Abstract: Aggregated baseline load（ABL）refers to the sum of all customer baseline loads（CBL）managed by the load aggregator,which is the basis for system operators and load aggregators to make compensation and settlement in incentive-based demand response. Customer clustering based estimation methods are common ABL estimation methods at present. However, the two steps of “clustering” and “estimation” are implemented separately in this kind of method, and their goals are not unified, so the estimation accuracy needs to be further improved. Therefore, an ABL estimation method based on clustering-estimation linkage is proposed. The basic idea is to take the estimation accuracy as the guidance to adjust customer clustering, and find an optimal customers clustering to achieve the highest accuracy of ABL estimation. Firstly, the first-level clustering on all customers is performed, and all customer loads in the obtained clusters are accumulated as the input of second-level clustering. Secondly, the second-level clustering is carried out, and the clustering is adjusted according to the performance of the estimation results in the validation set, so as to obtain the optimal clustering results. The proposed method is compared with the existing methods on a real data set, and the results show that the proposed method can effectively improve the estimation accuracy of ABL.