Random signal identification method based on multilayer neural network and support vector machine

Abstract

The invention relates to a random signal identification method based on a multilayer neural network and a support vector machine. The method comprises the following steps: 1) establishing a finite element model; 2) determining a response point position, an excitation input position and the amplitude of an acceleration excitation power spectrum; 3) carrying out random response analysis, and modifying the finite element model to obtain a finite element model closest to the actual situation; 4) designing different acceleration excitation power spectrums, and generating a plurality of groups of data sets; 5) completing data processing; 6) dividing the data set into a test set and a training set, and training the random signal prediction model by using a support vector machine and a multilayer neural network; 7) based on training results of the multi-layer neural network and the support vector machine, modifying parameters to find an optimal random signal prediction model; 8) based on the optimal random signal prediction model, adding noise, and verifying the generalization of the random signal prediction model. According to the method, the prediction precision and generalization of the multi-layer neural network are improved.

Description

technical field [0001] The invention belongs to the technical field of random signal recognition in the frequency domain, and relates to a random signal recognition method based on a multilayer neural network and a support vector machine. Background technique [0002] With the continuous development of the aerospace industry, the requirements for the strength of engineering structures are constantly increasing. In order to ensure the reliability and stability of the structure of equipment parts during operation and avoid resonance problems, designers need to accurately understand the random signals acting on the outside of the structure, but in many cases the outside of the structure It is very difficult to obtain random signals, and the measurement accuracy is poor, or even unmeasurable. For example, the random signal received by the shell during product transportation, and the random signal received by the shell during the high-altitude operation of the aircraft. Indirect...

AI Technical Summary

Problems solved by technology

[0003] The purpose of the present invention is to provide a random signal recognition method based on a multilayer neural network and a support vector machine, which solves the problem of few recognition methods, low recognition accuracy, simple recognition structure, and poor recognition principle in the process of random signal recognition in the prior art. complicated question

Images