Research on Automatic Fault Diagnosis System of Coal Mine Drilling Rigs based on Drilling Parameters

Aiming at the problem that the underground fully hydraulic drilling rigs mainly relies on manual experience to diagnose faults, which requires high professional skills of the operators and can not accurately and reasonably judge the fault types, this paper puts forward a method to identify faults by using the sensitive drilling parameters in the drilling rigs construction. Firstly, principal component analysis(PCA) algorithm is used to extract features from drilling sensitive data sets, which eliminates the correlation between drilling data sets. Then, the self-organizing maps(SOM) neural network clustering algorithm is performed on the trained samples, and the cluster centers are obtained. By calculating the Euclidean distance between each sample in the test sample set and the cluster center in the high-dimensional feature space, the Bayesian discriminant method is used to judge the test samples, and the final fault diagnosis result is obtained. The diagnostic accuracy rate is up to 90%. This method provides a new and reliable fault warning method for the fully hydraulic drilling rigs in coal mine.