Impr ovement of the method of automation of monitoring of gas pipeline wear using vibration data

The features of monitoring the wear of gas pipeline surface and the steering rod of the equipment are considered. An easy-to-use algorithm for automatic detection of damage to the main pipeline based on vibration data is proposed. The use of the modal analysis method, data cleaning procedure, damage detection algorithm based on the Mahalanobis distance squared method, and construction of logistic curves allow for automation of the process of steering rod monitoring. It is shown that with an increase in the damage area and an increase in the modal attenuation coefficient from 5/8L to 8/10L, the proportion of true positive results is 100%, and the predictive ability of the model increases with the use of a moving average of the modal attenuation.