Method Of Anomaly Detection Of Temperature Data In Vacuum Thermal Test Based On Data Mining

In order to improve the degree of automation for the anomaly detection of Temperature Data in Vacuum Thermal Test, based on the similarities of the Temperature Data, a kind of Similarity Metric Algorithm named DTW-SBD is carried out and the optimal ranges of parameters are obtained. The Average Accuracy Rate of this algorithm is above 0.9. Based on the preceding algorithm, an Anomaly Detection method of Temperature Data is proposed, which can accurately identify four kinds of typical abnormal modes in Vacuum Thermal Test. This method achieves the similar effect as manual detection.