XJXiaohang Jin

Xiaohang Jin

*This is an algorithmically generated author record based on publicly available information.

Current Organizations / Institutes

The organizations/institutes from the scholar's latest article

Previous Organizations / Institutes

The organizations / institutes from the scholar's other articles in the past ten years


Xiaohang Jin has a total of 97 co-authors. Here are the top co-authors by number of shared publications:


Latest Publications

  • Open Access

    A two-stage multi-task early prediction model for lithium ion battery with uncertainty quantification

    by Xing-Yan Yao, Liwei Chen, Weijun Hu, Xiaohang Jin, Xiangyang Xu, Li Wang, Cheng Qian, Ijaz Haider Naqvi, Hassan Abbas Khan, Daeil Kwon, et al.

    Accurate early prediction of lithium-ion battery degradation is essential for safe and reliable energy storage systems, yet it remains challenging due to complex aging mechanisms and prediction uncertainty. This paper proposes a novel two-stage multi-task early prediction framework with uncertainty quantification. The model decomposes the prediction task into inflection point identification, cycle life prediction, and trajectory reconstruction, ensuring the reconstructed SOH curve strictly follows physical decay patterns. Three characteristic curves (V-Q, IC-V, DV-Q) are integrated by using a multi-scale architecture combining parallel CNNs and iTransformer for local and global feature extraction. Furthermore, conformal prediction is employed to provide not only high-accuracy point estimates but also prediction intervals with a guaranteed 95% confidence level, offering quantifiable risk indicators. Validated on the HUST and XJTU datasets, our results demonstrate that the proposed model achieves superior prediction accuracy and provides well-calibrated uncertainty bounds, representing a significant advancement towards trustworthy and intelligent battery management systems. The errors of the model with complete input of three characteristic curves are the lowest compared with other three single characteristic curves. The MAE and the RMSE of the proposed framework on XJTU datasets achieved to 0.0058% and 0.0071%.

  • Open Access

    A Local-Adaptive MSET Framework for Lithium-Ion Battery SOH Estimation from Partial Charging Data

    by Gaochao Wu, Zhi Rui, Ziqian Kong, Zijun Que, Xiaohang Jin, Zhengguo Xu

    Accurate state-of-health (SOH) estimation of lithium-ion batteries remains challenging in practical battery management scenarios, where available observations are often limited to partial charging segments. Under such conditions, handcrafted health indicators (HIs) may become sensitive to the selected charging window, while data-driven models may suffer from reduced robustness under varying operating protocols and limited interpretability. To address these issues, this paper proposes a local-adaptive multivariate state estimation technique (MSET) framework for SOH estimation from partial charging data. Handcrafted HIs and neural representations are extracted from charge-related curves and organized as grouped features for local-adaptive MSET-based capacity reconstruction. The framework assigns optimized weights to different feature groups and uses a local-adaptive kernel operator to account for variations in the data distribution. Experiments on a self-built lithium-ion battery aging dataset with four operating protocols examine the method under mixed-protocol SOH estimation, cross-protocol settings, and multiple voltage-window analysis. The proposed framework shows strong estimation accuracy, robustness to protocol shifts, and stable performance across voltage windows. External validation on the XJTU benchmark further confirms its generalization ability.

  • Open Access

    A Local-Adaptive MSET Framework for Lithium-Ion Battery SOH Estimation Using Partial Charging Data

    by Gaochao Wu, Zhi Rui, Ziqian Kong, Zijun Que, Xiaohang Jin, Zhengguo Xu

    Accurate state-of-health (SOH) estimation of lithium-ion batteries remains challenging in practical battery management scenarios, where available observations are often limited to partial charging segments. Under such conditions, handcrafted health indicators (HIs) may become sensitive to the selected charging window, while data-driven models may suffer from reduced robustness under varying operating protocols and limited interpretability. To address these issues, this paper proposes a local-adaptive multivariate state estimation technique (MSET) framework for SOH estimation from partial charging data. Handcrafted HIs and neural representations are extracted from charge-related curves and organized as grouped features for local-adaptive MSET-based capacity reconstruction. The framework assigns optimized weights to different feature groups and uses a local-adaptive kernel operator to account for variations in the data distribution. Experiments on a self-built lithium-ion battery aging dataset with four operating protocols examine the method under mixed-protocol SOH estimation, cross-protocol settings, and multiple voltage-window analysis. The proposed framework shows strong estimation accuracy, robustness to protocol shifts, and stable performance across voltage windows. External validation on the XJTU benchmark further confirms its generalization ability.

  • Knowledge Graph Transformer Network for Health Status Assessment

    by Yuanming Zhang, Jiacheng Huang, Wenbo Zheng, Gang Xiao, Xiaohang Jin

    Health status (HS) assessment has important scientific significance and practical value to ensure safe equipment operation and reduce maintenance costs. Graph neural network-based methods extract spatiotemporal features from sensor monitoring data, and show great promise. However, such methods focus on monitoring data, to the exclusion of other important data, such as component data or prior knowledge, rendering them unable to accurately characterize the relationships between monitoring data and HS. This paper proposes a knowledge graph transformer network (KGTN) for HS assessment. An HS knowledge graph is constructed by integrating monitoring data, component data, and prior knowledge. Meta-paths are utilized to describe the monitoring relationships among heterogeneous data, thereby achieving a comprehensive representation of health status. A knowledge graph transformer network extracts accurate spatio-temporal features from the HSKG, automatically extracting meta-paths closely related to HS, and performs feature transformations to obtain accurate feature representations, which are utilized to classify the HS in a vector space. In experiments on C-MAPSS public datasets, the proposed model achieves average improvements of 1.12% in Micro-F1 and 3.08% in Macro-F1 compared to the state-of-the-art baselines.


