The problem of "dimensionality disaster" of design variable is a key technical problem that restricts the application of current surrogate-based optimization algorithms. In order to solve the problem of declining accuracy and poor optimization effect of surrogate model caused by dimensionality disaster problem, this paper improves a sensitivity analysis method based on physical knowledge driven partition surrogate model optimization. Based on physical knowledge and manual design experience, the design space is partitioned, and the molecular space units are divided by three-dimensional wing design section to achieve dimension reduction. In order to further improve the design effect of the partition optimization, the sensitivity of different partition design variables to the objective function is studied. On the basis of the sequential partition optimization, the sensitivity is taken as the order of the partition surrogate model optimization. The results show that the partitioning surrogate model optimization algorithm based on physical knowledge-driven sensitivity analysis can search for a better optimal solution, and the time spent on establishing the surrogate model is much lower than that of the EGO algorithm, and it can maintain good accuracy of the surrogate model, which can provide a new idea for the dimension disaster problem, and has certain engineering reliability and applicability.