Jin-Zhen Kong

Jin-Zhen Kong

Auto-generated profile, based on publicly available information

Jin-Zhen Kong

School of Mechanical Engineering, Hebei University of Technology, Tianjin, China

Research Focus & Metrics

Academic Career Length

8Years (2019-Present)

(Calculated once Scilit publication data is synced: current year − year of first publication + 1)

Representative Publications

View All Publications
1 December 2019 in Measurement
Verified
Journal Article
Closed access
DWDong Wang
Dong Wang
JKJin-Zhen Kong
Jin-Zhen Kong
YZYang Zhao
Yang Zhao
KTKwok-Leung Tsui
Kwok-Leung Tsui

16 November 2021 in IEEE Transactions on Industrial Electronics
Verified
Journal Article
Closed access
JKJin-Zhen Kong
Jin-Zhen Kong
DWDong Wang
Dong Wang
TYTongtong Yan
Tongtong Yan
JZJingzhe Zhu
Jingzhe Zhu
XZXi Zhang
Xi Zhang
Accurate remaining useful life (RUL) of batteries plays an imperative role in ensuring safe operations and avoiding catastrophic accidents. However, in practice, complicated working conditions bring challenges to accurate battery prognostics. In this article, an accelerated stress factors-based nonlinear Wiener process model is proposed to enrich inadequate battery prognostic works at various operating conditions. To realize online individual battery prognostics, once a new measurement is available, the parameters of a state-space model constructed by the proposed model are posteriorly updated. Then, based on the Peukert law and the Arrhenius equation, two specific accelerated stress-relevant drift functions and their associated degradation models at different discharge rates and temperatures are respectively designed. Subsequently, RUL predictions are conducted using the proposed method. RUL predictions at different discharge rates and different temperatures demonstrate the accuracy and robustness of the proposed prognostic models. According to some general prognostic metrics, the proposed method is proved to be superior to four existing RUL prediction approaches.
1

1 February 2020 in Measurement
Verified
Journal Article
Closed access
DWDong Wang
Dong Wang
JKJin-Zhen Kong
Jin-Zhen Kong
FYFangfang Yang
Fangfang Yang
YZYang Zhao
Yang Zhao
KTKwok-Leung Tsui
Kwok-Leung Tsui

12 October 2022 in IEEE Transactions on Industrial Electronics
Verified
Journal Article
Closed access
JKJin-Zhen Kong
Jin-Zhen Kong
DCDi Cui
Di Cui
BHBingchang Hou
Bingchang Hou
XLXingchen Liu
Xingchen Liu
DWDong Wang
Dong Wang
Battery health prognostic plays a crucial role in the safe operation of electrical devices. Battery degradation modeling is a crucial step for battery health prognostics. Currently, since most existing battery prognostic models only consider entire long-term degradation trends during modeling, battery degradation modeling is still not precise and lacks short-term deviations to sufficiently reflect strong complexity and volatility of degradation processes. To solve this problem, a new battery prognostic method is proposed by simultaneously considering long-term degradation tendency and short-term volatility. Because Wiener process and Ornstein–Uhlenbeck process have ability to respectively model long-term degradation tendency and short-term volatility, a short-long-term degradation model is accordingly proposed by linearly fusing two stochastic processes. Additionally, the proposed model can characterize modeling uncertainty and observation uncertainty. Besides, this article proves that the proposed model follows a log-normal distribution, which provides solid foundations for state-space modeling. Here, the purpose of formulating the proposed model to a state-space model with multiple cross-correlation hidden state variables is to better describe individual battery degradation using online battery capacity data. Health prognostics can be easily realized by extrapolating the updated model to a failure threshold. Results show that our proposed method performs well than contrast methods for battery health prognostics.

26 April 2021 in IEEE Transactions on Instrumentation and Measurement
Verified
Journal Article
Closed access
TYTongtong Yan
Tongtong Yan
DWDong Wang
Dong Wang
JKJin-Zhen Kong
Jin-Zhen Kong
Tangbin XiaTangbin Xia
Tangbin Xia
ZPZhiKe Peng
ZhiKe Peng
LXLifeng Xi
Lifeng Xi
Machine performance degradation assessment (MPDA) aims to use health indicators (HIs) to first detect an incipient failure time (IFT) and then monotonically track machine degradation trajectory. The IFT can indicate the time when a machine is entering a defective stage and it is also termed as first predicting time (FPT). Moreover, the benefit of monotonic degradation trajectories of HIs in a defective stage can well describe irreversible machine degradation mechanism, which in turn is beneficial to the simplification of prognostic models. In the literature, most existing HI construction methodologies only focus on one aspect of the MPDA, namely either PFT detection or degradation tracking. In this article, in view of spectrum amplitude fusion, a new methodology that aims to enhance HIs of MPDA is proposed. Considering intuitive machine degradation characteristics, four essential properties of HIs for FPT detection and subsequently monotonical degradation assessment are put forward and they are integrated into a uniquely hybrid property coined as HI-signal-to-noise ratio (HI-SNR) in this article. After the HI-SNR of HIs is maximized, spectrum fusion weights can be analytically derived. Finally, the sum of weighted spectrum amplitudes (SOWSA) is used as a generalized HI for MPDA. To demonstrate the effectiveness of the proposed new methodology, several experimental bearing and gear degradation cases are studied to show that the proposed SOWSA can detect FPT and characterize monotonic gear and bearing deterioration processes in a unified manner.

17 April 2025 in IEEE Sensors Journal
Verified
Journal Article
Closed access
QHQidi Hu
Qidi Hu
BSBinyi Su
Binyi Su
JKJinzhen Kong
Jinzhen Kong
LWLiu WeiPeng
Liu WeiPeng
The prediction of the Remaining Useful Life (RUL) of stochastically degrading devices is of critical importance for Prognostics and Health Management (PHM). However, most of the existing approaches overlook the inherent coupling of trends and fluctuations in the degradation process. Meanwhile, these methods are struggling to effectively model the random fluctuations component in the degradation process and may lead to inaccurate prediction. To address these limitations, we propose a modified RUL prediction framework WTSAGNN. The framework extends the spatiotemporal dependencies of multi-sensor signals into the frequency domain. By leveraging the proposed adaptive orthogonal wavelet convolution to more effectively decouple the trend and fluctuation components of the degradation process. Furthermore, we introduce a stochastic differential network (SDN) to replace the traditional multilayer perceptron (MLP). The proposed method enhances the robustness of neural networks against stochastic variations in degradation processes through explicit stochastic modeling. Experimental results demonstrate that WTSAGNN outperforms existing state-of-the-art(SOTA) methods on benchmark datasets, achieving up to a 9.1% improvement in prediction accuracy with only a limited increase in computational overhead. In addition, the proposed modules are plug-and-play, allowing seamless integration with existing data-driven prediction frameworks.