To address the issues of the limitations of the current transformer oil-paper insulation system moisture assessment model selection and the low directivity of the feature indexes for insulation moisture assessment under the synergistic effect of aging and moisture，a quantification method of the moisture degree of oil-paper insulation based on Stacking heterogeneous integration is proposed.Firstly，the mechanism analysis is conducted about the change trend of the macro insulation state under the synergistic effect of aging and moisture synergism at the micro-medium level，and multi-source heterogeneous moisture feature indexes that can eliminate the interference of aging are extracted，forming a highly correlated feature set with moisture.Secondly，based on four types of heterogeneous algorithms that are integrated based on the fusion learning idea of Stacking，the primary assessment system of the overall model is constructed，and output noise is reduced through the Optuna hyper-parameter tuning framework.Finally，the stack data after weight assignment is utilized to train the sub-assessment system，and then a weighted and improved Stacking model is built to quantify the moisture degree of the oil-paper insulation system.Lastly，the effectiveness of the proposed model is validated in moisture assessment by taking the measured data as an example.The model can serve as a reference for assessing moisture in transformer oil-paper insulation systems.