Based on electromechanical impedance (EMI) technology, we studied the indication effect of each characterization index on flange bolt looseness under different noise levels. A new evaluation index T is constructed to solve the problem that the minor looseness and noise under the actual monitoring environment cannot be distinguished with existing indexes. The results show that among the five characterization indexes, the root mean square deviation (RMSD) and the correlation coefficient deviation (CCD) can reflect the minor looseness sensitively and are positively correlated with the degree of looseness. They also have good performance of looseness identification under the noise of 30 dBW. However, the indexes’ indication range of the case under 1 dBW-6 dBW noise overlaps with the case under 36 N·m-39.5 N·m torque. So it is impossible to judge the structure state with mere characterization indexes. Evaluation index T can extract the correlation difference of the conductance spectrum under the noise conditions and looseness conditions with the benchmark respectively. And there are different indication ranges of T on the noise and minor looseness. By determining a threshold and comparing T with it, whether bolt looseness occurs in an unknown condition can be effectively judged. Finally, T is applied to other flange structure with 8 bolts to verify the effectiveness and general applicability of the method. The evaluation index plays an important role in avoiding misjudgment of flange bolt caused by noise influence on minor looseness and improving the ability of identification when the bolt is at the critical state.
1 January 2018 in Intelligent Automation and Soft Computing
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.
The accuracy of structural state evaluation may be affected by the damaged piezoelectric sensors. Therefore, it is necessary to identify the sensor fault during monitoring. This paper proposes a method based on classification and regression tree (CART) and particle swarm optimization (PSO) to improve the efficiency of potential feature sets selection for sensor fault classification and build an identification model with the best performance. Firstly, the signal features of three structural changes and four sensor faults were extracted with five indexes. Then the decision trees (DT) for sensor fault classification were built based on different index combinations whose performances were then evaluated by the designed fitness function. Finally, PSO was used to optimize the searching for the best index combination. The results show that compared with the exhaustive method, adopting PSO for DT optimization can greatly simplify the search process. When the particle population is 5 and 10, the fitness converges to the optimal solution after only 6 and 4 iterations respectively. Although the DT with the best fitness is trained with only two indexes, its accuracy is higher than those trained with more indexes and the classification accuracy of 64 samples reaches 98.4% which shows the feasibility and practicability of the method.
1 January 2024 in Structural Control and Health Monitoring
The multiple intersecting stiffeners on the chemical milling stiffened panel (CMSP) limit the application of active health monitoring methods on it. An imaging algorithm based on electromechanical impedance (EMI) and probability‐weighting is proposed to achieve quantitative evaluation and localization of the damage on CMSP. The proposed algorithm compensates for the difference in sensor performance with coefficients and there is no need to determine the key parameters of the algorithm through prior experiments. In the paper, the applicability of ultrasonic guided wave (GW) and EMI on CMSP was first studied through the finite element method. Based on EMI and the mean absolute percentage deviation (MAPD), the selected damage indicator (DI), a probability‐weighted damage imaging algorithm are proposed and experimentally verified. The results indicate that due to the reflection and attenuation effects of stiffeners on GW, the signal characteristics of damage scattering waves are contaminated, making it difficult to quantitatively characterize the damage from GW signals through DIs. MAPD is positively correlated with the damage degree and has consistency in characterizing the signal of different PZTs under the same working condition. The feasibility and accuracy of the proposed algorithm are verified through experiments which show a strong engineering application capability.
8 May 2024 in IEEE Transactions on Instrumentation and Measurement
Mode conversion and wave scattering will occur when ultrasonic guided wave (GW) propagates to the stiffener which makes the received signals complex and poorly interpretable, thus limiting the application of GW in damage detection of chemical milling stiffened panel (CMSP). This article proposes a deep learning (DL) model inception-convolutional neural network (Inception-CNN) to realize damage localization of CMSP. First, a model analysis was conducted to get the resonant frequency of the piezoelectric wafer, and the mode conversion at the stiffener was explored through numerical analysis. Then, the identification effect based on the conventional damage imaging method was discussed. Finally, for training the proposed DL model, the residual signals were collected as a dataset after setting damages in different zones of CMSP. The model was then trained and tested, and its performance was analyzed and demonstrated. The results indicate that S0 mode has a greater conversion degree than A0 mode at the stiffener; GWs do not propagate in the form of a uniform wavefront on CMSP and conventional damage imaging methods based on the wave propagation paths are not applicable to CMSP. The model proposed can automatically extract the signal spatial features, accurately identify the corresponding damage zone, and its accuracy reaches 94% in the testing set. As the input features increase, the classification ability of the model will be further improved. The noise interference experiment shows that the model has good noise resistance performance at noise levels below 15%, indicating the feasibility of the model for practical applications.