Complex structure damage cooperative identification method based on ultrasonic guided-wave

Abstract

The invention relates to a complex structure damage cooperative identification method based on ultrasonic guided-wave. The method comprises a preliminary identification mechanism and a feedback decision mechanism. The preliminary identification mechanism is used for realizing preliminary judgment of damage types of various complex structures; and the feedback decision mechanism is used for realizing optimization of the damage identification result, and thus recognition rate is effectively improved. In the preliminary identification mechanism, damage signals are comprehensively analyzed in the aspects of time domain, frequency domain and detail features of the signals by utilizing a characteristic quantity extracting method for ultrasonic guided-wave characteristics to obtain corresponding characteristic quantity; and a three-layer "input unit layer-order parameter layer-output unit layer" damage cooperative identification network model is constructed. The feedback decision mechanism is introduced on the basis of complex structure damage primary recognition; characteristic quantity assessment based on within-class among-class mean distance is adopted, and damage characteristic quantity most suitable for cooperative training is selected; and prototype characteristic vector reconstruction is carried out by utilizing an information superposition method, and thus error identification rate is effectively reduced.

Description

technical field [0001] The invention relates to the technical field of non-destructive testing, in particular to a method for cooperative identification of damage to complex structures based on ultrasonic guided waves. Background technique [0002] In various fields such as transportation, transmission lines, bridges and housing construction, it is often necessary to use various complex long-shaped structures, such as stranded wires, pipes, steel wire ropes, etc. These complex structures are often affected by the external environment and bearing capacity for a long time in the work. The influence of changes, resulting in various damages during use, such as wear, rust, broken strands, etc., these damages will endanger the safety of personnel and equipment and property. Different damage will cause different effects, and each type of damage also has its corresponding treatment plan, which requires the detection system to be able to quickly and effectively classify and identify ...

AI Technical Summary

Problems solved by technology

However, due to the complex external working environment where the complex structure is located, there is often a lot of noise, the extreme conditions and the complex structure will bring great interference to the detection signal of the ultrasonic guided wave

At the same time, since there is no clear boundary between the extracted eigenvalue classes and classes in the received signal of the ultrasonic guided wave, misidentification often inevitably occurs.

Images