To address the problems of semantic heterogeneity of multi-source data and difficulty in reusing operation and maintenance knowledge in the intelligent operation and maintenance of gas stations, this paper proposed a construction method of gas station ontology model for intelligent operation and maintenance. First, according to the operation and maintenance business process of gas stations, this paper sorted out the concepts and correlation relations of core knowledge dimensions such as equipment asset, process system, operation and maintenance task, and fault diagnosis. Second, this paper adopted a 4-stage modeling method of “requirement analysis-concept extraction-hierarchy construction-rule definition”, constructed the intelligent operation and maintenance ontology model of gas stations based on OWL language, and realized the formalized expression and consistency verification of the model through Protégé tool. Finally, this paper connected the ontology to the intelligent operation and maintenance prototype system, and selected the fault diagnosis scenario of pressure regulation equipment in typical stations for application verification. The results indicate that the ontology model can realize the semantic fusion of multi-source operation and maintenance data, and the accuracy of fault diagnosis improves by over 34% compared with the traditional empirical method, which can provide effective knowledge support and decision-making assistance for the intelligent operation and maintenance of gas stations.