Fusion of Multi-time Section Measurements for State Estimation of Power System
Xiaoli Liu, Lei Yue Yao, Xianghui Zeng, Shuaidong Zhang, Muqing Liang, Changhong Deng, Yujie Ding · 2019
With the large-scale development and utilization of renewable energy such as wind power, photovoltaics and hydro- power, as well as the gradually increasing scale and complexity of the power grid in China, it is an important implementation to establish a real-time monitoring, analysis and control system (RTMCS) for ensuring the safe and stable operation of the power system. To great extent, the performance of RTMCS depends on the quality of the data which could be available from power system state estimation. In this paper, a data-driven fusion method based on convolutional neural network (CNN) was proposed to enhance the performance of power system state estimation. Firstly, the CNN was trained offline by historical measurements. Then, the real-time measurements were fused and estimated online by the forward calculation function with the neural network. Finally, we took state estimation as an example to test the performance of CNN proposed in this paper. Compared with the typical state estimation method, the simulation results demonstrated that the proposed method presented a higher estimation accuracy with a simpler model.