Search Results (5,555)

The increasing automation of mechanized forage production has improved operational efficiency, but achieving consistent product quality remains challenging because material conditions evolve continuously across sequential operations. Current control strategies primarily focus on machine-level objectives, such as load regulation, throughput stabilization, and actuator response, while the linkage between operational decisions and final forage quality is still insufficiently established. Machinery-, sensor-, and stage-centered reviews chiefly catalog technologies and prediction performance, without consistently testing whether observations are representative, transferable, timely, and assigned to actionable material. This critical narrative review integrates recent evidence identified through Web of Science Core Collection and CNKI searches, supplemented by citation tracing of foundational studies. It presents a state-centered framework for smart sensing and sensing-to-control pathways throughout the mechanized forage production chain. Key intermediate states, including moisture distribution, leaf integrity, windrow geometry and linear density, mass flow, particle characteristics, bale density, and pore structure, are analyzed as the links between mechanical actions and downstream outcomes, such as drying performance, material loss, fermentation, heating risk, and storage stability. Sensing technologies based on electrical, mechanical, optical, microwave, weighing, and spectroscopic principles are evaluated from the perspectives of measurement representativeness, transferability, latency, uncertainty, and operational validity rather than prediction accuracy alone. Prediction accuracy describes agreement with a reference, whereas decision utility reflects whether a valid, representative, and timely estimate can change an operational decision and reduce downstream quality risk. The reviewed evidence indicates that practical quality-oriented automation requires rapid proxy-state feedback for immediate machine regulation combined with batch-linked quality measurements for delayed verification and model updating. Future progress depends on establishing causal machine–quality relationships, transferable sensing models, cross-stage data continuity, and production-scale validation. Full article

Battery-electric and hybrid ships combine battery energy storage systems with networked control, machine learning state estimation, digital twins, high-capacity charging, and, in some applications, second-life cells. These configurations introduce risks for which maritime approval instruments may lack verifiable acceptance criteria. This paper defines a Regulatory Readiness Level (RRL) from 0 (unrecognised) to 4 (verifiable acceptance) and applies it to nine frontier risks in three families: cyber-physical, AI-driven, and second-life/lifecycle. The assessment covers IMO, EMSA, IACS, IEC, UL, NFPA, and battery requirements of five classification societies, using instruments in force on 31 July 2026. No risk reaches RRL 4 under the study’s target. Median readiness is RRL 2: one risk is at RRL 0, two at RRL 1, four at RRL 2, and two at RRL 3. The two RRL 3 cases illustrate different limitations: state-of-health estimation is verified periodically without qualification between tests, whereas mixed-chemistry and mixed-state-of-health packs are addressed through prohibition rather than qualification. Baseline priority places AI3, auditability of AI safety functions, first, with AI2 and SL1 in the next band; SL1 is sensitive to aggregation. Ordinal dominance excludes CP2, CP3, and SL1 from the lead without assuming equal spacing. Sensitivity analysis shows that uncertainty in readiness classification has a larger effect on ranking than uncertainty in severity and exposure. Adjacent-sector assurance mechanisms provide templates for maritime incorporation, and the paper proposes an acceptance objective, verification method, and regulatory vehicle for each risk. Full article

In the context of current climate issues and environmental constraints, green and renewable energies have experienced considerable growth in recent decades. Thus, there has been a renewed interest in the development of electric machines. To achieve this, the optimization of mass saving, volume and magnetic performance in the development of these machines must accelerate and be implemented industrially, particularly in the aeronautical field. The aircraft of the future will therefore require the use of electric machines with very high power densities and efficiencies. To achieve these objectives, one of the essential components to improve is the permanent magnet, and in particular its magnetic properties (increases in coercive field, residual magnetization and specific energy density). These magnets are likely to operate at temperatures above 200 °C. In this context, hard ferromagnetic magnets seem to be the most appropriate. SmCo₅ magnets have a Curie temperature between 720 and 920 °C. In this work we show the impact of grinding time on the magnetic properties of these SPS-treated magnets. This method allows the sintering SmCo₅ magnets while limiting their oxidation and their grain growth. By optimizing the grinding parameters, we obtain magnets with a coercive field of the order of 18,800 Oe. The novelty of this study lies in the optimization of EBSD (Electron Backscatter Diffraction) image analysis using the NLPAR (Non-Local Pattern Averaging Reconstruction) method, combined with advanced image correlation techniques. This approach enabled the precise determination of a grain size distribution centered around 50 nm, offering new perspectives on microstructural control and the enhancement of magnetic performance in SmCo₅ magnets. Full article

