The purpose of the paper is to present the elaboration of models development methods for solving the tasks of electric power control for mechatronics systems such as public electric transport. The paper presents the model and the Menger’s algorithm for servicing time optimization of several high-voltage substations by optimizing the route. An intelligent multi-agent system is used for task solution. The algorithm is workable as in normal mode for maintenance team as for emergencies, when the fastest repair is needed. The workability of the algorithm is checked by an abstract numerical example. The model and the algorithm is realized in computer experiment for intelligent agent system. As an object, the part of Latvian high-voltage network model is taken. The optimal route for servicing and repair is calculated. All data for computer experiment coincide with the real object data. The computer experiment shows the optimal usage of supplied electric energy by mechatronics system such as Riga tram system.