A deep learning approach to impact localization and uncertainty assessment in CFRP composites using sparse PZTs: Integrating experiments and simulations
Huai Yan, Weihua Xie, Bo Gao, Fan Yang, Songhe Meng
{"title":"A deep learning approach to impact localization and uncertainty assessment in CFRP composites using sparse PZTs: Integrating experiments and simulations","authors":"Huai Yan, Weihua Xie, Bo Gao, Fan Yang, Songhe Meng","doi":"10.1016/j.tws.2025.113143","DOIUrl":null,"url":null,"abstract":"<div><div>The propagation of elastic waves in CFRP composites is dispersive and multimodal, making impact localization using wave signals challenging. An end-to-end deep learning model with an encoder-decoder architecture was developed for impact localization in CFRP composites, using sparse piezoelectric ceramic transducer (PZT) arrays and a data-driven approach to enable online impact sensing. The model used a segmented training strategy and transfer learning to capture shared features between experimental and simulated data. According to the piezoelectric equation, it links experimental piezoelectric signals with simulated stress responses. The results show that the feature encoder trained can extract their shared features effectively. Meanwhile, the model successfully applied the laws from simulation data to the localization of experimental impacts based on the fine-tuning strategy, alleviating the challenge of limited experimental data. The prediction results in the test set show that good generalization, and impact localization can be achieved in milliseconds during inference. The average error in localization is only 3.42 mm over a 100 mm × 100 mm monitoring area. Compared to the traditional transfer strategy, the three-stage training proposed shows better generalization by constraining the encoded features. In addition, the Monte-Carlo dropout strategy is used to assess prediction uncertainty, analyzing the effects of dropout rate and repeated predictions on confidence intervals. The study provides a prospective solution for large CFRP structures to achieve fast localization of impacts under sparse PZT arrays.</div></div>","PeriodicalId":49435,"journal":{"name":"Thin-Walled Structures","volume":"212 ","pages":"Article 113143"},"PeriodicalIF":5.7000,"publicationDate":"2025-03-10","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":"0","resultStr":null,"platform":"Semanticscholar","paperid":null,"PeriodicalName":"Thin-Walled Structures","FirstCategoryId":"5","ListUrlMain":"https://www.sciencedirect.com/science/article/pii/S026382312500237X","RegionNum":1,"RegionCategory":"工程技术","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":null,"EPubDate":"","PubModel":"","JCR":"Q1","JCRName":"ENGINEERING, CIVIL","Score":null,"Total":0}
引用次数: 0
Abstract
The propagation of elastic waves in CFRP composites is dispersive and multimodal, making impact localization using wave signals challenging. An end-to-end deep learning model with an encoder-decoder architecture was developed for impact localization in CFRP composites, using sparse piezoelectric ceramic transducer (PZT) arrays and a data-driven approach to enable online impact sensing. The model used a segmented training strategy and transfer learning to capture shared features between experimental and simulated data. According to the piezoelectric equation, it links experimental piezoelectric signals with simulated stress responses. The results show that the feature encoder trained can extract their shared features effectively. Meanwhile, the model successfully applied the laws from simulation data to the localization of experimental impacts based on the fine-tuning strategy, alleviating the challenge of limited experimental data. The prediction results in the test set show that good generalization, and impact localization can be achieved in milliseconds during inference. The average error in localization is only 3.42 mm over a 100 mm × 100 mm monitoring area. Compared to the traditional transfer strategy, the three-stage training proposed shows better generalization by constraining the encoded features. In addition, the Monte-Carlo dropout strategy is used to assess prediction uncertainty, analyzing the effects of dropout rate and repeated predictions on confidence intervals. The study provides a prospective solution for large CFRP structures to achieve fast localization of impacts under sparse PZT arrays.
期刊介绍:
Thin-walled structures comprises an important and growing proportion of engineering construction with areas of application becoming increasingly diverse, ranging from aircraft, bridges, ships and oil rigs to storage vessels, industrial buildings and warehouses.
Many factors, including cost and weight economy, new materials and processes and the growth of powerful methods of analysis have contributed to this growth, and led to the need for a journal which concentrates specifically on structures in which problems arise due to the thinness of the walls. This field includes cold– formed sections, plate and shell structures, reinforced plastics structures and aluminium structures, and is of importance in many branches of engineering.
The primary criterion for consideration of papers in Thin–Walled Structures is that they must be concerned with thin–walled structures or the basic problems inherent in thin–walled structures. Provided this criterion is satisfied no restriction is placed on the type of construction, material or field of application. Papers on theory, experiment, design, etc., are published and it is expected that many papers will contain aspects of all three.