Lost circulation is a common complex condition in drilling operations that severely compromises safety and efficiency.Most existing early-warning models focus mainly on predicting when lost circulation occurs, with insufficient attention to its severity andrisk assessment. To address this gap, an intelligent lost circulation diagnostic model integrating logging time series data and multi-scaledecomposition was proposed, combining ICEEMDAN ( improved complete ensemble empirical mode decomposition with adaptivenoise) , IMF ( intrinsic mode function) screening criteria, and SVM ( support vector machine) optimized by PSO ( particle Swarmoptimization) . First, from drilling and completion reports of Iran''s Azadegan Oilfield, logging parameters that effectively characterizedifferent leakage degrees were selected. Second, the logging data undergo multi-scale decomposition via ICEEMDAN. IMF componentsstrongly correlated with leakage characteristics were filtered using Pearson correlation coefficients and energy entropy, and the data arereconstructed to eliminate medium-to-high-frequency noise. Finally, a diagnostic model was built with the PSO-optimized SVM, inwhich the PSO optimizes the penalty coefficient C and kernel parameter γ, to enhance accuracy and generalization. Results indicate the ICEEMDAN-PSO-SVM model''s test set accuracy exceeds 98% , outperforming PSO-SVM, RF ( random forest) and SVM across keyperformance metrics. This study offers valuable support for mitigating drilling safety risks and enhancing engineering efficiency andeconomic benefits.