Adaptive Dynamic State Estimation Method for Distribution Networks with Enhanced Robustness

In order to improve the robustness and accuracy of dynamic state estimation for uncertain systems, a dynamic state estimation method for distribution network based on adaptive mixed Kalman/H∞ filtering is proposed. The filtering gain of this method is obtained by weighted sum of the gain of extended Kalman filtering (EKF) and H∞ filtering. By mixing the two algorithms, the method achieves the performance balance of the two filters. Different from the traditional estimator, this method uses the performance index of Kalman filtering to adjust the weights of both filters, so as to improve the tracking ability of state of the dynamic system represented by the distribution networks. At the same time, a scaling factor is introduced to enlarge the performance bound of H∞ filter to reduce the bad influence of exceedingly conservative external factor assumptions in traditional H∞ filter on the accuracy of the result. Experimental results on various states of distribution networks further verified the feasibility of the proposed dynamic state estimation method.