Rising costs, supply uncertainty, and geopolitical dependence associated with rare-earth permanent magnets have created an urgent need to reduce permanent magnet (PM) usage in electric machine designs while maintaining high electromagnetic performance. This paper presents a comprehensive comparative investigation of PM reduction strategies for three widely adopted traction machine topologies: Surface Permanent Magnet (SPM), Interior Permanent Magnet (IPM), and Permanent Magnet-Assisted Synchronous Reluctance (PMASynRM) machines. We developed a unified evaluation framework by integrating each machine’s electromagnetic characteristics, electrical operating constraints, and practical inverter limitations across a wide operating speed range. Finite element analysis (FEA) is employed to systematically investigate the effects of progressive PM volume reduction by independently varying magnet width and length while keeping stator geometry, winding configuration, material properties, and operating conditions. We reduced PM dimensions from 100% to 30% of the baseline design to evaluate their impact on torque-speed characteristics, efficiency, and flux-weakening capability. The results reveal that the impact of PM reduction depends strongly on both machine topology and the magnet reduction strategy. Across all machine topologies, reducing magnet width consistently preserves higher torque capability, efficiency, and high-speed operating performance compared with reducing magnet length. Among the machines investigated, the SPM topology exhibits the greatest sensitivity to PM reduction, particularly in the constant-torque region, owing to its complete reliance on permanent magnet excitation. In contrast, the IPM and PMASynRM machines show much greater tolerance to reduced PM volume because the additional contribution of reluctance torque from rotor saliency improves torque retention, extends the constant-power speed range, and reduces efficiency degradation. These findings provide valuable design insights into the trade-offs between permanent magnet utilization and machine performance, offering practical guidelines for minimizing rare-earth material consumption while maintaining competitive traction performance in next-generation electric vehicles and other sustainable electrification applications. Full article

The rapid adoption of electric vehicles (EVs) has intensified the demand for advanced battery technologies capable of delivering higher energy density, improved safety, faster charging, and longer operational life. This review provides a comprehensive assessment of the evolution of EV battery technologies, covering conventional batteries, lithium-ion batteries, and emerging next-generation chemistries, including solid-state, lithium–sulfur, sodium-ion, and lithium–air batteries. Key battery performance characteristics, such as energy density, power density, efficiency, cycle life, charging and discharging behavior, and degradation mechanisms, are critically discussed to highlight their influence on battery performance and lifespan. The review further examines the application of artificial intelligence (AI) in battery management systems, emphasizing machine learning, deep learning, reinforcement learning, and hybrid AI techniques for improving battery state estimation, fault diagnosis, predictive maintenance, thermal management, and charging optimization. The roles of State of Charge (SOC), State of Health (SOH), and Remaining Useful Life (RUL) estimation in enhancing battery reliability and operational safety are also reviewed. In addition, current research gaps related to battery degradation, fast charging, thermal management, recycling, and explainable AI are found, together with future research trends involving digital twins, smart charging, Vehicle-to-Grid (V2G) integration, and sustainable battery technologies. The review concludes that the integration of next-generation battery chemistries with AI-driven battery management systems offers significant potential to improve battery efficiency, extend service life, enhance safety, and accelerate the widespread adoption of electric vehicles while supporting the global transition toward sustainable and intelligent transportation systems. Full article

