Self-learning dynamic graph neural network with self-attention based on historical data and future data for multi-task multivariate residential air conditioning forecasting

In the context of escalating energy consumption in buildings, particularly from air conditioning (AC), the intelligent control of AC has become increasingly crucial. Accurately predicting future energy consumption for AC, the indoor environment, and determining the optimal settings have emerged as k...

Celý popis

Uloženo v:

Podrobná bibliografie
Vydáno v:	Applied energy Ročník 364; s. 123156
Hlavní autoři:	Hu, Zehuan, Gao, Yuan, Sun, Luning, Mae, Masayuki, Imaizumi, Taiji
Médium:	Journal Article
Jazyk:	angličtina
Vydáno:	Elsevier Ltd 15.06.2024
Témata:	Graph neural network Multi-task multivariate time-series forecasting Residential building air conditioning Self-attention Residential building air conditioning Graph neural network Multi-task multivariate time-series forecasting Self-attention
ISSN:	0306-2619, 1872-9118
On-line přístup:	Získat plný text
Tagy:	Přidat tag Žádné tagy, Buďte první, kdo vytvoří štítek k tomuto záznamu!

Popis
Shrnutí:	In the context of escalating energy consumption in buildings, particularly from air conditioning (AC), the intelligent control of AC has become increasingly crucial. Accurately predicting future energy consumption for AC, the indoor environment, and determining the optimal settings have emerged as key challenges in intelligent AC control. In this study, a hybrid self-learning dynamic graph neural network with self-attention mechanism is proposed for AC forecasting. Addressing the gaps in the existing graph neural network applications, this model overcomes the limitations of static graph structures by constructing evolving adjacency matrices integrated with a gated recurrent unit and self-attention, effectively capturing the dynamic relationships between changing feature quantities. Additionally, a multi-task prediction (MTP) module that utilizes both past and future data is proposed. The MTP enables the application of a single model to multiple prediction tasks, thereby obviating the need for separate model training for each task. An experiment in an actual outdoor environment was designed to verify the predictive performance of the proposed model. The results indicate that the proposed model achieves superior accuracy for all target variables across different tasks under various AC conditions, particularly for variables with strong non-linearity, which showed a maximum improvement of 24.94% in correlation coefficient (R2) compared to long-short term memory network. With the MTP, the single model applied to multiple prediction tasks exhibited only a minimal sacrifice in accuracy, resulting in a mere 0.64% decrease in average R2 of all target variables for the proposed model. •A novel graph neural network for air conditioning forecasting is proposed.•The proposed model overcomes static graph structure limitations.•Developed a module that enables handling multiple predictive tasks.•An experiment was conducted to validate the accuracy of the model.•Proposed model shoed the best performance across all scenarios.
ISSN:	0306-2619 1872-9118
DOI:	10.1016/j.apenergy.2024.123156