Fractal elimination method for mass data in ultrasonic intelligent detection

Abstract

The invention relates to a fractal elimination method for massive data in ultrasonic intelligent detection. In the long-term, long-distance ultrasonic automatic detection and ultrasonic array automatic detection, a large amount of test data will be generated, and the amount of test data is very large, which makes it very difficult to store the data in real time. In view of the fact that most of the test data are defect-free signals, the present invention proposes to eliminate defect-free ultrasonic signals and store only defective ultrasonic echo signals, thereby solving the technical problem of data storage. For this reason, the storage problem of massive data is attributed to the binary identification problem. In the present invention, the fractal R / S algorithm of limited noisy sequence based on fractal technology is used to extract the characteristic vector (Hurst index) of the signal, and a fractal linearity index to measure the quality of the fractal characteristic is proposed, and the two characteristic distributions are used to distinguish Signal. A large number of experiments show that the non-defective signal has good fractal characteristics, and its Hurst exponent is concentrated in a small interval; while the defective signal has no obvious fractal characteristics, the fitted fractal parameters have poor linearity, and its Hurst exponent is Larger spread. The test results also show that the method is easy to implement, has good real-time performance, is less affected by noise, and has high judgment accuracy and stability.

Description

technical field [0001] The invention relates to a fractal elimination method for massive data in ultrasonic intelligent detection, and is a processing method especially suitable for generating massive detection data during long-time and long-distance ultrasonic automatic detection and ultrasonic array automatic detection. It can accurately and quickly distinguish the types of data, eliminate useless parts of the data, and retain useful data, which can greatly reduce the amount of stored data and solve the current technical bottleneck of data storage. Background technique [0002] Ultrasonic intelligent detection, automatic detection and multi-sensor array detection have a broad application market, and these test processes are often carried out for a long time and over a long distance (or large area), requiring the test system to be able to effectively record test data for post-analysis and analysis. Evaluate. Because the ultrasonic waves used for detection often have a high...

AI Technical Summary

Problems solved by technology

However, how to make the detection system autonomously and in real time determine whether the signal is defective or not without human participation is a technical problem

However, there are very few reports on fractal algorithms and their applications for ultrasonic signal processing.

Images