Authorization

 

Journal

Volume 22, №3, Special Issue on the Artificial Intelligence; Where Do We Stand?

A multi-stage, multi-strategy-improved marine predator algorithm and its application in wind turbine fault detection (pp.372-408)
Mingzhu Tang1, Jiabiao Yi1, Wen Long2,*, Fei Feng3, Weiqiang Chen3, Liang Liu1
https://doi.org/10.30546/1683-6154.22.3.2023.372


1School of Energy and Power Engineering, Changsha University of Science & Technology, Changsha, China, e-mail: tmz@csust.edu.cn, 20106011037@stu.csust.edu.cn, liuliang_hn@126.com  

2School of Mathematics and Statistics, Guizhou University of Finance and Economics, Guiyang, China

3Tongdaoziwu Wind Energy Development Limited Company, Hunan, Huaihua, P.R. China, e-mail: cggc_fengfei@163.com, pfdsjdcc@163.com

∗Corresponding author’s e-mail: lw227@mail.gufe.edu.cn

 

 

Abstract. This paper proposes a multi-stage, multi-strategy improved marine predator algorithm based on the basic marine predator algorithm, which introduces the oscillating adaptive control step factor, the drunkard walking strategy, and the elite population leadership strategy. Finally, a combined model is established by integrating the proposed algorithm with the traditional classification algorithm. Comparative experimental results on 23 benchmark functions show that the proposed algorithm has a more accurate and faster merit-seeking capability. In addition, the application results on a real wind turbine fault detection (WTFD) field indicate that the proposed algorithm can obtain lower data dimensionality and improve the accuracy of the traditional classification algorithm in the the WTFD field significantly in terms of both feature selection and parameter optimization.

 

Keywords: Marine Predator Algorithm, Fault Detection, Parameter Optimization, Feature Selection.

 

AMS Subject Classification: 65K10.

  



COIA 2027

7-9 July 2027
Baku, Azerbaijan

Search