It is necessary to accurately identify the mechanical properties of skin tissue for diagnosis,evalu-ation and treatment of skin tissue diseases by means of mechanical modeling.Therefore,this paper propo-ses a skin tissue constitutive parameter identification method using adaptive simulated annealing optimiza-tion algorithm combined with approximation model technology.First,the finite element method was used to simulate the skin uniaxial tensile test,and the numerical mechanical response data of skin tissue were ob-tained under different parameter combinations.In order to improve the computational efficiency of parame-ter identification,response surface model,Kriging model,and ellipsoidal neural network were constructed to replace the repeated simulation calculation process,and the fitting accuracy of the three approximation models was verified by the determination coefficient R2.Finally,an adaptive simulated annealing optimiza-tion algorithm was used to identify the constitutive parameters that best matched the uniaxial tensile test re-sults of skin tissue of common pig belly through inversion with the objective of minimizing the root mean square error of the test curve and the numerical calculation curve:C10=0.140 1 MPa,k1=24.51 MPa,k2=0.496 1,κ=0.317 1,and cp=13.86°.The results show that the ellipsoidal neural network model is more suitable for fitting the nonlinear relationship between skin constitutive model parameters and stress-strain response.The comparison between the identified numerical and experimental curves shows that the adaptive simulated annealing algorithm combined with the approximation model is a fast and reliable meth-od to identify the anisotropic hyperelastic constitutive parameters of skin tissue.