In response to the challenges posed by the strong randomness and volatility of new energy power generation during the transition to a new power system, demand-side flexible resources hold significant potential. As major electricity consumers, large industrial users have considerable untapped potential in utilizing their flexibility resources. This paper employs the main equipment analysis method to perform clustering analysis on large industrial loads, summarizing loads into Curtailable, Shiftable, and Segmentable types. On this basis, an optimization model for demand-side response of clustered large industrial loads is established, with the objective of maximizing new energy consumption, and subject to constraints including clustered load constraints and power system constraints. Finally, using typical daily data from a provincial grid as a case study, the model’s effectiveness in accommodating new energy is verified. Among the load types, the Shiftable load shows the greatest potential, increasing new energy utilization by 9.7323%, followed by Segmentable load with an increase of 7.1023%, while simultaneous participation of clustered loads can also improve the output curve of thermal power units.