Ensuring Physical Consistency in Power Flow Sample Generation via Automated Tie-Bus Adjustment

With the wide application of data-driven methods in power systems, the requirements for the number and quality of power flow samples are increasingly higher. DC power flow model has the advantages of fast calculation speed and can avoid convergence problems, which has high efficiency in large-scale sample generation. However, the low accuracy and limited calculation scope restrict its further application in sample generation. For instance, the power injection at tie-buses always violates the physical consistency. To this end, this paper aims to study an automated adjustment method to ensure the tie-buses physical consistency of power flow samples. Our proposed method enhances the DC power flow model by integrating an automated tie-bus adjustment algorithm that dynamically corrects power injections to maintain zero injection at tie-buses, thereby adhering to physical laws and improving the realism of power flow simulations. We conduct case studies on the IEEE 118-bus test system to validate the effectiveness of our approach in generating physically consistent power flow samples. The results demonstrate improvements in the tie-bus power injection of the samples compared to other methods, offering a promising solution for power system operators and planners in their quest for high-quality power flow samples.