Abstract: A dam is an important piece of infrastructure for ensuring economic and social development. During operation, because of environmental changes, aging materials, and other factors, a dam may develop accident risks and once it fails, it poses a great threat to society. Therefore, it is of great significance to use reasonable methods for analyzing the monitoring data collected by a dam safety monitoring system and evaluate a dam’s behavior to ensure operation safety. At present, the existing methods are mainly devoted to evaluating the local state of a dam according to the monitoring information of a single measuring point. Relatively few studies are available on the evaluation methods for the overall state of a dam, and the existing methods are mainly qualitative and subjective. To address this problem, the residual between the model calculated value and the measured value was taken as the research basis. The concept of the fusion residual, an important index for characterizing the overall behavior of a dam, was promoted. Combined with the information entropy theory, the variation of residuals at different measuring points was studied, and the fusion weight of residuals at each measuring point was analyzed. The fusion residual was calculated. Based on the distribution analysis of the fusion residual, a concept cloud representing the different states of a dam, namely, the evaluation criteria, was established using a reverse cloud generator and a forward cloud generator. On this basis, an evaluation model of the overall behavior of a dam was established and combined with the cloud similarity algorithm. The example shows that the evaluation method can effectively identify the abnormal value of a dam and evaluate its overall behavior. The evaluation results are reasonable and reliable. The model can evaluate the overall behavior of a dam quantitatively and objectively, and the evaluation results are reasonable and reliable, providing an important reference for the safe operation of a dam.