Piezoelectric Active Sensor Self-Diagnosis for Electromechanical Impedance Monitoring Using K-Means Clustering Analysis and Artificial Neural Network

Xie Jiang,Xin Zhang,Yuxiang Zhang,Dario Richiedei

doi:10.1155/2021/5574898

Abstract

Piezoelectric sensor is a crucial part of electromechanical impedance technology whose state will directly affect the effectiveness and accuracy of structural health monitoring (SHM). So carrying out sensor self-diagnosis is important and necessary. However, it is still difficult to distinguish sensor faults from structural damage as well as identify the cases and degrees of sensor faults. In the study, three characteristic indexes of admittance which have different indication intervals for damages of structure and sensors were selected from six indexes after comparison. To improve the discrimination effect, three principal components (PC) were extracted by principal component analysis (PCA). And the damage information represented by PCs was clustered by the K-means algorithm to identify the cases of damage. Then, the degrees of sensor damages were classified with the artificial neural network (ANN). The results show that the K-means clustering analysis based on admittance characteristics can accurately distinguish and identify the structural damage and four kinds of sensor damages, namely, pseudosoldering, debonding, wear, and breakage. The trained ANN model has a good recognition effect on the damage degrees and the accuracy of recognition reaches 100%. This study has a certain reference value for piezoelectric sensor self-fault identification.

Highlights

Electromechanical impedance technology using piezoelectric materials has aroused extensive attention in structural health monitoring (SHM) [1, 2]
The accurate recognition of Piezoelectric lead zirconate titanate (PZT) fault is the key to improve the ability of long-term SHM based on the electromechanical impedance method. is paper proposes a PZT self-diagnosis method based on Kmeans clustering analysis and artificial neural network (ANN) given in the existing study cannot achieve the intelligent identification and evaluation of the cases and degrees of sensor faults
By clustering the admittance spectrum characteristics under different damages, the PZT fault can be distinguished from structural damage, and the damage types of PZT can be classified and identified. en we used the ANN trained model to realize the accurate evaluation of the damage degree. e specific conclusions are as follows: (1) Combined with the characteristics of conductance and susceptance, six damage indexes are used to extract the signal feature under different damage conditions

Summary

Introduction

Electromechanical impedance technology using piezoelectric materials has aroused extensive attention in structural health monitoring (SHM) [1, 2]. There are many methods combining electromechanical impedance and data mining algorithms that have been used to solve practical application problems. Park et al [14] verified the effectiveness of the method which incorporates the principal component analysis (PCA) and K-means clustering in the practical application of electromechanical impedance-based wireless SHM system. Most of the existing monitoring systems are not intelligent enough to distinguish the signal changes caused by structural damage and sensor faults. En the K-means algorithm was used to cluster different cases of damages represented by the PCs. the degrees of four PZT damages were identified by the ANN model

Electromechanical Impedance-Based Damage Detection Method

Experimental Investigation on EMI-Based Sensor Self-Diagnosis

Serial number

Clustering Analysis of Damage Data Based on K-Means Algorithm

Findings

Summary and Conclusions