thereby enabling high-accuracy RUL prediction to be achieved across different operating conditions. Experimental validation demonstrates that the proposed method is proven to provide an effective solution for reliability assessment of power devices under different operational conditions.

Eco-driving and bus bunching are two major challenges for connected and electric buses (CEBs). Eco-driving aims to minimize energy consumption, whereas bus bunching occurs when consecutive buses arrive at the same station simultaneously. The existing research rarely addresses both issues together. To address this gap, this article proposes a novel speed planning approach that addresses both problems via nonlinear model predictive control (NMPC) and imitation learning. A multiobjective NMPC model is developed that considers practical factors, such as energy consumption, time headway deviation, and traffic lights. To save computational resources, a speed planning network (SPN) based on transformer and long short-term memory (LSTM) architectures is designed to mimic the NMPC planner. Additionally, a knowledge distillation method is introduced to reduce the SPN's memory footprint by incorporating mixed knowledge. Extensive experiments show that the NMPC model ensures nonstop passage through intersections and performs better across multiple metrics, including energy consumption and time headway deviation, than several baselines do. The SPN achieves similar performance to NMPC while significantly improving real-time efficiency, and the proposed distillation method further reduces memory usage while maintaining acceptable performance.

Next-generation intelligent battery management systems (BMS) require accurate real-time estimation of battery state of health (SOH). However, existing studies often underestimate challenges arising from large volumes of online data with varying quality, as well as the resulting pressures on data storage, transmission, and computation. This paper proposes a lossy counting-based gated dual-attention Transformer (LC-GDAT) framework that substantially reduces historical data storage needs while maintaining high accuracy in SOH estimation. To overcome errors due to information loss from data compression, two critical modules are introduced. The first is the parallel temporal-spatial lossy counting feature extraction module (PTS-LC). It uses frequent-item extraction to identify important voltage and charging capacity patterns during battery operation. This significantly reduces storage demands and effectively transforms frequent items into two-dimensional features. The second module is the gated dual attention Transformer (GDAT). It uses a dual-branch structure to adaptively explore battery degradation characteristics from positional and channel dimensions. A gating mechanism is introduced to enhance interaction between these dimensions. The performance of LC-GDAT is comprehensively evaluated using data from 124 batteries under laboratory conditions, as well as real-world data from 20 electric vehicles collected over approximately 29 months. The experimental results show that LC-GDAT achieves the lowest SOH estimation errors of 0.46 % under laboratory conditions and 2.23 % under real-world conditions.

Energy efficiency has become one of the major concerns that hinder the penetration of freight electric commercial vehicles (CEV), due to frequent and long-distance transportation tasks. However, how vehicle kinematic states would influence the energy requirement of a given journey and how design modifications might affect the energy conversion efficiency of CEV powertrains have not been comprehensively considered in the development phase of CEVs. This work presents a novel mechanism for saving required energy output by actively adapting differences in wheel speed and explores how modifications to powertrain design would shape the energy utilization within CEV powertrains. Correspondingly, a novel powertrain configuration for improving energy conversion efficiency and exploiting the energy-saving mechanism is proposed. Validation suggests a 54.8 % reduction in energy consumption of CEVs by the synergy of powertrain modifications and the novel energysaving patterns, compared with the current CEV design. This result also opens the avenue for the next generation of CEVs.

Connected cruise control (CCC) is an advanced system that extends traditional cruise control by incorporating vehicle-to-everything (V2X) communications. However, as such technologies continue to iterate, conventional intra-vehicle communication systems struggle to meet escalating bandwidth demands, while cross-vehicle communication delays degrade CCC's closed-loop performance and increase collision risks. Motivated by these issues, a zone-centralized architecture integrating automotive passive optical networks and cellular V2X is proposed to reduce the upper bounds of loop delays compared to those in Ethernet-based domain-centralized systems. A novel cross-vehicle loop delay analysis framework is introduced to characterize communication system uncertainties with provable upper bounds. Building on this, a scheduling and control coordination scheme is developed to mitigate the cross-vehicle loop delays in intra-vehicle and inter-vehicle communications, while ensuring precise tracking platoon motion. For scheduling, a fraction-type basic period methodology and an earliest-deadline-priority strategy are employed to manage deterministic communications and alleviate the cross-vehicle loop delays. In terms of control, a delay-robust model predictive control approach is applied for decision-making and an H infinity -based linear quadratic regulator technique is used for vehicle longitudinal acceleration tracking while combating vehicle communication delays. Finally, the proposed scheduling and control coordination scheme undergoes validation across diverse driving scenarios, demonstrating its effectiveness and robustness through hardware-in-the-loop verification.

