Large-scale power outage events bring serious economic losses to national and social development and cause bad impacts, and the reasonable formulation of a unit black-start strategy is the basis of power outage recovery. Firstly, the Dijkstra shortest path algorithm is used to search for the optimal recovery path of the unit to be started after a major blackout occurs, and secondly, a comprehensive index of the unit start sequence considering the unit characteristics and unit distance is defined to guide the unit start sequence formulation by combining the unit capacity, climbing rate, start power, and other characteristics with the recovery time and capacitance value of the recovery path. Finally, based on the unit start sequence and optimal recovery path, and taking into account the unit start constraints, a complete unit black-start strategy is proposed, and the effectiveness of this strategy is verified by IEEE30 node system simulation. The results show that the proposed unit black-start strategy can reduce the unit recovery time and improve the recovery success rate.

The fluctuation of new energy output makes the power system operation more complex, and the traditional power system tide calculation is no longer fully applicable. Therefore, the determination of wind farm output and site selection is important and fundamental research at present. Based on this, a combination of wind farm output scenario and probability is proposed. The site selection model with the minimum network loss of the distribution network as the optimization objective is proposed. First, the wind farm output is clustered into several typical scenarios by K-Means, and the output time series model and the probability of each typical scenario are obtained. The probabilities of each scenario are multiplied by the network losses under that scenario. Then, the network losses obtained for all individual scenarios are summed up and used as the minimum objective function. The model is solved by a genetic algorithm with tidal current constraint and node voltage constraint to derive the optimal wind farm access point. With the IEEE33 node, simulations are performed to derive the minimum value of the daily average network loss, which verifies the rationality and feasibility of the optimized model.