Analysis Model of Drilling Tool Failure Based on PSO-SVM and Its Application

Accurate drilling tool failure diagnosis is an important issue. This paper proposes a new forecasting method, using the support vector machine (SVM) to forecast drilling tool failure. First, select several major factors that affecting drilling tool failure as input features of SVM, then SVM's nuclear parameters are optimized with particle swarm optimization (PSO) in order to enhance its accuracy. This method takes full advantages of special advantages of SVM in treating small sample classification study problems, and the overall parallel search of PSO. Compared with actual engineering results, it is proved to have high performance and accuracy, which provides a new method to forecast drilling tool failure.