Legacy computer numerical control (CNC) machines often lack controller connectivity or sensing suites, making it difficult to monitor utilization and energy demand or deploy data-driven maintenance without mechanically invasive modifications. This study develops a black-box method that links net input current signatures to CNC operating states. Four current-sensor channels recorded 70,000 samples per second each, generating approximately 504 million samples across 44 repetitions of an unloaded Haas Mini Mill cycle containing dwell periods and bidirectional motion of the spindle and each linear axis. Cycle-to-cycle variability supported semiautomated, sequence-guided transition detection, while frequency-domain signatures were evaluated across similarity measures, baseline compensation, window functions, overlap, multicycle aggregation, and multichannel Bayesian analysis. The active cycle was distinguishable from a powered-idle baseline, and the reconstructed sequence represented 39.99 s of the measured 40.94 s cycle. Cross-correlation produced stronger intra-dataset screening results than Pearson correlation under the magnitude-based procedure; baseline compensation increased the Bayesian screening score from 36.1% to 81.0%. With a rectangular window and positive-match evidence across all four channels, the conservative minimum estimated confidence reached 95% across the eight states. This value describes segmented-signature distinguishability within the dataset, not independently tested classification accuracy. The results provide proof-of-concept evidence for controller-independent monitoring under one fixed unloaded program and a basis for broader experimental validation. Full article

Software-defined electric vehicles have reached micromobility, creating a growing need to evaluate how controller software shapes drive performance and energy use. This requires experimental platforms with quantified emulation fidelity and measurement performance. This paper presents a scenario-based hardware-in-the-loop platform combining a production e-scooter drive, a controlled magnetic particle brake, and a real-time road-load model. The platform reproduces rolling resistance, aerodynamic drag, and speed-gated inertial demand for a nominal 90 kg vehicle. Across eight repetitions of a scaled ECE-15 urban cycle, including a cold first run, the protective load cap retained 99.51% of the commanded load impulse. A post hoc sensitivity analysis of the seven warm runs gave 99.71%. Measured torque and speed tracking root-mean-square errors over all eight runs were 0.704 N·m and 0.593 km/h. Separate electrical characterization produced a machine efficiency map peaking at 86.3% and cycle-energy measurements with a coefficient of variation of 0.72%. Subsampling unfiltered 200 kHz recordings at every sampling offset gave a maximum absolute cycle-energy error of 0.14% at 2 kHz, a hundredfold reduction in retained samples under the tested conditions. These results provide a quantified experimental basis for micromobility drive testing and subsequent controller-software comparisons. Full article

Direct-drive, low-speed, high-torque (LSHT) applications operating at a fixed rated frequency require electrical machines with high pole numbers. However, synchronous reluctance machines (SynRMs) are often considered unsuitable in this range because increasing the pole number can reduce saliency and power factor. This paper investigates the feasible pole-number range of magnet-free SynRMs for direct-drive operation and compares inner- and outer-rotor configurations. A systematic optimisation campaign, comprising 93,650 two-dimensional finite-element-evaluated designs, was performed for 12- to 24-pole machines operating at 50 Hz. The designs were evaluated under fixed constraints on machine envelope, terminal voltage, and total losses, while efficiency, torque density, power factor, and torque ripple were considered as performance objectives. Inner-rotor designs satisfied all targets up to 18-pole; beyond this point, the reduction in saliency required higher stator current, thereby increasing copper losses. By contrast, the outer-rotor configuration enabled a larger air-gap diameter within the same envelope, preserving the saliency ratio and satisfying all targets up to 24-pole at the 20 °C reference condition adopted for the screening, the best 24-pole design reaching 91.2% efficiency and 17.1 kNm/m³ at 250 rpm. The results demonstrate the potential of outer-rotor SynRMs as magnet-free candidates for low-speed, high-torque direct-drive applications. Full article

Manufacturer-defined driving modes modify traction-power availability, accelerator response, regenerative braking, and auxiliary-system operation, but their comparative effects under real-world urban and suburban conditions remain insufficiently characterized. This study evaluated Normal, ECO, and ECO+ modes of a Volkswagen e-UP under two real-world route conditions. Each mode–route combination was tested in three complete-route runs, resulting in 18 tests. Electric-machine power recorded through the diagnostic interface was processed using the original timestamps to determine route-normalized net energy consumption and regenerative-energy recovery, while battery state of charge was used as a supporting indicator. On the urban route, mean energy consumption was 9.90, 9.08, and 8.83 kWh/100 km in Normal, ECO, and ECO+ modes, corresponding to reductions of 8.3% and 10.8% relative to Normal. On the suburban route, the respective values were 10.69, 10.41, and 10.04 kWh/100 km, corresponding to reductions of 2.6% and 6.1%. Although the percentage reductions were larger under urban conditions, the driving-mode × route interaction was not statistically significant (p = 0.111). Methodologically, the study demonstrates a structured research approach based on repeated complete-route runs, run-to-run motion comparability assessment, and decoupled accounting of propulsion demand and recuperation. The results show that route-normalized net electric-machine energy consumption and regenerative-energy recovery should be evaluated jointly when assessing manufacturer-defined EV driving modes. Full article

