Method, system and medium for damage location based on Lamb wave and neural network

Abstract

The invention discloses a method, system and storage medium for damage location based on Lamb wave and neural network. This method does not need to extract signal feature values ​​as the input of neural network. Considering that the main advantage of convolutional neural network is to process images Data, image data can be used as the input to get the best results. By converting the original waveform signal of the sensor into a multi-channel pseudo-color image as the input of the neural network, and directly taking the damage position coordinates as the output of the neural network, the original waveform is preserved. The effective information in the signal improves the accuracy of the positioning results, and can directly output the location of the damage. At the same time, it does not need to optimize the design of the sensor layout scheme in advance, and can realize automatic and intelligent damage positioning and identification. More importantly, by using multiple frequency excitation signals, more abundant damage diagnosis information can be obtained, further improving the accuracy of detection results.

Description

technical field [0001] The invention relates to the technical field of damage location and identification of plate structures, in particular, to a method and system for damage location based on Lamb waves and neural networks, and a computer-readable storage medium. Background technique [0002] With the increasing attention to the safety of aircraft structures and large-scale infrastructure structures, structural health monitoring technology has attracted more and more attention. The plate structure is widely used in various industrial structures, especially in the field of aerospace, such as space station cabins, aircraft doors, etc. Affected by factors such as manufacturing process, working load, and operating environment, these structures will inevitably have defects such as cracks, corrosion, and holes, which will affect the safe operation of the structure, and even cause major safety accidents and casualties. Therefore, in order to ensure the safe operation of these im...

AI Technical Summary

Problems solved by technology

[0005] The present invention provides a method and system for damage location based on Lamb wave and neural network, and a computer-readable storage medium to solve the signal characteristic value existing in the existing damage identification method based on Lamb wave and neural network It is difficult to extract, the algorithm is complicated, and a lot of effective information in the original waveform is lost, resulting in poor detection accuracy, and the technical problems of sensor layout need to be optimized in advance

Images