The contact resistance of the wire mesh is very important to the current receiving performance of the wire mesh system. In this paper, an optimized support vector machine (SVM) method based on improved Grey Wolf optimization algorithm (IGWO) is proposed to analyze and model the contact resistance characteristics of the bow mesh. The sliding electrical contact experimental machine was developed to simulate the strong current sliding electrical contact conditions of the arch net under fluctuating load, and the data of multiple working conditions were collected to deeply explore the relationship between the contact resistance and sliding speed, contact current, fluctuation frequency, fluctuation amplitude and other factors. By introducing dimension learning hunting (DLH) search strategy, GWO was improved to optimize SVM parameters, and a prediction model was constructed, which was compared with traditional modeling methods. Experimental results show that IGWO-SVM model has excellent performance in evaluation indicators under different working conditions. Compared with BP, XGBOOST, RF, SVM and other models, R2 of the test set under three different working conditions are as follows: 98.84%, 98.63% and 96.807%, the accuracy of predicting the contact resistance of the arch net is higher, which can more accurately reflect the change of actual working conditions, and provide strong support for the optimization design of the arch net system.