[{"data":1,"prerenderedAt":-1},["ShallowReactive",2],{"doc-detail-125233-en":3,"doc-seo-125233-105":30,"detail-sidebar-cat-0-en-105":91},{"code":4,"msg":5,"data":6},0,"success",{"doc_id":7,"user_id":8,"nickname":9,"user_avatar":10,"doc_module":4,"category_id":11,"category_name":12,"doc_title":13,"doc_description":14,"doc_content":15,"file_id":16,"file_url":17,"file_type":18,"file_size":19,"view_count":4,"is_deleted":4,"is_public":20,"is_downloadable":20,"audit_status":20,"page_count":21,"language":22,"language_code":23,"site_id":24,"html_lang":23,"table_of_contents":25,"faqs":26,"seo_title":27,"seo_description":14,"update_tm":28,"read_time":29},125233,1099514067438,"River Wang","https://ap-avatar.wpscdn.com/avatar/100002539ee87300030?x-image-process=image/resize,m_fixed,w_180,h_180&k=1780474512215547542",6,"Technology","A Machine Learning Approach to Ultrasonic Testing Signal Degradation","The presentation addresses signal degradation in ultrasonic non-destructive testing and its impact on layer thickness measurements. It explains how probe-emitted ultrasound reflections are used to infer material thickness, while external factors such as water-frequency requirements and material roughness can cause destructive interference and unreliable interface echoes. The work proposes machine learning reconstruction using high-quality signals for training, evaluates A-scan time-series data, applies a structured pre-processing and classification workflow, and analyzes results through reflection patterns and cluster distributions to support accurate thickness determination.","A Machine Learning Approach to Ultrasonic Testing Signal Degradation  \nAbigail Brown  \nComputer Science & Data Science B.A. Student Pacific Northwest National Laboratory Intern  \nProblem Formulation  \n• Non-Destructive Testing – Ultrasonic Testing  \n􀂃 A probe emits a high-frequency sound wave onto a plate submerged under water and records the amount of time it takes for the wave to reflect back  \n• Ultrasound reflections can be used to measure the thickness of layers of materials in a plate  \n• The accuracy of layer thickness measurements can be incorrectly measured due to various external factors  \n􀂃 Water vs Frequency (standard operating procedure – requires certain frequency)  \n􀂃 Smoothness of material (causes destructive interference)  \n• Incorrect accuracy measurements factor into the acceptance tolerance of the plate and can result in the plate being deemed unusable  \n2  \nLayer Thickness Measurement Accuracy (Implications of a Bad Signal)  \n• A major challenge in detection lies in the interface echo (black bracket) occasionally not being a well-defined wave, which introduces uncertainty in signal detection  \n• The thickness of the plate has very little margin of error:  \n􀂃 A mismeasurement of ½ a wavelength at 50 MHz (red dot to blue dot) results in an error of 2.5 mils (64 µm)  \n• Proper thickness classification is vital so that no material is wasted; mis-recorded scans may result in an erroneous thickness measurement, resulting in material waste  \nDescription and Image taken from Jacobs, R. E. (2024) . Ultrasonic Testing Techniques [PowerPoint Slides] . Non-Destructive Evaluation, Pacific Northwest National Laboratory.  \n3  \nMain Goal/Approach  \n• Use ML models toreconstruct bad-wavescans (due to destructive interference caused by rough interfaces of materials)  \n• Train using good signals for reconstruction models  \n• Predict at the start of an echo of interest  \n4  \nDescription of the Data  \n• A-Scans: A one-dimensional graph plotting the signal amplitude against time  \n• B-Scans: A two-dimensional view depicting material thicknesses scanned at different points over time  \n• C-Scans: A three-dimensional view depicting material thickness plotted over positions both vertically and horizontally  \n• We are working with A-Scans and developing an algorithm that measures the thickness of the layers of material based on how long it took for the signal to reflect back  \nDescription and Images taken from Gecko Marketing Team. “A-Scan, B-Scan, and C-Scan Ultrasonic Data from Robotic Inspections.” Gecko, September 29, 2021. [https://blog.geckorobotics.com/unpacking](https://blog.geckorobotics.com/unpacking)  \na-scans-b-scans-and-c-scans-in-robotic-ultrasonic-inspection. 