Energy storage is a critical step for the decarbonization of our society by exploiting renewable energies. Carnot batteries are promising technologies for storing excess electricity from renewables, but their efficiency is generally limited. The present work suggests a novel idea for enhancing the performance and flexibility of Carnot battery units. Specifically, a hybrid system that exploits both electricity and waste heat streams is suggested, incorporating two stages of latent thermal storage to increase flexibility. Moreover, the present system uses an innovative cascade high-temperature heat pump, with an absorption heat transformer in the low stage and a vapor-compression unit in the upper stage. The power production machine is a recuperative organic Rankine cycle. Practically, the absorption heat transformer exploits the waste heat stream by upgrading it and reducing the temperature lift that the vapor-compression heat pump produces, so there is increased electrical performance with suitable consumption of the waste heat stream. Also, the respective configuration without the absorption heat transformer is studied to make a suitable comparison. The results of this analysis show that the adoption of the AHT enhances the roundtrip efficiency from 26.30% up to 92.31% (relative enhancement) compared to the system without it. For the main scenario with a waste heat stream temperature at 100 °C, the roundtrip efficiency was found to be 72.86% and the exergetic efficiency as 35.02% for the system with the absorption heat transformer. Full article

Material removal rate and material removal per unit energy are key indicators for evaluating EDM efficiency. By studying the relationship between these indicators and EDM discharge parameters, it is possible to improve machining efficiency and material removal performance. Taking the EDM structure with magnetic-levitation micro-motion compensation as the research object, this paper combines the dynamic levitation position of the moving electrode with the formula of the gap between electrodes and establishes a mathematical model among the macroscopic feed position of the spindle, the dynamic compensation position of the moving electrode, the pulse energy and the heat flux density of the workpiece. It also analyzes the conversion process of single-pulse energy and the Gaussian heat flux density on the workpiece surface. By combining the response surface method, the mapping relationship between electrode discharge parameters and material removal rate, as well as the material removal amount per unit energy, was obtained. Finite-element simulations and single-factor continuous machining experiments show that the MRR is more sensitive to the low-voltage current and pulse interval, whereas the material removal per unit energy exhibits a non-monotonic dependence on the discharge parameters. Specifically, it shows a decreasing trend followed by an increasing trend with increasing low-voltage current and pulse width, whereas the opposite trend is observed with increasing high-voltage current and pulse interval. These results suggest that a moderate low-voltage current, a relatively high-voltage current, a short pulse width, and an appropriate pulse interval are favorable for magnetic-levitation micro-motion-compensated EDM to achieve both high MRR and high material removal per unit energy. Full article

This study presents a comprehensive machine learning (ML) evaluation methodology, grounded in physical principles, for range estimation in battery-electric vehicles (BEVs). In a simulation environment consisting of 50 driving and 12 virtual vehicles, 2000 observations were generated by classifying 23 input variables into three main categories: battery and energy status (SOC, voltage, current, battery power, rolling energy consumption, SOH, battery temperature, regenerative power), driving and route context (speed, acceleration, cumulative distance, route gradient, traffic density, driving style, route type, load), and environmental and thermal context (ambient temperature, relative humidity, wind speed, headwind component, precipitation intensity, solar radiation, HVAC power). The Gradient Boosting (GB) model, selected from among eight algorithms, achieved an R² = 0.8949, a mean absolute error (MAE) of 12.58 km, and a root mean square error (RMSE) of 16.83 km in the independent test set; In 79.25% of the test observations, the error remained below 20 km. Permutation significance analysis revealed the overwhelming dominance of battery charge state (SOC) and instantaneous energy consumption (R² reductions of 1.435 and 1.084, respectively). Meanwhile, group-based permutation tests showed that the battery/energy group provided an R² reduction of 2.0815, compared to reductions of only 0.0419 and 0.0413 for the environmental/thermal and driving/route groups. Ablation tests showed that the model created with only the battery/energy variables achieved R² = 0.8472 and MAE = 14.17 km, performing close to the full model (R² = 0.8949, MAE = 12.58 km); however, the model created with only SOC proved unable to generalize in independent driving (R² = −0.0458). These findings quantitatively confirm that remaining energy and current consumption are the primary determinants of range estimation, while driving, route, and climate variables provide only minor, corrective contributions, and demonstrate that the proposed synthetic framework offers a methodological basis for validation studies using real-world data. Full article

