试验以4种不同工况的OTLC振动信号作为试验数据进行分析，试验具体细节

The test uses four OTLC vibration signals of different working conditions as test data for analysis. The specific details of the test are shown in [29]. The 4 different working conditions are: normal contact, burnt contact, loose contact and contact shedding. 50 sets of data are collected under each working condition, each set of data is 7380 samples. Randomly draw 80% as training and 20% as test, and compare the method in this paper with the four algorithms of WOA-KNN, GOA-KNN, PSO-KNN and CBA-KNN. The average, maximum, minimum and standard deviation of the fitness of the model were used to evaluate the models involved in the experiment.

The test was analyzed with THE OTLC vibration signal in 4 different operating conditions as the test data, and the specific details of the test were shown in the literature. 4 different operating conditions are: normal contact, contact burned, contact loose and contact off, each under the condition of the collection of 50 sets of data, each set of data for 7380 samples. The method of this paper was compared with THE four algorithms of WOA-KNN, GOA-KNN, PSO-KNN and CBA-KNN, and CBA-KNN, which were selected at random as training and 20% as tests. The models involved in the experiment were evaluated by mean, maximum, minimum and standard deviation of the model's adaptation.

In the test, the OTLC vibration signals of four different working conditions are used as test data for analysis. The specific details of the test are shown in reference [29]. Four different working conditions are: contact normal, contact burning, contact loosening and contact falling off. Under each working condition, 50 groups of data are collected, each group of data is 7380 samples. 80% of them are randomly selected as training and 20% of them are tested. This method is compared with woa-knn, goa-knn, pso-knn and cba-knn. The average, maximum, minimum and standard deviation of the fitness of the model were used to evaluate the model involved in the experiment.