A Data-Driven Approach to Super-Real-Time Simulation of City Power Grids

With the rapid development of the city power grid, the power system has become increasingly complex, making research more challenging. To better understand the dynamic characteristics and control strategies of the power grid, power grid simulation has become a critical approach. In this paper, we present a super-real-time simulation method for city power grids using deep learning techniques. Our approach takes the running state of the power grid as input and outputs high-resolution power grid simulation data. Specifically, our method generates power grid operation data using a power grid data generation model and forecasts long-term yields using a forecasting model. We use a super-resolution model to obtain high-resolution city power grid simulation data. We validate our approach using actual power grid data. The experimental results demonstrate that our model accurately simulates the operation data of the power grid, confirming the successful simulation of the running state of the power grid.