A Method for Identifying Uncertain Dynamic Loads of Wing Structure Based on Support Vector Regression

Abstract

The invention discloses a method for identifying uncertain dynamic loads of a wing structure based on support vector regression. The method expresses the uncertain parameters of the wing structure with interval variables, and combines support vector regression and interval dimension-by-dimension calculation to identify all the factors of the wing structure. subject to dynamic loads. First, various types of training load samples are determined, and the heterogeneous responses of the wing structure under different parameters are obtained as training response samples. Aiming at the training sample data, a load regression hyperplane based on support vector regression is established, and its direction vector and position parameters are solved based on the optimization theory, and the load identification model is obtained. Enter the measured normalized heterogeneous response to obtain the load history under various uncertain parameters. Finally, based on the best square approximation of the polynomial, the functional relationship between the uncertain load and each dimension interval variable is calculated dimension by dimension. The load interval boundary at a time.

Description

technical field [0001] The invention relates to the technical field of aerospace, in particular to a method for identifying uncertain dynamic loads of a wing structure based on support vector regression. Background technique [0002] In the fields of structural health monitoring, structural vibration control, and structural optimization design, it is crucial to accurately grasp the input information of external loads. In practical engineering, due to the limitation of sensor technology and the form of external load, it is quite difficult to directly measure the external excitation through the force sensor. Therefore, load identification techniques for indirect inversion of external excitations based on easily obtained structural response signals have received extensive attention in recent years. [0003] The dynamic load identification technology emerged in the 1970s and was initially studied to improve the mechanical properties of helicopter composites. In the past few de...

AI Technical Summary

Problems solved by technology

In engineering practice, the acceleration response is relatively easy to measure, and the cost of the acceleration sensor is low, but when only the acceleration response is used for load identification, the low-frequency load identification results will be unstable.

The use of displacement response for load identification can avoid the phenomenon of unstable identification results, but the displacement sensor is expensive and difficult to install in large quantities on the structure

Images