- Smart Grid Security and Resilience
- Distributed Control Multi-Agent Systems
- Machine Fault Diagnosis Techniques
- Fault Detection and Control Systems
- Stability and Control of Uncertain Systems
- Security in Wireless Sensor Networks
- Industrial Vision Systems and Defect Detection
- Neural Networks Stability and Synchronization
- Mathematical and Theoretical Epidemiology and Ecology Models
- Network Security and Intrusion Detection
- Anomaly Detection Techniques and Applications
- Face and Expression Recognition
- Structural Health Monitoring Techniques
- Engineering Diagnostics and Reliability
- Advanced Memory and Neural Computing
- Network Time Synchronization Technologies
- Fuzzy Logic and Control Systems
- Energy Load and Power Forecasting
- Non-Destructive Testing Techniques
- Stability and Controllability of Differential Equations
- Neural Networks and Applications
- Infrastructure Resilience and Vulnerability Analysis
- Welding Techniques and Residual Stresses
- Adversarial Robustness in Machine Learning
Northeastern University
2019-2024
State Key Laboratory of Synthetical Automation for Process Industries
2021
This article studies the consensus problem of multiagent systems (MASs) under denial-of-service (DoS) attacks and switching topology. Compared with previous papers, topology, impact DoS on MAS is considered, thus, difficulty protocol design further increased. First, a piecewise resilient proposed to ensure that can achieve topology attacks. Then, by considering different convergence rates caused in each sleep period attacks, Lyapunov–Krasovskii function established. The sufficient conditions...
Condition monitoring (CM) has been playing an important role in the operation and maintenance of industrial equipment. However, changeable environments where some equipment, such as wind turbines (WTs) are installed, have put a negative effect on CM. To deal with this problem, article proposes CM method WTs based assembled multidimensional membership functions (MFs) using fuzzy inference system. First, MFs, which formed by fusing environment factor into conventional proposed to reduce from...
Abstract This article is concerned with the issue of observer‐based event‐triggered control for networked switched systems (NSSs) under hybrid attacks. Since attacks can cause more damage to system than a single cyber attack, this considers situation where stochastic deception and aperiodic denial‐of‐service (DoS) coexist in NSSs. To address issue, dynamics initial combined mechanism (ETM), observers are formulated as new augmented system. Thereinto, unmeasurable states estimated by newly...
Abstract This paper studies event‐triggered containment control problem of multi‐agent systems (MASs) under deception attacks and denial‐of‐service (DoS) attacks. First, to save limited network resources, an mechanism is proposed for MASs hybrid cyber Different from the existing mechanisms, negative influences DoS are considered in triggering function. The communication frequencies between agents reduced. Then, based on mechanism, a corresponding protocol ensure that followers will converge...
Summary This article focuses on the issue of event‐triggered fault detection filters (FDF) design for networked switched systems. Compared with existing literature, this investigates situation that all subsystems are unstable, and considers stochastic cyber attacks will disrupt detection. An mechanism (ETM) excludes Zeno behavior naturally is designed to economize limited network resources. Then considering ETM‐induced delay term, dynamics initial system combined ETM, filters, effects...
In industrial fields, Nondestructive Testing (NDT) has become an important method to test the quality of welds. For low-contrast pipe weld defect x-ray image, traditional detection low precision. this paper, automatic for defects based on a triplet deep neural network is proposed. First, original X-ray image changed into relief so that feature more obvious. Second, vector obtained by mapping through network. The makes similar vectors are closer, and distances different farther. It first time...
With the rapid development of machine learning, model interpretability has become a major topic in recent studies, especially field measurement. However, there always been dilemma interpreting learning models: those intrinsically interpretable methods cannot achieve desirable results complex applications, whereas with good performances are not interpretable. To resolve this dilemma, article proposes an method based on fuzzy inference system (FIS), capable handling complicated applications....
Fuzzy logic system is commonly used in wind turbine fault detection. However, traditional fuzzy systems are built through human experience. The constructed this way will have inaccurate problems. In order to solve problem, paper proposes a novel (FLS) based on neural network construction method improve accuracy rate of First, proposed. Using method, membership function can be more accurately. Then, FLS the extended data-driven When environment changes, such detection with function. Finally,...
For a cyber physical system under multiple attacks, including non-periodic denial-of-service (DoS) attack and stochastic deception attack, we design dynamic output feedback controller with event-triggered strategy. We adopts control strategy based on trigger conditions, which reduces the number of triggers saves network resources. Besides, establish switched model to describe presence attacks scheme. Then, according asymptotic stability theory, ensuring switching stable is designed by using...