Abstract: Load clustering is an important part of the strategy to achieve "peak carbon dioxide emissions" and "carbon neutrality". Based on the shortcomings of the existing algorithms, this paper proposes an integrated algorithm for multiple clustering of load profiles considering the overlapping relationships between clusters. The label trust space is constructed by using the manually defined hierarchical clustering division center to evaluate the label trust of the hierarchical clustering algorithm and form multiple trusted clusters at first. Secondly, based on the concepts of the potential clusters and the potential clustering centers, the inter-cluster similarity matrix of the credible clusters is constructed by using the similarity measure function that considers both the numerical and morphological characteristics of the load curves. Finally, the alignment of the labels is completed and the final clustering results are obtained by using spectral clustering as the final integration means. The comparisons of the results in this paper with those of the hierarchical clustering and with the spectral clustering algorithms show that the proposed integrated algorithm in this paper has a better performance in SC coefficients, DBI metrics and CH metrics, and that it has more reasonable curve clustering images, thus the effectiveness of the algorithm is verified.