A State of Health (SOH) estimation of lithium-ion batteries is a core function of battery management systems, directly affecting the safe operation, lifetime prediction, and economic efficiency of batteries. However, existing methods still face challenges in balancing feature robustness and model generalization ability; for instance, some studies rely on features whose physical correlation with SOH lacks strict verification, or the models struggle to simultaneously capture the temporal dynamics of health factors and nonlinear mapping relationships. To address this, this paper proposes an SOH estimation method based on incremental capacity (IC) curves and a Temporal Convolutional Network—Relevance Vector Machine (TCN-RVM) model, with core innovations reflected in two aspects. Firstly, five health factors are extracted from IC curves, and the strong correlation between these features and SOH is verified using both Pearson and Spearman coefficients, ensuring the physical rationality and statistical significance of feature selection. Secondly, the TCN-RVM model is constructed to achieve complementary advantages. The dilated causal convolution of TCN is used to extract temporal local features of health factors, addressing the insufficient capture of long-range dependencies in traditional models; meanwhile, the Bayesian inference framework of RVM is integrated to enhance the nonlinear mapping capability and small-sample generalization, avoiding the overfitting tendency of complex models. Experimental validation is conducted using the lithium-ion battery dataset from the University of Maryland. The results show that the mean absolute error of the SOH estimation using the proposed method does not exceed 0.72%, which is significantly superior to comparative models such as CNN-GRU, KELM, and SVM, demonstrating higher accuracy and reliability compared with other models.

To address the challenges of short dwell time near top dead center (TDC) and uneven heat release, this paper presents a comprehensive analysis of the effects of different ignition schemes on combustion characteristics, flame formation and development, and emissions. A three-dimensional model of coupled reaction’s kinetic mechanism was established using Converge 3.0 and validated by experimental data. The results show that ignition position, whether synchronous or asynchronous changes, significantly influence pressure. The pressure in synchronous cases can reach up to 62.5 bar, representing a 10.8% increase, exhibiting a distinct upward trend with advanced ignition position. In asynchronous cases, the pressure variation shows a distinct nonlinear characteristic due to the negative effects of in-cylinder airflow and flame core collision. When the ignition position is advanced, the ignition delay increases for both synchronous and asynchronous strategies. However, for synchronous cases, the combustion duration is reduced by up to 1.5 ms, whereas for asynchronous cases, the reduction is only 0.135 ms. Regardless of the schemes, the layout and the strong counterclockwise swirl lead to the flame core gradually developing from right to left, ultimately engulfing the left-side flame core. Compared then to that case, the left and right flame kernels may collide prematurely, leading to incomplete local combustion and consequently reducing combustion efficiency. Compared to synchronous changes, the emission differences during asynchronous changes are smaller and maintained at a relatively low level.

This paper focuses on the challenges of the free-piston internal combustion linear power generation system, which is characterized by a wide frequency and amplitude variation range of output electric power and difficulties in rectification. The study delves into the power generation characteristics during the system’s stable operation and explores strategies for rectifying three-phase electric power. A mathematical model of the power generation system has been established. The model of the linear motor has been verified, and an in-depth analysis of the electric power output characteristics has been carried out. On this basis, this paper proposes several rectification schemes, including uncontrolled rectification, PWM rectification based on a PI control strategy, and compound control based on model predictive sliding mode control. In terms of control strategy, a model predictive sliding mode compound control algorithm is introduced to optimize traditional PWM rectification. By comparing and analyzing the electric power output effects under different rectification schemes, this paper aims to explore the rectification strategy most suitable for the system. Simulation results indicate that while uncontrolled rectification can achieve power conversion, it leads to further phase current distortion on the linear motor side. This paper examines the last two types of rectification strategies in terms of robustness, accuracy, and responsiveness. It is found that model predictive sliding mode compound control performs better than PI control. It can stably regulate the output voltage, and the phase current on the linear motor side will not be distorted. It is more suitable for power processing in this system.