In the context of building a new power system, the problem of new energy consumption has become increasingly severe. Studying the factors that hinder the consumption of new energy and quantitatively describing their impact is of great value to the optimized operation of high-proportion new energy power systems. Based on machine learning technology, this paper proposes an intelligent identification method for key factors that hinder the consumption of new energy in regional power grids. First, determine the set of blocking factors based on actual measurement data and engineering experience; then use the blocking factor list data as input and the new energy blocking amount as output to establish a multi-layer feedforward neural network to fit the functional relationship between the input and output, and further The BP-MIV algorithm is used to calculate the contribution of each blocking factor to the blocking amount of new energy, so as to determine the key factors causing the blocking of new energy. Finally, a case analysis is conducted based on the actual operation data of a provincial power grid in northwest China. The results show that the proposed method supports Quantitatively analyze the key factors affecting new energy consumption in a data-driven manner.