A5101909402- Advanced Control Systems Optimization
- Fault Detection and Control Systems
- Control Systems and Identification
- Process Optimization and Integration
- Gene Regulatory Network Analysis
- Model Reduction and Neural Networks
- Complex Network Analysis Techniques
- Distributed Control Multi-Agent Systems
- Iterative Learning Control Systems
- Microbial Metabolic Engineering and Bioproduction
- Advanced Optimization Algorithms Research
- Protein Degradation and Inhibitors
- Advanced Battery Materials and Technologies
- Target Tracking and Data Fusion in Sensor Networks
- Neural Networks and Applications
- Neural dynamics and brain function
- Guidance and Control Systems
- Fuel Cells and Related Materials
- Optimization and Variational Analysis
- Metabolomics and Mass Spectrometry Studies
- Computational Drug Discovery Methods
- Neural Networks Stability and Synchronization
- Advanced Battery Technologies Research
- Adaptive Control of Nonlinear Systems
- Machine Learning in Materials Science
North Carolina State University
2023-2025
The University of Sydney
2023-2025
University of Hong Kong
2024
Beijing Information Science & Technology University
2023-2024
China Southern Power Grid (China)
2024
Hong Kong University of Science and Technology
2024
Dalian University
2024
Dalian University of Technology
2024
Sichuan University
2023-2024
State Key Laboratory of Biotherapy
2023
Research on trajectory tracking is crucial for the development of autonomous vehicles. This paper presents a scheme by utilizing model predictive control (MPC) and preview-follower theory (PFT), which includes reference generation module MPC controller. The could calculate lateral acceleration at preview point PFT to update state variables, generate yaw rate in each prediction point. Since range increased, makes calculation more accurate. Through physical constraints, controller can achieve...
Process systems are characterized by nonlinearity, uncertainty, large scales, and also the need of pursuing both safety economic optimality in operations. As a result they difficult to control effectively. Data-driven techniques such as machine learning algorithms can provide complementary tools insights classical model-based enhancing capability modeling dynamics complex maintenance performance. Moreover, behavior plants controllers black boxes, data-driven enable completely model-free...
To test the biomechanical properties of 3D printed tantalum and titanium porous scaffolds.Four types scaffolds with four alternative pore diameters, #1 (1000-700 μm), #2 (700-1000 #3 (500-800 #4 (800-500 were molded by selective laser melting technique, tested scanning electronic microscope, uniaxial-compression tests, Young's modulus tests; they compared same size pig femoral bone scaffolds.Under equivalent stress scaffold was 411 ± 1.43 MPa, which significantly larger than (P < 0.05). 2.61...
An information-theoretic framework for constructing data injection attacks on process systems, from the attacker's standpoint, is studied. The attack construction aims to distract stationary distributions of variables and stay stealthy, simultaneously. problem formulated as designing a multivariate Gaussian distribution maximize Kullback-Leibler divergence between states state estimates under without attacks, while minimizing that sensor measurements. When attacker has limited access...
To identify hub metabolic biomarkers that constructively shape the type 2 diabetes mellitus (T2DM) risk network. We analysed data from 98 831 UK Biobank participants, confirming T2DM diagnoses via medical records and International Classification of Diseases codes. Totally 168 circulating metabolites were quantified by nuclear magnetic resonance at baseline. Metabolome-wide association studies with Cox proportional hazards models performed to statistically significant metabolites. Network...
Abstract In this article, we study the resilience of process systems in an information‐theoretic framework , from perspective attacker capable optimally constructing data injection attacks. The attack aims to distract stationary distributions variables and stay stealthy, simultaneously. problem is formulated as designing a multivariate Gaussian distribution maximize Kullback‐Leibler divergence between states state estimates under attacks without attacks, while minimizing that sensor...
Input‐output partitioning for decentralized control has been studied extensively using various methods, including those based on relative gains and degrees sensitivities. These two concepts are characterizations of long‐time short‐time input‐output response, respectively. A unifying new interaction measure, called time‐averaged gain, which characterizes the interactions during a time scale interest linear time‐invariant systems is proposed. This measure used as basis community detection in...
Abstract Ease of control complex networks has been assessed extensively in terms structural controllability and observability, minimum energy criteria. Here we adopt a sparsity-promoting feedback framework for undirected with Laplacian dynamics distinct topological features. The objective considered is to minimize the effect disturbance signals, magnitude signals cost channels. We show that depending on channels, different network structures become least expensive option control....
Alternating direction method of multipliers (ADMM), as a powerful distributed optimization algorithm, provides framework for model predictive control (MPC) nonlinear process systems based on local subsystem information. However, the practical application classical ADMM is largely limited by high computational cost caused its slow (linear) rate convergence and non-parallelizability. In this work, we combine recently developed multi-block parallel algorithm with Nesterov acceleration technique...
Abstract Decomposition‐based solution algorithms for optimization problems depend on the underlying latent block structure of problem. Methods detecting this are currently lacking. In article, we propose stochastic blockmodeling (SBM) as a systematic framework learning in generic problems. SBM is generative graph model which nodes belong to some blocks and interconnections among stochastically dependent their affiliations. Hence, through parametric statistical inference, interconnection...