Data-Driven Power Flow Calculation Based on Deep Learning

With the rapid advancement of modern power system reform and the popularization of large power grid system and supervisory control and data acquisition (SCADA) system, data-driven method has become a new field in power system research. Aiming at the problem that the traditional power flow calculation method cannot update the system operation status in real time due to the real-time change of power system topology, this paper proposes a data-driven power flow calculation method based on deep neural network. The calculation results are compared with Newton's method and PQ decomposition method. The results show that this method is superior to the above traditional algorithms in accuracy and calculation time.