Lithium-ion batteries, essential components in new energy vehicles and energy storage stations, play a crucial role in health-status investigation and ensuring safe operation. To address challenges such as limited estimation accuracy and a weak generalization ability in conventional battery state of health (SOH) estimation methods, this study presents an integrated approach for SOH estimation that incorporates multiple health indicators and utilizes the multi-scale convolutional neural network (MSCNN) model. Initially, the aging characteristics of the battery are comprehensively analyzed, and then the health indicators are extracted from the charging data, including the temperature, time, current, voltage, etc., and the statistical transformation is performed. Subsequently, Pearson’s method is employed to analyze the correlation between these health indicators and identify those with strong correlations. A regression-prediction model based on the MSCNN model is then developed for estimating battery SOH. Finally, validation using a publicly available lithium-ion battery dataset demonstrates that, under similar operating conditions, the mean absolute error (MAE) for SOH estimation is less than 0.67%, the mean absolute percentage error (MAPE) is less than 0.37%, and the root mean square error (RMSE) is less than 0.74%. The MSCNN has good generalization for datasets with different working conditions.

This paper proposes a new Asymmetric Permanent Magnet Vernier Motor (A-PMVM) for low-speed high-torque applications. Unlike conventional symmetric V-shaped PMVMs (SV-PMVMs), the A-PMVM features irregular U-shaped magnet arrays composed of asymmetric V-shaped magnets. Finite element analysis confirms its superior performance: 10.6% higher torque (19.67 N·m vs. 17.78 N·m), 22% reduced PM volume (37,500 mm³ vs. 48,000 mm³), and 53% lower cogging torque (0.32 N·m vs. 0.68 N·m peak-peak). While exhibiting higher initial torque ripple ratio (8.65%), multi-objective optimization suppresses torque ripple ratio by 5.32% (from 8.65% to 8.19%), reduces cogging torque 12.5% (from 0.32 N·m to 0.28 N·m), and enhances torque by 0.76% (from 19.67 N·m to 19.82 N·m). The optimized A-PMVM achieves a significant reduction in cogging torque and torque ripple ratio, demonstrating significant potential for applications like wind turbines and electric vehicles. Additionally, this paper confirms that the proposed motor maintains consistent performance during both clockwise and counterclockwise operation.

Permanent magnet vernier motors (PMVMs) have significant advantages in low-speed direct-drive fields on account of their high torque density, and their performance improvement is still a research hotspot. To enhance the overall electromagnetic performance and provide an alternative solution for low-speed direct-drive applications, this paper proposes a permanent magnet vernier motor with rotor auxiliary teeth (denoted as “RAT-PMVM”). Firstly, the structure and working principle of RAT-PMVM are introduced. Then, the two-dimensional (2D) finite element method (FEM) is used to comparatively study the influence of the number, position, and tooth profile of the rotor auxiliary teeth on the electromagnetic performance of the proposed motor. The results show that the RAT-PMVM with trapezoidal teeth (denoted as “TT-PMVM”) achieved improvement in output torque, efficiency, and power factor: the output torque increased from 11.32 Nm to 14.19 Nm, the efficiency increased from 88.5% to 92.2%, and the power factor increased from 0.60 to 0.71. Finally, in order to further reduce the torque ripple and improve the torque, power factor, and efficiency, multi-objective optimization of the TT-PMVM is carried out. The optimization yields a 27.3% increase in torque, a 31.8% reduction in torque ripple ratio, an efficiency improvement from 92.2% to 93%, and a power factor enhancement from 0.73 to 0.81, demonstrating significant potential for low-speed direct-drive applications like industrial robots and wind power generation.

Rolling bearings are indispensable elements in mechanical equipment, and their condition is closely related to system reliability and operational safety. To enhance diagnostic performance with limited samples and reduce the dependence on manual parameter selection, this study proposes a fault diagnosis approach that combines Continuous Wavelet Transform (CWT), Convolutional Neural Network (CNN), Black-winged Kite Algorithm (BKA), and Least Squares Support Vector Machine (LSSVM). The original 1-D vibration signals are first processed by CWT to obtain 2-D time–frequency representations. CNN is then employed to learn deep fault-sensitive features, which are subsequently fed into LSSVM for state classification. To further improve classification performance, BKA is used to automatically search for the optimal LSSVM parameters, with validation accuracy adopted as the fitness criterion. Experiments conducted on three public bearing datasets, namely CWRU, JNU, and SEU, indicate that the proposed method outperforms CNN, CNN-SVM, and CNN-BiGRU under small-sample conditions. In addition, t-SNE results show more distinct feature clusters, while BKA exhibits faster convergence and better global search capability than Particle Swarm Optimization (PSO) and Genetic Algorithm (GA).