GA-BP neural network used for knock detection of aviation composite materials

In order to better solve many problems such as low detection accuracy, high detection cost and low detection efficiency of aviation composite materials. This paper proposes an intelligent tapping nondestructive testing method, and establishes a GA-BP neural network optimization model. Data collection is performed by tapping the sensor, and feature factor extraction is performed on the collected data. The extracted feature quantity is used as the input of the model. After training and testing, the accuracy of model detection can reach 95%. The results show that the knock detection method optimized based on the GA-BP neural network model can meet the detection accuracy requirements of aviation composite materials. The introduction of the algorithm model provides a new method for tapping data processing, and greatly improves the detection accuracy, which has high research and application value.