Experimental data is an important resource for detecting and evaluating the performance and safety of automobiles. The current experimental equipment has a wide variety of data types and a large amount of data, which can be effectively managed and used through information models. Therefore, this article focuses on the field of automotive testing, dividing test equipment data into equipment, projects, results, and controls, and abstracting the categories, attributes, and association relationships of the data to construct an automotive test equipment information model. Based on this model, XML language is used to define semantic standards, which are applied to edge gateways to achieve semantic conversion and format encapsulation of collected data, ensuring the standardization and uniformity of test equipment data, Lay the foundation for the application of experimental data.