A hybrid energy management approach for home appliances using climatic forecasting
energy forecasting, particle swarm optimization, artificial bee colony, artificial neural network
Energy management refers to saving the power by employing effective monitoring and control strategies. The demand for energy is rising in all sectors, such as residential, industrial, transportation, and agriculture, owing to our dependency on electronic appliances. Hence, of late, energy management in the households has become a pertinent issue. Electricity consumption depends on various factors, including climatic conditions, number of occupants in the household and their behavior, usage of appliances, etc. The utilization of electronic appliances and the climate conditions are inter-related; if the outside temperature is high, the usage of ACs in the house increases and vice-versa. Therefore, the climatic conditions are the most relevant factors in energy consumption. There is a need to manage the energy demand by using certain optimization approaches and by predicting the demand based on different climatic conditions. In this study, the prediction model of Artificial Neural Network (ANN) was merged with the optimization methods such as Particle Swarm Optimization (ANN-PSO) and Artificial Bee Colony (ANN-ABC). The experimental results revealed that ANN-ABC performed in a superior manner by reducing the energy consumption by up to 41.12 kWh per day. Finally, the prediction results were compared with the existing model and it was verified that the energy prediction with climatic conditions gave the better results of the power usage in the households.
Tsinghua University Press
Jasmeet Kaur, Anju Bala. A hybrid energy management approach for home appliances using climatic forecasting. Build Simul, 2019, 12(6): 1033–1045.