  • Knowledge Graph Metric Learning Network for Few-Shot Health Status Assessment

    by Gang Xiao, Ying Cao, Jiacheng Huang, Xiaohang Jin, Yuanming Zhang

    Health status assessment plays an important role in guaranteeing equipment operating safety and reducing maintenance cost. Traditional deep learning methods require large amounts of labeled samples for training, making them ineffective in most practical scenarios with few-shot labeled samples. To address this challenge, we propose a novel knowledge graph metric learning network for few-shot health status assessment. We first construct a health status knowledge graph to fuse sensor monitoring data, component data, and prior knowledge, providing a new approach to describe the spatiotemporal characteristics of equipment. Then, even with few-shot samples, our knowledge graph metric learning network effectively extracts spatiotemporal features from the health status knowledge graph. This network can embed overall structure and node features of health status knowledge graph into low-dimensional vectors, and can learn the similarity of few-shot samples in the vector space. By transforming the health status assessment into a few-shot classification problem in the graph representation space, our model achieves high performance with few-shot labeled samples. Experimental results on public datasets demonstrate that our method is significantly comparable to existing full-shot approaches.


  • Remaining Useful Life Prediction of Radioisotope Thermoelectric Generator Using a Component-Level Mechanism-Based Nonlinear Additive Wiener Process

    by Zixuan He, Ziqian Kong, Xiaohang Jin, Zijun Que, Zhengguo Xu

    Accurate prediction of the remaining useful life (RUL) of a general-purpose heat source radioisotope thermoelectric generator (GPHS-RTG) is critical for mission scheduling and power allocation in spacecraft. The key challenges in this task are long-term predictability and uncertainty quantification due to the limited data and variations in load conditions. However, little reported literature can be found addressing both issues. In this study, a component-level mechanism-based nonlinear additive Wiener process framework was proposed to improve the effectiveness of RUL prediction. Generally, a novel component-level mechanism empirical function developed via energy transfer path and kinetics was adopted to regulate the degradation trend, and a modified nonlinear additive Wiener process achieved uncertainty quantification and probabilistic prediction of RUL. In practice, the empirical function was regarded as a nonlinear drift of Wiener process, while the drift coefficients were randomized to represent component heterogeneity. This study derived parameter estimation by expectation-maximization algorithm and provided a probability density function under the first hitting time of RUL. Experimental results on NASA’s GPHS-RTG dataset demonstrate that the proposed empirical function outperforms existing counterparts and LSTM networks, exhibiting strong long-term predictive capability. The proposed RUL prediction framework exhibits higher accuracy than traditional one- and two-stage nonlinear Wiener processes, which is helpful for task planning. Moreover, the proposed empirical function enables the identification of component-level power losses, helping the detailed power analysis.


  • Incremental Learning Method for Wind Turbine Fault Detection Models Considering False Negatives

    by Ziqi Wang, Xiaohang Jin, Zhengguo Xu

    Fault detection (FD) algorithms based on supervisory control and data acquisition (SCADA) data have been widely used in the operation and maintenance of wind turbines (WT). However, the performance of FD models will degrade due to the time-varying operating conditions (TVOC). Incremental learning (IL) methods can be employed to update the models online to adapt to TVOC, but the error accumulation caused by false negatives will lead to a continuous decrease in the fault detection rate (FDR). A novel IL method considering false negatives is proposed to improve the performance of WT FD models. Firstly, a data buffer is built to cache some new normal data used in model updating. Secondly, a processing strategy for false negatives is proposed to block some high-risk data from being added to the buffer, thereby weakening the error accumulation. Thirdly, when the buffer is full, update the FD model using the data in it. A real-world SCADA dataset with gearbox faults and four different FD algorithms with various model updating strategies are used in the experiments. The results demonstrate that the proposed method can lower the false alarm rate (FAR) of all four FD algorithms. After processing false negatives, for example, for the FD model based on Gaussian kernel regression, its FDR increased by about 11%, and for the FD model based on multivariate state estimation technique, its FDR increased by about 14%. The results of hyperparameter experiments and early stopping experiments show that the proposed method has good potential for practical applications.


  • A Guided Multiobjective Optimization Method for Attenuation Calibration in Distributed Fiber Sensing

    by Jiamu Ling, Chenli Shi, Feng Wang, Wei Ye, Xiaohang Jin, Zhuangwei Xu, Zhengguo Xu

    Optical fibers serve as both signal transmission channels and sensing units in distributed fiber sensing systems. The sensor’s detection signal may be adversely affected by multiple forms of attenuation due to unavoidable bending and splices during fiber installation. Ensuring signal reliability requires precise calibration of signal attenuation. The attenuation calibration of multiple measurement points along the optical fiber is transformed into an error-related large-scale multi-objective optimisation problem, and resolved via a priori knowledge guiding evolutionary algorithm, which makes full use of the redundancy of data from adjacent measurement points and ensures the accuracy of each measurement point. As a means of introducing a priori knowledge to guide the search, the dynamic beta distribution of the population initialization in the evolutionary algorithm is used to achieve better convergence performance. To assess the calibration efficiency, experiments are conducted at various temperatures, revealing that the calibration at a temperature of 0°C led to significant reductions of 24.29 %, 0.23 % and 99.91 % for each objective function. Moreover, post-calibration temperature measurements at multi temperature points and the dynamic temperature measurement exhibit reduced mean absolute error, enhancing the accuracy of temperature measurements. This research offers a fresh perspective on the utilization of optimization algorithms, presenting an innovative approach to simultaneously calibrate multiple measurement points.