A high-rise building passive aerodynamic optimization method based on agent model and wind tunnel test interaction

Abstract

The application is a high-rise building passive aerodynamic optimization method based on agent model and wind tunnel test interaction. The application relates to the field of high-rise building wind resistance technology. Based on pre-test and parameter sensitivity analysis, the application determines the high-rise building passive aerodynamic parameters and their value range, carries out corresponding wind tunnel test, constructs a target domain training set, and further constructs a transfer learning model. The Pareto optimal front is determined, wind tunnel test verification is carried out on multiple candidate working conditions in the Pareto optimal front, the wind tunnel test verification result is compared with the model prediction result, the optimal solution is output, and the aerodynamic control mechanism is analyzed. Through the transfer learning model, the application establishes a nonlinear mapping relationship between the high-rise building passive aerodynamic optimization parameters and the target response indicators under a small sample. The application can reduce the number of wind tunnel tests, reduce the model manufacturing cost, and improve the high-rise building passive aerodynamic optimization efficiency.

Images