Fixed-threshold battery management systems can miss faults that appear across several signals at once, which has encouraged the use of hybrid machine-learning detectors for electric-vehicle (EV) batteries. This study examines when combining such detectors actually improves anomaly detection. Using a public EV Battery Charging Dataset of 1900 one-second observations, four controlled fault types were injected: voltage imbalance, thermal instability, resistance growth, and sensor malfunction. The record was segmented into 125 overlapping windows and represented by 36 statistical and temporal features, which principal component analysis reduced to seven components. Isolation Forest, Local Outlier Factor (LOF), and an autoencoder were evaluated individually and in majority-vote and weighted-score ensembles. Six more recent machine-learning and deep-learning detectors were evaluated on the same windows for comparison. A Support Vector Machine was excluded from fusion because its outputs could not be aligned with the ground-truth windows. LOF gave the best discrimination (ROC-AUC 0.852), whereas the weighted ensemble reached 0.805 and majority voting 0.623. None of the recent detectors outperformed LOF, and no difference was statistically significant. Fusion therefore did not improve on the strongest single detector. Detector admissibility, complete score alignment, and leakage-free validation should be verified before ensemble gains are claimed for battery monitoring. Full article

Self-powered wearable photonic sensors enable energy-autonomous operation by integrating optical sensing, energy harvesting, and signal transduction within a unified platform. These systems exploit photophysical mechanisms such as photovoltaic, photogating, and triboelectric effects to convert ambient or body-derived energy into electrical signals while simultaneously performing sensing. Advances in hybrid semiconductors, two-dimensional materials, and nanocomposites have significantly improved light absorption, carrier transport, and mechanical compliance, enabling integration into flexible and textile-based wearable formats. Device architectures have evolved from simple junctions to heterostructured and tandem configurations that enhance spectral selectivity and energy utilization under variable illumination. However, reliable operation under low-intensity and dynamically changing conditions remains a key challenge, particularly in balancing sensing performance with energy autonomy. This review analyzes the interplay between materials, photophysical mechanisms, device architectures, and system-level design, with emphasis on energy harvesting–sensing co-optimization. Challenges related to mechanical stability, signal reliability, and real-world deployment are discussed, alongside emerging solutions based on adaptive photonic systems and intelligent data processing. These insights provide a framework for next-generation wearable photonic platforms for healthcare, environmental monitoring, and human–machine interfaces. Full article

Hydropower is recognized as a clean and low-carbon energy source, playing an important role in the green development of various countries. Cavitation is likely to occur at the turbine blades and pipes under complex working conditions. Machine performance is significantly affected by this phenomenon. The prediction of cavitation evolution trends in turbines is considered to have considerable engineering application significance for improving the operational efficiency of turbines. However, due to noise interference, cavitation features are easily submerged. Long-distance cavitation information is difficult to capture because of the limitations of existing network structures. A double hierarchical residual long short-term memory (DHRLSTM) network has been proposed. Cavitation information is captured from the underwater acoustic signal through three one-dimensional convolutional layers arranged in two branches. Cavitation information can be directly transmitted from input to output through the residual module. The vanishing gradient problem in deep networks is effectively solved. Early cavitation features are retained, allowing cavitation information to be mined more deeply. More accurate predictions are provided as a result. Finally, comparisons were made between the underwater acoustic signal data from the mixed-flow turbine model test bench of Harbin Large Electric Machinery Research Institute and various prediction methods, followed by ablation experiments. It has been proven by experimental results that the proposed method achieves good prediction accuracy. Full article