Application of Artificial Intelligence Algorithm in Building Electrical Fault Diagnosis

With the continuous development of modern building system towards automation and intelligence, the importance of building electrical system is becoming more and more prominent. Its operation status and reliability directly affect the normal operation of building system. Therefore, research and analysis of the building electrical system fault diagnosis has important value for development of the construction industry. Most building electrical systems are neutral grounding systems, with large fault current. Once grounding or short circuit faults occur, the whole system will lose stability. Therefore, the failure will cause greater harm to the distribution system and electrical equipment. In this paper, the main research object is the failure of the electrical system in the building. Common building electrical faults are simulated by building electrical comprehensive experimental platform. And use low-voltage electrical comprehensive tester to collect fault data. Firstly, a fault diagnosis algorithm for building electrical system based on wavelet analysis and support vector machine algorithm is proposed. At the same time, considering the low efficiency of building electrical system diagnosis, this paper proposes a fault diagnosis algorithm based on compressed sensing and K nearest neighbor algorithm. It aims to improve the efficiency of building electrical system fault diagnosis.