Relay protection cloud testing system mainly refers to clarify whether the relay protection device functions normally or not through offline testing and online testing technology, relying on the collection of data, signal analysis and so on. The article proposes a data object modeling method based on the IEC61850 standard protocol and designs a framework for a relay protection cloud testing system. Under the framework, relay protection is modeled, a database data storage and query indexing method based on a quadtree algorithm is constructed, and a relay protection fixed value inspection and alarm module is designed. In order to realize the intelligent warning of relay protection fixed value calibration, the intelligent warning model of relay protection fixed value calibration is established based on sensitivity and selective fixed value calibration and combined with RBF network. In order to verify the application feasibility of the relay protection cloud testing system, quantitative data analysis is carried out through example data. The average response time of index query based on quadtree algorithm is only 3.17 s, and the average time required to write grid monitoring data in parallel is only 93.54 ms. The intelligent warning effect of relay protection fixed-value calibration is more obvious, which can help to locate the faulty relay protection equipment of the grid, and when the system is introduced into inter-area grid scheduling, the maximal difference is 5.31%, which is within the permitted error range in engineering. Within the scope of engineering allowable errors. The application of data object modeling for industrial Internet of Things to relay protection modeling can better assist in identifying the accurate identification of grid relay protection and improve the stability of grid operation.