Defect Recognition in Concrete Ultrasonic Detection Based on Wavelet Packet Transform and Stochastic Configuration Networks

Jinhui Zhao,Qichun Zhang,Tianyu Hu,Penghui Ba,Ruifang Zheng,Congli Mei

doi:10.1109/access.2021.3049448

Abstract

Aiming to detect concrete defects, we propose a new identification method based on stochastic configuration networks. The presented model has been trained by time-domain and frequency-domain features which are extracted from filtering and decomposing ultrasonic detection signals. This method was applied to ultrasonic detection data collected from 5 mm, 7 mm, and 9 mm penetrating holes in C30 class concrete. In particular, wavelet packet transform (WPT) was then used to decompose the detected signals, thus the information in different frequency bands can be obtained. Based on the data from the fundamental frequency nodes of the detection signals, we calculated the means, standard deviations, kurtosis coefficients, skewness coefficients, and energy ratios to characterize the detection signals. We also analyzed their typical statistical features to assess the complexity of identifying these signals. Finally, we used the stochastic configuration networks (SCNs) algorithm to embed four-fold cross-validation for constructing the recognition model. Based upon the experimental results, the performance of the presented model has been validated and compared with the genetic algorithm based on BP neural network model, where the comparison shows that the SCNs algorithm has superior generalization abilities, better fitting abilities, and higher recognition accuracy for recognizing defect signals. In addition, the test and analysis results show that the proposed method is feasible and effective in detecting concrete hole defects.

Highlights

Concrete is essential material in modern architecture
The existing manual recognition methods are low efficient and the traditional intelligent recognition methods have low accuracy. It is a difficult and important task to extract the effective features of the complex signal so that we can identify the type of concrete ultrasonic detection signal faster with high accuracy
The method uses wavelet packet transform to extract the main frequency node of the detection signal which leads to five types of statistics as feature vectors

Summary

INTRODUCTION

Concrete is essential material in modern architecture. it is widely used in the construction of many facilities, including buildings, bridges, and dams. The existing manual recognition methods are low efficient and the traditional intelligent recognition methods have low accuracy It is a difficult and important task to extract the effective features of the complex signal so that we can identify the type of concrete ultrasonic detection signal faster with high accuracy. Traditional machine learning algorithms such as neural network and support vector machine have been used in concrete ultrasonic detection signal recognition cases [7]–[9]. These algorithms are not of sufficient accuracy for identifying complex concrete defect signals. Experimental results showed that the method described in this article can effectively identify defective concrete ultrasonic detection signals.

BASIC STEPS AND ALGORITHM FLOWCHART FOR THE PROPOSED ALGORITHM

ULTRASONIC SIGNAL FEATURES SELECTION

STOCHASTIC CONFIGURATION NETWORKS

RUNNING AND ANALYSING THE PROPOSED ALGORITHM

Findings

CONCLUSION AND FUTURE WORKS