The invention discloses a demand response resource clustering method. During initial running, the demand response resource clustering method comprises the following steps: 1, performing discrete Fourier transform on previous-day loads of all resources in a garden, and extracting first five Fourier components; 2, standardizing the extracted components and the characteristic quantity of the resources to form a single-day standard characteristic vector of the resources; 3, setting a clustering center and a clustering ending condition, and clustering the resources according to an FCM (Fuzzy C-Means) algorithm and the single-day standard characteristic vector of the resources till a clustering ending condition is satisfied; and S4, updating clustering according to a clustering result obtained in the step 3. Through adoption of the demand response resource clustering method, conventional resource classifying manners based on information such as workers, merchants and residents is broken through, and clustering of response resources is realized, so that the demand response efficiency is increased remarkably, and the resource response success rate is increased.