Adaptive Dynamic Thresholds for Unsupervised Joint Anomaly Detection and Trend Prediction

Ding, Fenglin; Zhao, Yilin; Li, Zongliang; Tang, Haibin; Liu, Yizhuo; Guo, Danhuai

doi:10.3390/s26010257

This is an early access version, the complete PDF, HTML, and XML versions will be available soon.

Open AccessArticle

Adaptive Dynamic Thresholds for Unsupervised Joint Anomaly Detection and Trend Prediction

by

Fenglin Ding

^1,2,

Yilin Zhao

³,

Zongliang Li

¹,

Haibin Tang

²,

Yizhuo Liu

¹ and

Danhuai Guo

^3,*

¹

Beijing Institute of Control Engineering, Beijing 100190, China

²

School of Astronautics, Beihang University, Beijing 100191, China

³

College of Information Science and Technology, Beijing University of Chemical Technology, Beijing 100013, China

^*

Author to whom correspondence should be addressed.

Sensors 2026, 26(1), 257; https://doi.org/10.3390/s26010257

Submission received: 19 November 2025 / Revised: 18 December 2025 / Accepted: 30 December 2025 / Published: 31 December 2025

(This article belongs to the Section Fault Diagnosis & Sensors)

Download Versions Notes

Abstract

Anomaly detection and degradation trend prediction are two pivotal tasks in system health management. However, most existing approaches treat them as independent problems and fail to exploit their intrinsic interdependence. In addition, the scarcity of labeled data in real-world scenarios limits the applicability of supervised learning methods. To address these challenges, we propose an adaptive thresholding strategy framework for unsupervised joint anomaly detection and trend prediction. Our framework introduces a self-adaptive threshold strategy from historical data distributions and dynamically updates them in response to evolving system behavior. The anomaly detection results are integrated to enhance degradation trend forecasting, while the predicted degradation trends, in turn, refine the anomaly thresholds through a feedback mechanism. Experiments on both public and real-world industrial datasets demonstrate that the proposed framework achieves superior detection accuracy, robust trend prediction, and high computational efficiency under diverse operational conditions.

Keywords: anomaly detection; degradation prediction; adaptive threshold; prognostics health management (PHM); time series

Share and Cite

MDPI and ACS Style

Ding, F.; Zhao, Y.; Li, Z.; Tang, H.; Liu, Y.; Guo, D. Adaptive Dynamic Thresholds for Unsupervised Joint Anomaly Detection and Trend Prediction. Sensors 2026, 26, 257. https://doi.org/10.3390/s26010257

AMA Style

Ding F, Zhao Y, Li Z, Tang H, Liu Y, Guo D. Adaptive Dynamic Thresholds for Unsupervised Joint Anomaly Detection and Trend Prediction. Sensors. 2026; 26(1):257. https://doi.org/10.3390/s26010257

Chicago/Turabian Style

Ding, Fenglin, Yilin Zhao, Zongliang Li, Haibin Tang, Yizhuo Liu, and Danhuai Guo. 2026. "Adaptive Dynamic Thresholds for Unsupervised Joint Anomaly Detection and Trend Prediction" Sensors 26, no. 1: 257. https://doi.org/10.3390/s26010257

APA Style

Ding, F., Zhao, Y., Li, Z., Tang, H., Liu, Y., & Guo, D. (2026). Adaptive Dynamic Thresholds for Unsupervised Joint Anomaly Detection and Trend Prediction. Sensors, 26(1), 257. https://doi.org/10.3390/s26010257

Note that from the first issue of 2016, this journal uses article numbers instead of page numbers. See further details here.

Article Metrics

Article metric data becomes available approximately 24 hours after publication online.

Article Menu

Adaptive Dynamic Thresholds for Unsupervised Joint Anomaly Detection and Trend Prediction

Abstract

Share and Cite

Article Metrics

Further Information

Guidelines

MDPI Initiatives

Follow MDPI