Battery mechanical properties degrade progressively with aging, manifesting as expansion pressure in module-constrained cells. Here, an in situ pressure operating system was developed to replicate the mechanical environment of lithium iron phosphate (LFP) prismatic batteries, enabling long-term monitoring under different loads and temperatures. Coupled with quasi-static compression tests on internal components, stress–strain curves and elasticity moduli were obtained to link microscopic behavior with macroscopic pressure response. Results show that irreversible pressure growth is jointly governed by state of health (SOH) and load: under low-load conditions, irreversible pressure increases nonlinearly with SOH, whereas higher loads yield more linear trends. A multilevel physical model encompassing electrodes, cells, and modules was proposed to explain these behaviors. This model takes into account the influence of external pressure on the modulus of the battery, and indicates that SOH and load influence reversible pressure curves through their effect on modulus. A theoretical method was derived to calculate in-module modulus, confirming its linear correlation with the fluctuation amplitude of reversible pressure. Differential pressure-capacity analysis further demonstrated that characteristic changes in expansion pressure reflect modulus evolution, and deviations from this relationship reveal degradation pathways such as gas generation, solid electrolyte interphase (SEI) growth, or lithium plating. This study establishes pressure signals as mechanistic indicators of modulus evolution and provides a framework for diagnosing mechanical degradation in batteries.

The pressing need to reduce greenhouse gas emissions and optimize traffic demand underlines the importance of effective travel demand management. Previous studies have explored budget-based and aggregated incentive programs, which diminish a heavy financial burden on governments and tend to be limited in contributing to effective behavior change in practice due to budget issues. This study proposes a personal carbon trading travel incentive (PCTTI) mechanism, to encourage private car commuters using low-carbon travel routes. The revenue obtained from the sale of carbon emission reductions, resulting from changes in commuter routes, serves as a partial budget for the incentives under PCTTI. To determine the optimal incentives, we developed an incentive scheme optimization model based on a bi-level programing model. Numerical analysis reveals the substantial potential of PCTTI to reduce carbon emissions and travel costs in the road traffic system, but also highlights the sensitivity of these results to the carbon trading price and the commuters' value of time. Specifically, the effectiveness of PCTTI diminishes when the carbon price drops below $6.494 per ton or when commuters’ value of time exceeds $3.99 per hour. These results indicate that the PCTTI mechanism offers a scalable and economically sustainable approach to enhance travel demand management and achieve environmental benefits.

Cooperative Adaptive Cruise Control (CACC) systems based on real-time vehicle-to-vehicle (V2V) communication are pivotal for enhancing traffic efficiency and safety in vehicular network environments. However, the adoption of open communication channels renders these systems more susceptible to attacks, particularly false data injection (FDI) attacks, which manipulate vehicle states and disrupt platoon stability. Firstly, the system delay under FDI attacks is meticulously analyzed based on the concept of multi-link loop delay, and its upper bound is derived. Then, a hierarchical control framework resilient to cyberattacks is proposed to address loop delays and implement FDI compensation control. The upper layer develops a model predictive controller (MPC) for decision-making and planning under uncertainties. The lower layer employs an $\boldsymbol{H}_{\infty}$ controller combined with a linear quadratic regulator (LQR) to mitigate the effects of loop delays and provide reliable acceleration tracking control. Finally, the effectiveness of the proposed method is validated through comprehensive hardware-in-the-loop testing.

