Application of energy scheduling algorithm based on energy consumption prediction in smart home energy scheduling

To effectively optimize household energy management, enhance user comfort, and ensure the stability of the power grid, this paper directs its focus towards household equipment and presents inter-household and indoor energy scheduling algorithms. Grounded in a two-dimensional dynamic device priority framework and predictive device operation modeling, this study demonstrates that the utilization of scheduling algorithms can avert grid overload scenarios during periods of low electricity prices. This results in more consistent hourly electricity consumption and a substantial peak-to-average ratio reduction of up to 44.86 %. Additionally, the application of the continuous wavelet neural network algorithm proves to be proficient in accurately forecasting home device operations. When the time was 17.5 h, the actual energy consumption of the water heater was 7.12 kWh, which was 0.78 kWh more than the predicted energy consumption. Utilizing scheduling algorithms, as opposed to not employing them, led to notable reductions in electricity expenses, demonstrating a cost savings rate of 15.32 % on the third day. Additionally, it facilitated decreased usage of the power grid. When the time was 18 o'clock, the power consumption of the grid using scheduling algorithms was 0.24 kWh, which was 0.25 kWh lower than that without scheduling algorithms. Following the implementation of the scheduling algorithm, minimal deviation in device startup times was observed during periods of high comfort, with offsets ranging from 0 h. The research methods employed demonstrate efficacy in orchestrating energy distribution among households and indoor spaces, thereby striking a harmonious balance between energy conservation and user comfort.