Abstract: Power systems are developed to perform with high reliability, even though, critical outages may trigger a cascade of events ending up in a large-area black out. In order to minimize the economic losses caused by large-area blackouts, an effective and rapid black-start restoration strategy is becoming very important. Black-start is defined as the process that a power system suffered from a complete blackout is restarted through reconfiguring its networks and recovering its loads depending on self-starting units (black-start units), without relying on other systems. In a typical black-start scenario, black-start units provide power to non-black-start units located close to them, critical system loads are recovered when these non-black-start units come on-line, and the re-supplied area is gradually expanded until the entire power system is restored. Evaluating black-start schemes is one of the key issues having impacts on the restoration speed of the power system concerned. Usually power system operators or experts are willing or able to provide only incomplete information, because of time pressure, lack of knowledge or data, and their limited expertise. However, most studies focused on the black-start decision making problem with complete information. In this article, a comprehensive black-start decision making method based on EM algorithm and weighted rank sum ration is proposed, which addresses the issue of evaluating black-start schemes with incomplete information. Firstly, the absent values are imputed by EM algorithm to obtain a complete black-start decision matrix. Then, the difference weight approach is used to determine index weights. Finally, the score of each scheme is computed according to the weighted rank sum ration approach. Based on the data of Guangdong power system of China, experiments are carried out to investigate the effectiveness of the proposed method. Experimental results show that the proposed method can reasonably fill the missing data of the black-start decision matrix, and performs better than the existing black-start decision making methods. This article can enrich the theories of black start decision-making, and can provide series of useful guidance for the decisions of power system operators.

Key words: power system restoration; black-start; incomplete information; EM algorithm; weighted rank sum ration