Development of Detailed Power Control Technology within Buildings Using Artificial Intelligence

GIST Professor Hwang Ui-seok's Research Team "Usable for Smart Grid System Development"

[Asia Economy Reporter Kim Bong-su] A technology that can monitor and control detailed power consumption inside small and medium-sized buildings using artificial intelligence has been developed by domestic researchers.

The Gwangju Institute of Science and Technology (GIST) announced on the 30th that the research team led by Professor Hwang Ui-seok of the Department of Electrical, Electronics and Computer Engineering developed a non-contact detailed power consumption monitoring technology based on an AI-based time-frequency mask method.

This technology is a non-contact power identification and separation technology that estimates the power consumption of characteristic sub-loads from the total power measured by smart meters. It can be applied in the field of power demand response control and management in smart grids (intelligent power networks).

The research team improved the decomposition accuracy of sub-loads through an approach that estimates time-frequency masks based on a deep neural networks (DNN) AI model. It can identify and separate which loads consume more power in small and medium-sized buildings.

Previously, load decomposition was mainly performed in the time domain, but this method has difficulty separating or identifying each target sub-load pattern from the total power when sub-loads show similar patterns or have complex consumption forms.

The research team overcame the limitations of the existing time-domain decomposition method by applying a deep neural network-based time-frequency mask method, which can simultaneously consider time and frequency features, to non-contact power consumption monitoring for the first time. To reduce computational complexity and ensure proper learning of the AI model, environmental information highly correlated with power loads was used. Clustering of load data was performed in advance, and for each cluster, a load decomposition method using a deep neural network-based time-frequency mask was applied.

Professor Hwang Ui-seok said, “This research enables monitoring of power loads with similar or complex patterns that were difficult to identify with existing non-contact power consumption monitoring technologies, and improves load decomposition accuracy,” adding, “It is expected to be utilized in energy ICT convergence platforms for power demand response control and management systems in the future.”

The research results were published online on the 17th in the world-renowned journal ‘IEEE Transactions on Smart Grid.’

This content was produced with the assistance of AI translation services.