Showing papers by "Zhiwei Gao published in 2020"

Actuator and Sensor Fault Classification for Wind Turbine Systems Based on Fast Fourier Transform and Uncorrelated Multi-Linear Principal Component Analysis Techniques

Abstract: In response to the high demand of the operation reliability and predictive maintenance, health monitoring and fault diagnosis and classification have been paramount for complex industrial systems (e.g., wind turbine energy systems). In this study, data-driven fault diagnosis and fault classification strategies are addressed for wind turbine energy systems under various faulty scenarios. A novel algorithm is addressed by integrating fast Fourier transform and uncorrelated multi-linear principal component analysis techniques in order to achieve effective three-dimensional space visualization for fault diagnosis and classification under a variety of actuator and sensor faulty scenarios in 4.8 MW wind turbine benchmark systems. Moreover, comparison studies are implemented by using multi-linear principal component analysis with and without fast Fourier transform, and uncorrelated multi-linear principal component analysis with and without fast Fourier transformation data pre-processing, respectively. The effectiveness of the proposed algorithm is demonstrated and validated via the wind turbine benchmark.

Fault Tolerant Control Using Reinforcement Learning and Particle Swarm Optimization

Abstract: Diversity, uncertainty and suddenness of unexpected faults bring a challenge for fault tolerant control due to the lack of valid data especially for a fault during an early stage. In this study, a reinforcement learning approach with a critic action architecture is proposed to overcome this challenge by designing an online learning fault-tolerant controller so that the faulty system can approximate the performance index of the fault-free system. Different from the traditional Hebb enhancement rules in the reinforcement learning, the training process is speeded up by introducing a supervisory learning on the basis of the training dataset which is built with the states and the virtual optimal control acquired by particle swarm optimization. The effectiveness of the algorithm is demonstrated by a test bed of a three-tank system.

Hyperplane design for discrete‐time sliding mode control with event‐trigger strategy and disturbance observer

Abstract: In this study, a simple hyperplane design method based on the discrete-time Riccati equation is proposed, and the inter-relations are investigated for the basic requirement on sliding surface selection, the assumption of stabilisability, the Riccati inequality and the standard Riccati equation. A state and disturbance observer is embedded in discrete-time sliding mode control (DSMC) to achieve its applicability when only partial system states can be measured. In order to improve the performance of the system in control updating times, a novel event-triggered DSMC with a state and disturbance observer is proposed. It is shown that the proposed method achieves a quasi sliding mode with a small boundary layer. Simulation examples are presented to show the effectiveness and advantages of the proposed design schemes.

Observer-Based Event-Triggered Predictive Control for Networked Control Systems under DoS Attacks.

Abstract: This paper studies the problem of DoS attack defense based on static observer-based event-triggered predictive control in networked control systems (NCSs). First, under the conditions of limited network bandwidth resources and the incomplete observability of the state of the system, we introduce the event-triggered function to provide a discrete event-triggered transmission scheme for the observer. Then, we analyze denial-of-service (DoS) attacks that occur on the network transmission channel. Using the above-mentioned event-triggered scheme, a novel class of predictive control algorithms is designed on the control node to proactively save network bandwidth and compensate for DoS attacks, which ensures the stability of NCSs. Meanwhile, a closed-loop system with an observer-based event-triggered predictive control scheme for analysis is created. Through linear matrix inequality (LMI) and the Lyapunov function method, the design of the controller, observer and event-triggered matrices is established, and the stability of the scheme is analyzed. The results show that the proposed solution can effectively compensate DoS attacks and save network bandwidth resources by combining event-triggered mechanisms. Finally, a smart grid simulation example is employed to verify the feasibility and effectiveness of the scheme's defense against DoS attacks.

Data-driven Sensor Fault Estimation for the Wind Turbine Systems

Abstract: As the need for early fault detection increases day by day in large industries, the importance of a reliable fault diagnosis becomes more obvious. Moreover, sensors in industrial systems are prone to faults or malfunctions due to aging or accidents. Motivated by the above, in this study, a neural network sensor fault diagnosis approach is proposed and the stability and convergence of the algorithm are proven by using the robust estimation theorem and input-to-state stability (ISS). The proposed algorithm is applied to a wind turbine benchmark with 4.8 MW rated power. 10% to 30% of the sensor performance reduction is considered to illustrate the effective performance of the addressed algorithm.

An Online Control Approach for Forging Machine Using Reinforcement Learning and Taboo Search

Abstract: It is noticed that offline-training and online-implementation method is dominant in the data-driven control. However, the inconsistence existing in offline data and online data may degrade the control performance. To address the aforementioned issue, an online control strategy is developed so that the control parameters can be updated online based on the real-time data measured to ensure satisfactory control performance in this study. Specifically, an online control algorithm is addressed to control the pressing-down speed of the forging machine based on the framework of the reinforcement learning that has a capability of building a complete mapping from state space to action space only according to the neighbour samples. Rather than using the way of trials and errors which is too slow to be online implementation, a taboo search is addressed to speed up the learning-working process by directly searching the control on the current states, followed by the stability conditions, derived from Lyapunov stability theory. A coarse model that is limited to get the cost information of the reinforcement learning is used to make the best of mechanism information, which prevents the occurrence of the invalid states that do not conform to system characteristics. The effectiveness of the algorithm is demonstrated by an ultra-low forging machine, which outperforms the conventional approaches such as PID and neural network control approaches. The proposed algorithm has advantages in parameter adjustments so that it is easier to implement in a practical system.