Measuring Method of Aircraft Wing Load Based on Hybrid Feature Selection Algorithm

Abstract

The invention discloses an aircraft wing load actual measurement method based on a hybrid feature selection algorithm. The method comprises the following realization steps: (1) generating an initial training set; (2) performing data pre-processing; (3) generating an initial population matrix; (4) generating an individual matrix; (5) solving a coefficient matrix of the individual matrix; (6) calculating the adaptability of the coefficient matrix; (7) generating a population matrix; (8) judging whether to terminate or not; (9) obtaining the optimal coefficient matrix; (10) calculating aircraft wing load in actual flight. According to the aircraft wing load actual measurement method, an independent feature selection method is adopted for preprocessing the initial training set, strain bridge data which is unrelated to and is weakly related to target load is deleted, then the population matrix is iteratively generated, and the adaptability of the population matrix is calculated to obtain the optimal coefficient matrix, so that the precision of aircraft wing load actual measurement is improved; meanwhile, a gradient matrix is generated, and the coefficient matrix is searched in the negative gradient matrix direction, so that the coefficient matrix is obtained.

Description

technical field [0001] The invention belongs to the field of aerospace technology, and further relates to an actual measurement method of an aircraft wing load based on a mixed feature selection algorithm in the technical field of flight load measurement. The invention can realize high-precision real-time measurement of aircraft wing loads. Background technique [0002] Flight load measurement plays an important role in the entire aircraft development process, and can verify whether the aircraft structure design is reasonable. In order to measure the flight load, the strain bridge is first modified on some typical sections of the aircraft, and then the ground load calibration test is carried out. By analyzing the corresponding relationship between the calibration load and the response of the strain bridge, the load model is established by using multiple linear regression. Finally, the actual flight strain bridge response data is substituted into the load model to obtain the...

AI Technical Summary

Problems solved by technology

The disadvantage of this method is that when solving the coefficient matrix, the least squares method is used, and the coefficient matrix may not be solved due to the irreversibility of the individual matrix, which makes this method unable to realize the actual measurement of the flight load; because the assumption is made by The optimal load model established by n-1 strain bridges must be obtained from the optimal load model established by n strain bridges, resulting in a low measurement accuracy of the load model established by this method for flight loads

Images