Data-Driven Key Influential Factors Identification for Regional Renewable Energy Curtailment

Renewable energy consumption has become more serious issue as the installed capacity of renewable energy continues to grow. It is urgent to study the influential factors of the obstruction of renewable energy. This paper proposes a data-driven method for identifying the key influential factors for renewable energy curtailment in regional power grids. First, the set of influential factors is determined based on the actual measurement data. Then, taking the influential factors as input and the shed renewable energy as output, a back propagation neural network is built to fit the functional relationship between the input and the output. Next, the mean impact value algorithm is used to calculate the contribution of each factor, based on which the key factors can be determined. Finally, a case analysis based on the actual operating data of a provincial power grid is conducted, and results show that the proposed method can quantitatively analyze the influential factors of renewable energy curtailment and clarify their importance.