The new artificial intelligence framework estimates the potential of demand response in improving power consumption and reducing carbon footprints. Source: Guangzhou Science and Technology College
The grid is almost too large and cannot meet the short-term surge in energy demand. Simply put, the power station needs to have excess generators to provide electricity during peak hours. This mismatch between this power supply and the inefficient operation of the power station have led to higher carbon dioxide emissions. In addition, increasingly popular distributed energy, such as roof solar panels, only exacerbate supply and demand.
Fortunately, communication technology provides a smart policy to solve this problem: demand response (DR) program. In this scenario, the user is encouraged to reduce electricity at peak hours by reducing the electricity price other than the peak time and in advance. In addition, they can be integrated with distributed energy management to reduce the load of the grid when necessary.
However, there are very few research attention to the potential benefits of using the real user behavior data assessment DR program. To this end, Scientists of the Guangzhou Science and Technology Research Institute have developed a new method based on artificial intelligence (AI), which analyzes and extracts the behavior of grid users based on energy consumption of each household. Their papers were published in "IEEE Transactions On Smart Grid), the author describes a data-driven framework that estimates the best disaster relief management of each family, considering User home appliances and behavior patterns, as well as predictive power generation of distributed energy.
Researchers test their model by simulating data in real world. "In our simulation, we consider and quantify the degree of user discomfort related to household appliance dynamics, and then use it to estimate the best disaster-disaster potential," Professor JINO KIM "Professor Leaders explained. The research team also calculated the potential contribution of the disaster preparation program in reducing carbon dioxide emissions and managing coal-fired generators.
In summary, this study shows how to use artificial intelligence to improve our power consumption, achieve lower prices and smaller carbon footprints. Professor Gold emphasizes: "Analysis based on big data can convert family energy demand information to large-scale comprehensive resources." "We believe this technology will further expand to improve the efficiency and coupling of other departments, including water, heat , Gas and electric car sector. "
We certainly hope that his ideal is achieved soon.