Use this URL to cite or link to this record in EThOS:
Title: Fault estimation and fault tolerant control with application to wind turbine systems
Author: Liu, Xiaoxu
ISNI:       0000 0004 7430 0559
Awarding Body: Northumbria University
Current Institution: Northumbria University
Date of Award: 2017
Availability of Full Text:
Access from EThOS:
Full text unavailable from EThOS. Please try the link below.
Access from Institution:
In response to the high demand of the operation reliability by implementing real-time monitoring and system health management, the three-year PhD project focuses on developing robust fault diagnosis and fault tolerant control strategies for complex systems with high-nonlinearities, stochastic Brownian perturbations, and partially decoupled unknown inputs, which are then applied to wind turbine energy systems. Integration of serval advanced approaches, including the augmented system method, unknown input observer design, Takagi-Sugeno fuzzy logic, linear matrix inequality optimization, and signal compensation techniques enable us to achieve robust estimations of both the system states and the faults concerned simultaneously, while removing/reducing the adverse influences from faults to the system dynamics. Prior to the existing work, the considered unknown inputs can be partially decoupled rather than completely decoupled, which can meet a wider practical requirement. Moreover, the systems under investigation can be linear, Lipschitz nonlinear, quadratic inner-bounded nonlinear, high-nonlinear characterized by a Takagi-Sugeno fuzzy model, and stochastic with Brownian perturbations, which can cover a wide range of real industrial plants. Specifically, the augmented system method is used to construct an augmented plant with the concerned faults and system states being the augmented states. Unknown input observer technique is thus utilized to estimate the augmented states and decouple unknown inputs that can be decoupled. Linear matrix inequality approach is further addressed to ensure the stability of the estimation error dynamics and attenuate the influences from the unknown inputs that cannot be decoupled. As a result, the robust estimates of the faults concerned and system states can be obtained simultaneously. Based on the fault estimates, a signal compensation scheme is developed to remove/offset the effects of the faults to the system dynamics and outputs, leading to a stable dynamic satisfying the expected performance. A case study on a 4.8 MW wind turbine benchmark system is proposed to illustrate and demonstrate the proposed integrated fault tolerant control techniques. Takagi-Sugeno modelling of a wind turbine system is presented as a by-product. To summarize, the proposed integrated fault estimation and fault tolerant control strategy can handle a system with highly nonlinear dynamics in a strong disturbance/noise environment (e.g., partially-decoupled process disturbances and stochastic parameter perturbations), which is validated by a real-time wind turbine system. As a result, the presented methods/algorithms have enriched fault diagnosis and tolerant control theory with high-novelty and great potentials for practical applications.
Supervisor: Gao, Zhiwei Sponsor: Not available
Qualification Name: Thesis (Ph.D.) Qualification Level: Doctoral
EThOS ID:  DOI: Not available
Keywords: H300 Mechanical Engineering ; H600 Electronic and Electrical Engineering