State of health estimation of lithium-ion batteries is essential for ensuring operational safety, extending service life, and optimizing production processes. To address the limitation of relying solely on features from either the charging or discharging process, this study proposes a health state estimation method based on features extracted from both charging and discharging phases. Specifically, three features are extracted from the battery's charging and discharging curves and then evaluated for their correlation with SOH using the spearman correlation coefficient to ensure the effectiveness. A particle swarm optimization-optimized gate recurrent unit-based model is constructed to estimate state of health, and the proposed method is validated using the CALCE dataset and NASA dataset. Experimental results show that the root mean square error is controlled to around 1%, demonstrating that the method can effectively and accurately assess the SOH of lithium-ion batteries, providing valuable guidance for practical applications in battery management systems.

The demand-responsive transit (DRT) system can offer an alternative mode for providing feeder services to enhance the first/last mile (FM/LM) connectivity for metro stations. This study considers the feeder bus network design problem (FBNDP), which designs customised routes for serving FM/LM to/from a metro station. First, the electric vehicle (EV)-based vehicle scheduling problem (EVSP) of fleet size optimisation is combined with the FBNDP under a static DRT service for FM and LM problems. Second, a hybrid solution algorithm that combines a heuristic approach with the cloud model and various optimisation operators is designed to improve the quality of solutions. The proposed design method is evaluated on a case study of a real-world instance, and a sensitivity analysis to explore the influence of different operating strategies on system cost is also conducted. The results demonstrate how the proposed feeder bus system offers feasible service schemes while balancing convenience and cost.

Utilizing renewables locally for EV charging in highway service areas (HSAs) represents an optimal synergy between EVs and renewable energy. However, uncertainties in collective EV charging load and renewable generation make it challenging for HSA renewable systems to maintain real-time power balance and long-term sustainability with limited grid support. To address this, this work develops a deep learning-aided integrated optimization framework for system control and design. The method introduces a data-driven representation of complex design constraints under stochastic operation environments and evaluates the expected optimal system performance. In parallel, a novel system topology is engineered to enhance the matching between EV charging load and photovoltaic generation. Compared with the traditional isolated design, conditioned on the same capital investment, the mutual-aid system demonstrates a 20.2 % average improvement in comprehensive performance and a 6.4 % increase in self-sustain rate. The proposed method also regulates the interaction between the HSA renewable system and the grid by conducting at most 99.2 % of requests for grid supply through regular and economy day-ahead electricity purchases, which leads to a friendly and economical interaction among EVs, renewable systems, and the grid.

This work presents a hierarchical electrochemical-thermal-mechanical coupled model of Li-ion cells that integrates multiple submodels across particle, electrode, and cell levels. The model divides a single cell into various parallel-connected unit cells, each associated with a physics-based pseudo-2D (P2D) model capable of predicting mass and charge transfers at the electrode level and diffusion-induced stress/strain at the particle level. At the cell level, a heat-transfer submodel calculates the temperature distribution across the cell thickness, along with an electrical submodel to predict current partition and a phenomenological mechanical submodel to estimate force and displacement under preloading conditions. These through-plane submodels are bi-directionally coupled with the P2D models, enabling the translation of particle-level stress/strain changes to cell-level force and displacement evolutions under various preloading conditions. The model is validated against experimental data from three types of cells with different thicknesses, demonstrating its capability in predicting current/voltage, temperature, and displacement/force evolutions at various C-rates and preloading forces. The calibrated model is then extrapolated to explore the coupled electrochemical, thermal, and mechanical behaviors in large-format cells with considerably greater thickness. Significant non-uniformities are observed at two levels: across the electrode thickness and across the cell thickness. The former is associated with non-uniform current densities within a single electrode, leading to varying rates of lithiation/delithiation at different positions and, consequently, varying rates of particle volume changes. The latter is associated with non-uniform current and temperature distributions, resulting in non-uniform reaction-induced and thermal-induced displacement and force.