Self-adaptive point adding strategy uncertainty optimization method and system based on chaotic polynomial

Abstract

The invention discloses a self-adaptive point adding strategy uncertainty optimization method and system based on a chaos polynomial. The method comprises the following steps: acquiring a certain number of initial samples by adopting a Latin hypercube method; constructing a Kriging approximation model of an objective function mean value and a variance of the initial sample on a Gaussian-Hermite integral point, and calculating a current optimal point of the Kriging approximation model by using a point adding criterion and a global optimization algorithm; adding the obtained current optimal point to the initial sample as a sample data point, updating the approximate model, and carrying out global optimization on the new approximate model to obtain a next optimal point; and comparing the error between the current optimal point and the next optimal point with the approximation precision of the Kriging approximation model near the next optimal point, and determining the optimal value point of the target function. According to the method, the efficiency of uncertainty problem analysis can be improved, the number of sample points for constructing the proxy model is reduced, and the efficiency of target function uncertainty optimization is improved.

Description

technical field

[0001] The invention belongs to the technical field of airfoil uncertainty optimization, in particular to an uncertainty optimization method and system based on chaotic polynomial adaptive point addition strategy. Background technique

[0002] For conventional airfoil optimization design, the design variables are deterministic, but in actual situations, there are many uncertain factors, such as the uncertainty of geometric parameter processing and the uncertainty of flight environment. The uncertainty of the airfoil flight angle of attack will affect the lift-drag ratio of the airfoil. For a supercritical airfoil, the uncertainty of the Mach number will seriously affect the change of the drag coefficient of the airfoil, resulting in an increase in the drag coefficient.

[0003] Traditional uncertainty optimization is mainly based on Monte Carlo simulation (MCS), through a large number of random sample point calculations to obtain the mean and variance of rand...

Images