Long-term operation of the aqueduct leads to great discrepancies between the actual parameters and the designed parameters. In order to obtain the true physical and mechanical parameters of the aqueduct material, the parameters of the aqueduct are inversely analyzed based on measured steel stress. First, the stress monitoring test of steel bar with water in the field was carried out to obtain the measured steel bar stress. Then the hydraulic pressure factor was determined based on the finite element calculation of the three-dimensional aqueduct and the statistical model was established based on the measured steel bar stress. Regression analysis was used to isolate the water pressure and temperature components and finally combined the neural network mode, orthogonal design and numerical calculation to optimize the inversion of mechanical parameters of aqueduct. The results showed that the accuracy of the statistical model for measured steel stress established in this paper is high and the elastic modulus of the concrete in the inner edge, the concrete in the outer edge protective layer, the trunk and end concrete and the rebar are 37.87% , 34.97% , 66.23% and 63.79%, respectively of the corresponding design parameters. The degradation of the aqueduct is serious.