Cyber-Physical System of Intelligent Diagnosis of Generator Winding Insulation

The article proves the urgency of developing a cyber-physical technical condition diagnosis system for high-voltage generating equipment. The most vulnerable part of such equipment is the winding insulation. The most simple and affordable method for determining this fault is the analysis of the electrodischarge activity (EDA). This method allows determining the insulation defect at the initial stage of development. It does not allow for a critical failure. One of the main problems of generating equipment diagnostics is the considerable distance of its elements from each other. At the same time, an important task is the exchange of information on the state of the equipment elements. To solve this problem, a cyber-physical prediction diagnosis system is proposed. The diagnosis method for generating equipment windings is proposed. This method allows determining the technical condition of the generators insulation during its operation without complicated measuring devices using artificial intelligence. The effectiveness of this method for turbogenerator diagnosing is shown.