Artificial Intelligence and Automated Operation and Maintenance Technology in Wireless Network Power Systems

Smart grid promotes the evolution of modern power system to intelligence and automation, which makes the application of artificial intelligence and automatic operation and maintenance technology in wireless network power system of great significance. In this paper, the application performance of three cutting-edge time series prediction technologies, including long-term and short-term memory (LSTM) network, autoregressive integrated moving average (ARIMA) and gated recursive unit (GRU), in wireless network power system is analyzed. Then the architecture design of wireless power system is put forward, and the coverage of wireless network and the security and reliability of data transmission are introduced. The experimental results show that the LSTM model has high time series prediction performance, the maximum failure rate is 1.3%, and the response time is 215-391 milliseconds, which reduces the system failure time and proves that the maximum failure time of the system is 157 minutes.