Xuzhi Li

Xuzhi Li

Auto-generated profile, based on publicly available information

Xuzhi Li

Key Laboratory of Space Utilization, Technology and Engineering Center for Space Utilization, Chinese Academy of Sciences, Beijing, China

Research Focus & Metrics

Academic Career Length

9Years (2018-Present)

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

Representative Publications

View All Publications
30 November 2023 in Journal of Hazardous Materials
Verified
Journal Article
Closed access
XHXinghua Huang
Xinghua Huang
XLXuzhi Li
Xuzhi Li
LZLiping Zheng
Liping Zheng
YZYa Zhang
Ya Zhang
LSLi Sun
Li Sun
GWGuoqing Wang
Guoqing Wang

30 December 2021 in IEEE Transactions on Instrumentation and Measurement
Verified
Journal Article
Closed access
HLHaoran Liang
Haoran Liang
Lei SongLei Song
Lei Song
JDJun-Rong Du
Jun-Rong Du
XLXuzhi Li
Xuzhi Li
LGLili Guo
Lili Guo
Multivariate time series is widely derived from industrial facilities, such as power plants, manufacturing machines, spacecraft, digital devices etc., and anomaly detection and location is of great importance to industrial preventive maintenance. However, anomalies in multivariate time series always result from their unusual change of temporal or correlative features, and it is challenging to capture these complex characteristics. Besides, achieving consistent anomaly detection and location performance is also a tricky issue. In this article, a novel anomaly detection and location framework that combines generative adversarial networks and autoencoder is proposed to capture time dependent and correlation features of multivariate time series with on need of anomalous sequences. Firstly, multi-time scale correlation computation is utilized to encode multivariate time series into multiple cross-correlation graphs, which can be fed into the proposed deep architecture for extracting more distinguishable features. On this basis, a robust cost function with multiple loss issues is designed, and reconstruction matrix deviation from original space of encoder-encoder structure is utilized to detect and locate abnormal time series, ensuring the consistency of detection and location tasks and the framework reliability. Extensive experiments on five industrial datasets are conducted to indicate our model is a generic and excellent framework for anomaly detection and location of multivariate time series.

15 August 2018 in Journal of Ecology
Verified
Journal Article
Closed access
PDF file
SVStavros D Veresoglou
Stavros D Veresoglou
LLLei Liu
Lei Liu
TXTianle Xu
Tianle Xu
MRMatthias C. Rillig
Matthias C. Rillig
MWMeie Wang
Meie Wang
Baodong ChenBaodong Chen
Baodong Chen

22 November 2024
Verified
Conference Paper
Open access
JZJian Zhang
Jian Zhang
LGLili Guo
Lili Guo
Lei SongLei Song
Lei Song
SGSong Gao
Song Gao
CHChuanzhu Hao
Chuanzhu Hao
XLXuzhi Li
Xuzhi Li

24 September 2026 in SSRN Electronic Journal
Verified
Preprint
Open access
LLLyurong Liu
Lyurong Liu
SLShuang Liu
Shuang Liu
QWQiaoying Wu
Qiaoying Wu
NZNa Zheng
Na Zheng
XZXiaolei Zhang
Xiaolei Zhang
YLYusong Liu
Yusong Liu
Background: Long-term maintenance of hemodialysis machines requires quantitative indicators of equipment use, recurrent failures, and service demand. This study evaluated whether routinely recorded maintenance data can characterize device-level maintenance burden in a long-serving hemodialysis fleet.Methods: Maintenance records from January 2022 to December 2025 were reviewed for 35 Fresenius hemodialysis machines. The primary analysis included 23 Fresenius 4008S Basic units and 186 corrective failure events. Maintenance burden was characterized using recurrent-event metrics, documented cumulative operating hours, documented maintenance time, fault manifestations, and component replacement demand. The mean cumulative function (MCF) quantified cumulative recurrence, and a power-law non-homogeneous Poisson process was fitted as a secondary temporal model. Associations of documented operating hours and calendar age with event burden were assessed using Spearman correlation.Results: Hydraulic/Valve failures accounted for 154 events (82.8%) and 155.5 of 187.5 h of documented maintenance time(82.9%). The MCF reached 8.087 events per machine at the end of 2025 (95% CI, 6.609–9.609). The power-law shape parameter was β=0.962 (95% CI, 0.816–1.148), providing no evidence of sustained fleet-wide acceleration. Documented cumulative operating hours were associated with four-year event burden (rₛ=0.727, p<0.001), whereas calendar age was not. Eight machines accounted for 68.8% of 2025 corrective events.Conclusions: Recurrent-event and operating-hour metrics provided complementary quantitative information on maintenance burden. The findings support the use of documented operating history and recurrent maintenance burden as complementary information for maintenance assessment, while highlighting the need for prospective exposure and condition-monitoring data before predictive thresholds can be established.