Internal Leakage Diagnosis of Natural Gas Station Valve Based on Acoustic Emission Parameters and Support Vector Machine
In order to ensure the safety of the natural gas station and yard system,the study on internal leakage diagnosis of gate valve in station and yard was carried out by acoustic emission technology.The valve internal leakage experiment was carried out to collect and analyze the acoustic emission signal data of valve internal leakage,and the wavelet packet threshold noise re-duction was used to process the valve internal leakage signal.In addition,feature extraction was carried out from the wave cha-racteristics,impact characteristics and energy contained in the signal,and the extracted six feature parameters were used as the input features of the subsequent SVM algorithm for valve internal leakage diagnosis.Lastly,K fold cross-validation method was used to find the optimal parameters c and g of the SVM algorithm model,and SVM valve internal leakage diagnosis model was es-tablished.The results show that the accurate identification rate of the gate value internal leakage of the established SVM model is 97.2%.
万方数据
维普
