Research On Diagnosis Method Of Bearing Fault In Shearer Cutting Department

The shearer plays an important role in the production of coal mines.Long-term operation in harsh environments causes coal mining machines to fail.In order to ensure the normal working schedule of the shearer,improve the accuracy and stability of the shearer fault diagnosis,and reduce the loss caused by the time of the shearer fault diagnosis,the paper takes the shearer cutting section bearing as an example and puts forward Research on the diagnosis method of bearing faults in the cutting section.The research on the fault diagnosis method of the cutting part bearing is mainly carried out in the following two aspects:(1)For the existing BP neural network based shearer fault diagnosis method,the noise sensitivity and generalization performance are weak.The paper proposes a BP_Adaboost integrated learning algorithm based on the shear fault diagnosis method of the shearer cutting section.Firstly,the Pearson correlation coefficient method is used to analyze the correlation of the data set.Then,the principal component of the data set is extracted by principal component analysis.Then,the BP neural network weak classifier is constructed.Finally,multiple BP weak classifiers are combined by the combination strategy for bearing fault diagnosis.Group BP Adaboost strong classifier.The experimental results show that the BP_Adaboost integrated learning algorithm solves the problem that the BP neural network based shearer fault diagnosis algorithm has noise sensitivity and low generalization performance,and improves the accuracy of equipment fault diagnosis.(2)For the traditional logistic regression model,the nonlinear relationship problem cannot be solved.The paper proposes a nonlinear logistic regression for the diagnosis of bearing faults in the cutting section.Firstly,the Newton iteration method is used to evaluate and optimize the parameters of the nonlinear logistic regression model.Then,the parameters after optimization are constructed into a fault diagnosis model for the diagnosis of bearing faults in the cutting faults in the cutting section.Finally,the test data is used to the equipment.The fault diagnosis model is tested.The experimental results show that the nonlinear logistic regression solves the problem that the traditional logistic regression can not solve the nonlinear relationship.Compared with the traditional logistic regression model,the nonlinear logistic regression improves the accuracy of bearing fault diagnosis and meets the actual requirements of equipment fault diagnosis.