2020 | vol. 68 | nr. 2 | art. 10

Fuzzy LSSVC-WKNN Combination Algorithm in Fault Diagnosis

Sheng Sun, Chuiwei Lu, Guojun Zhang
Abstract
In this paper, we propose a fuzzy least squares support vector clustering-weighted k-nearest neighbour (LSSVC-WKNN) combination algorithm. We use the ATS method to optimize feature vectors and adopt this algorithm for fault diagnosis. Firstly, the fault diagnosis model is described, and then the feature vector is reduced by ATS method. The fuzzy LSSVC algorithm is adopted to train the optimized feature vector of training samples, so as to obtain a trained clustering model. Then, the fuzzy LSSVC-WKNN combination algorithm is used for fault diagnosis of test samples. Taking the simulated circuit fault diagnosis test as an example, the experimental results prove that the method in this paper has the advantage of higher diagnostic accuracy than other methods, and it is a universal and feasible online fault diagnosis method.
Keywords: least squares, support vector, clustering, k-nearest neighbour, fault diagnosis
To cite this article: Sun Sheng, Lu Chuiwei, Zhang Guojun, “Fuzzy LSSVC-WKNN combination algorithm in fault diagnosis” in Electrotehnica, Electronica, Automatica (EEA), 2020, vol. 68, no. 2, pp. 93-101, ISSN 1582-5175.

 

Download