AN ENERGY CONSUMPTION PREDICTION SYSTEM FOR COMMUNICATION TOWER STATION EQUIPMENT ROOMS BASED ON THE COMBINATION OF GCN AND LSTM

This study designs an energy consumption monitoring and optimization system for communication tower station equipment rooms, based on a combination of Graph Convolutional Networks (GCN) and Long Short-Term Memory Networks (LSTM). The system aims to address issues such as low energy consumption data acquisition accuracy, high monitoring latency, and delayed energy-saving measures in traditional equipment rooms. The system collects energy consumption data through split-route acquisition, using hardware devices such as power transformers and AD7606 chips to monitor the energy consumption of key equipment in real-time. By integrating GCN and LSTM, the system can analyze the energy consumption relationships and trends of devices in the equipment room, providing accurate predictions of energy consumption for the next cycle. The research results show that this system can effectively predict node energy consumption, and provide an intelligent solution for the green transformation and energy-saving emission reduction in the telecommunications industry.