Abstract: Precise control of icing wind tunnel cloud parameters and the accuracy of measured parameters are two technical problems of icing wind tunnel experiments.They affect the accuracy of the experiment and bring errors to results.To solve the problems,this paper proposes an ice shape correction method based on cloud parameters.The key element of the ice shape correction is analyzed with the aerodynamic point of view,and the relationship between the cloud parame-ters and the typical geometric characteristic parameters of the ice is simulated by using artificial neural network technology.A TFI ice shape correction method is established.The NACA0012 airfoil is selected as the research obj ect.The ice shape has been corrected to account for the error caused by the cloud parameters of liquid water content and droplet diameter.The method has sig-nificantly improved the agreement between the corrected ice shape and the target shape.It is thus verified that our method is correct and it can be used for icing wind tunnel test.