Energy Efficiency Enhancement of a Multi-Agent Energy-Saving Heating Water Dispenser Through Collaborative Prediction Based on the MADDPG Algorithm
To improve the group performance of heating drinking w ater systems in public places and reduce overall energy consumption,a multi-agent collaborative prediction energy-saving heating water dispenser is proposed.The system uses a seasonal ARIMA model to predict the number of users of the water dispenser,combines decision trees and genetic algorithms to set critical thresholds for individual water dispensers,and applies multi-agent collaborative MADDPG algorithms to establish collaborative control between multiple water dispensers.By constructing an electricity cost regression model to test the performance of the prototype,the experimental results show that compared with commercial water dispensers on the market,the newly designed water dispenser can save 55%of electricity costs,and has potential application value in improving the energy utilization efficiency of water dispensers.
heating water dispensermulti-agent collaborative predictionMADDPG algorithmseasonal ARIMA
万方数据
维普
