The invention discloses an optical access network ODN intelligent clustering planning method for sparse user distribution, which belongs to the field of communication technology; specifically, it includes: firstly, for a planning area, in the distribution section of the ODN network, the users in the area The geographic location is input, and the position coordinates of k initial centroids are randomly generated as the second-level splitting points; then, the cost function T is constructed, and the number k of the initial centroids and the position coordinates are optimized; then, the position of the first-level splitting points is determined Coordinates and optimization; according to the optimized position of the secondary and primary splitting points, calculate the reserved length of the optical fiber, and complete the planning of the distribution section; for the planning of the feeder section of the ODN network, determine the weight connection through the connection relationship between the field areas As shown in the figure, the minimum spanning tree algorithm is used to find the minimum spanning tree; finally, according to the planning results of the distribution section and the feeder section, the optical access network ODN planning for sparse users is completed. The present invention combines upper and lower, and reduces the cost of optical fiber planning while realizing optimal planning.