Data driven approach for power system state estimation with consideration of multi-level noise

State estimation is vital tool in observing electrical power system. It is always important to know the operating state of the system with accuracy and less computational efforts. This paper presents a novel data driven approach to perform state estimation of power transmission system using deep neural networks (DNN). The network is trained offline using dataset prepared by generating multiple loading scenarios through programming. The proposed method is evaluated using IEEE 14 bus system with variable Gaussian measurement error at different load buses.