Output details

Robust state estimation using mixed integer programming

Conventional power network state estimation, which is widely applied in the industry, is able to reduce the effect of measurement noise in network monitoring. Robust state estimation extends this facility by eliminating the effect of gross measurement errors caused by instrumentation or communication faults. This paper provides the first algorithm to guarantee to find a solution which minimises the hypothesised number of faults needed to explain the measurement set. The method is attracting international interest and is being further developed by leading researchers in China (Guangyu He, Shufeng Dong, Junjian Qi and Yating Wang of Tsinghua University, Beijing).