Turkish Journal of Electrical Engineering and Computer Sciences
DOI
10.55730/1300-0632.3804
Abstract
Dynamic characteristics of the doubly-fed induction generator (DFIG)-based wind power generation (WPGS) are fully nonlinear. Therefore, issues such as stability and achieving high efficiency, especially under harmonics behavior, are challenges that assess the control strategy reliability to find the perfect dynamic solution. This discussion offers a control strategy for the separated stator-port power using a predictive sliding mode strategy with a resonant function (PSMC-R) based on a deep recurrent neural network (DRNN). DRNN is formed as a low-order Taylor series formula. PSMC-R predicts the perfect switching surface path and regulates the distorted nonlinear DFIG with several dynamic aims. This approach reduces excessive chatter while violating the sliding surface path range of the classical SMC switch-part. Also, PSMC-R handled the fundamental and 5th-/7th-type harmonic wave at the positive synchronous +dqreference level of the machine dynamic quantities without further dissociation computations of the components. Dynamic results of a 1.5 MW DFIG-WPGS are simulated by using Matlab-package and presented good dynamic characteristics, less pulsation ratio of variables, and optimal sliding chatter of PSMC-R during various operating scenarios compared to the other classical regulation approaches.
Keywords
Double fed induction machine, wind engine unit, sliding surface controller, harmonic wave, regulation of power system
First Page
659
Last Page
677
Recommended Citation
BUSATI, OMAR and LIU, XIANGJIE
(2022)
"Predictive optimization of sliding mode control using recurrent neural paradigmfor nonlinear DFIG-WPGS during distorted voltage,"
Turkish Journal of Electrical Engineering and Computer Sciences: Vol. 30:
No.
3, Article 13.
https://doi.org/10.55730/1300-0632.3804
Available at:
https://journals.tubitak.gov.tr/elektrik/vol30/iss3/13
Included in
Computer Engineering Commons, Computer Sciences Commons, Electrical and Computer Engineering Commons