A Fault Diagnosis Method for Key Equipment in Oil and Gas Stations Based on Decision-Level Multi Source Data Fusion

In this paper, a method for fault diagnosis of critical equipment in oil and gas station sites based on vibration data and deep learning models is proposed. The method utilizes 1DCNN and 2DCNN for diagnosis based on vibration fault data of rolling bearings in one and two dimensions, and performs decision-level multi-source data fusion through DS evidence theory, which integrally considers temporal and spatial features to improve the accuracy and reliability of fault diagnosis. The experimental results show that the multi-model fusion method based on DS evidence theory can more accurately identify the failure modes of the equipment and reduce the possibility of misjudgment and omission compared with a single 1DCNN or 2DCNN model. This method has important application value for realizing real-time monitoring and early warning of key equipment in oil and gas station sites, which provides strong support for ensuring the stability and safety of oil and gas production.