Abstract
It is very important to implement the fault diagnosis technology in industrial processes to make the process more reliable. In this paper, an improved particle filter (PF) method based on a modified beetle swarm antennae search (BSAS) algorithm is proposed and verified in a doubly fed induction generator (DFIG) fault diagnosis application. Firstly, the search strategy of BSAS is improved to ensure its global search ability. Secondly, it is introduced to the traditional PF algorithm to improve the particle diversity and impoverishment drawbacks. Finally, the fault diagnosis algorithm is verified by combining the DFIG state space model. The simulation experimental of fault detection and isolation results show that the proposed method is simple and effective, and it can effectively monitor the occurrence of faults. For the fault diagnostic application, the method proposed in this paper could be implemented in other model based processes, including chemical process, biochemical wastewater treatment process, etc.
Highlights
Wind energy, a clean and renewable energy, is the fastest-growing energy source in the world in the last decades, which is considered as an effective way to face and solve the world energy problem [1]. 2018 is a good year for wind energy with 51.3 GW newly added capacity worldwide, including 46.8 GW of onshore wind power and 4.5 GW of offshore wind power
Particle filter (PF) has been proved that it is an effective method to do the state estimation works based on the state space system model [12,13,14,15], which uses a set of weighted particles to approximate the posterior probability density function, and implements recursive Bayesian estimation to estimate the system state by nonparametric Monte Carlo method [16]
Wind power systems are usually installed in remote areas with few people tread, it is necessary to develop a remote monitoring system and a reliable fault diagnosis algorithm to monitor important components and equipment in wind turbine systems
Summary
A clean and renewable energy, is the fastest-growing energy source in the world in the last decades, which is considered as an effective way to face and solve the world energy problem [1]. 2018 is a good year for wind energy with 51.3 GW newly added capacity worldwide, including 46.8 GW of onshore wind power and 4.5 GW of offshore wind power. Faults in the generator and drive train (includes gear box, main shafts, and bearings) are the most crucial and widely observed failures, which dominate over 60% of the downtime in wind turbine systems [4] in some power plants located in Sweden, Finland, and Germany during 1997-2005, and it is similar to the other countries in the world. Having a reliable fault diagnosis and fault tolerant control (FTC) scheme is crucial to improve the reliability of wind turbines and reduce expensive repair costs, especially for the offshore wind generators because they are not accessible. Scholars and researchers worldwide have done many researches and applications of the fault diagnostic methods for key parts of wind power generation system, such as gearbox, generator, and other components [5,6,7,8,9], based on the three main categories of fault diagnosis: i) signal-based approach, ii) knowledge-based approach and iii) model-based approach. Dynamic adaptive inertia weight is employed to improve the global search capability of the beetle swarm antennae search (BSAS) algorithm, which is used to optimize the resampling process of the typical particle filter algorithm in avoiding the particle degeneracy with the iterations going on
Sign-in/Register to view highlights & summary Sign-in/Register to access full text options 
Talk to us
Join us for a 30 min session where you can share your feedback and ask us any queries you have
Schedule a callDisclaimer: All third-party content on this website/platform is and will remain the property of their respective owners and is provided on "as is" basis without any warranties, express or implied. Use of third-party content does not indicate any affiliation, sponsorship with or endorsement by them. Any references to third-party content is to identify the corresponding services and shall be considered fair use under The CopyrightLaw.