5  \nHow the Data is Collected  \n• A probe will move along each x, y coordinate on the plate  \n• The probe will take measurements every millisecond from 0 – 333 milliseconds, making time our z coordinate  \nExploring the Data – An Excerpt from the Dataset  \nA-Scan Analysis  \n• Looking at amplitudes over time for an a-scan  \n􀂃 One-Dimensional Time-Series Data  \n• Roughly 1.6 million points in our sampled data  \n• The length of time between reflections determines the thickness of the layer of material  \n7  \nDifferent Categories of Scans  \n• The following line plots depict each category of scan:  \n􀂃 ‘perfect’ scan (top left), ‘degraded’ scan (top right), ‘cladding’ scan (bottom left), ‘water’scan (bottom right)  \n8  \nPre-Processing Procedure  \n1. Exponentially smooth the data in order to reduce the noise in the plots  \n2. Count the numbers of peaks and negative peaks in a given amplitude sequence  \n3. Count the number of reflections of major peaks  \n4. Find the center of each reflection and ensure the centers for each scan are where they are expected to be  \n􀂃 Cladding-only scans will be missing a reflection around 180 milliseconds  \n5. Classify as fuel, cladding, or water based off reflection location  \n1. Pass fuel scans through a Fourier Transform algorithm  \n2. Lo","cbCaihHzj3XDXyZK","https://ap.wps.com/l/cbCaihHzj3XDXyZK","pdf",1176308,1,13,"English","en",105,"# Problem Formulation\n## Non-Destructive Testing – Ultrasonic Testing\n# Layer Thickness Measurement Accuracy\n## Implications of a Bad Signal\n# Main Goal/Approach\n# Description of the Data\n## A-Scans, B-Scans, C-Scans\n# How the Data is Collected\n# Exploring the Data\n## A-Scan Analysis\n# Different Categories of Scans\n# Pre-Processing Procedure\n# Results","[{\"question\":\"Why does signal degradation matter in ultrasonic testing?\",\"answer\":\"Degraded interface echoes can become poorly defined, introducing uncertainty in signal detection. Because thickness measurement has a very small allowable error margin, incorrect measurements can cause plates to fail acceptance tolerance and become unusable.\"},{\"question\":\"What is the main ML approach proposed in the presentation?\",\"answer\":\"Train machine learning reconstruction models using good signals to reconstruct degraded wave scans caused by destructive interference from rough material interfaces. The approach focuses on predicting the start of the echo of interest.\"},{\"question\":\"How does the pre-processing and classification workflow work?\",\"answer\":\"The method exponentially smooths the A-scan data, counts peaks and negative peaks, counts major reflections, and locates reflection centers to confirm expected timing. It then classifies scans into fuel, cladding, or water based on reflection location, using additional Fourier transform steps for fuel and identifying degradation via dips near reflection centers.\"}]","A Machine Learning Approach to Ultrasonic Testing Signal Degradation | PDF",1785897633,33,{"code":4,"msg":31,"data":32},"ok",{"site_id":24,"language":23,"slug":33,"title":13,"keywords":34,"description":14,"schema_data":35,"social_meta":86,"head_meta":88,"extra_data":90,"updated_unix":28},"a-machine-learning-approach-to-ultrasonic-testing-signal-degradation","",{"@graph":36,"@context":85},[37,54,68],{"@type":38,"itemListElement":39},"BreadcrumbList",[40,44,48,51],{"item":41,"name":42,"@type":43,"position":20},"https://docshare.wps.com","Home","ListItem",{"item":45,"name":46,"@type":43,"position":47},"https://docshare.wps.com/document/","Document",2,{"item":49,"name":12,"@type":43,"position":50},"https://docshare.wps.com/document/technology/",3,{"item":52,"name":13,"@type":43,"position":53},"https://docshare.wps.com/document/a-machine-learning-approach-to-ultrasonic-testing-signal-degradation/125233/",4,{"url":52,"name":13,"@type":55,"author":56,"headline":13,"publisher":58,"fileFormat":61,"inLanguage":23,"description":14,"dateModified":62,"datePublished":62,"encodingFormat":61,"isAccessibleForFree":63,"interactionStatistic":64},"DigitalDocument",{"name":9,"@type":57},"Person",{"url":41,"name":59,"@type":60},"DocShare","Organization","application/pdf","2026-08-05",true,{"@type":65,"interactionType":66,"userInteractionCount":4},"InteractionCounter",{"@type":67},"ViewAction",{"@type":69,"mainEntity":70},"FAQPage",[71,77,81],{"name":72,"@type":73,"acceptedAnswer":74},"Why does signal degradation matter in ultrasonic testing?","Question",{"text":75,"@type":76},"Degraded interface echoes can become poorly defined, introducing uncertainty in signal detection. Because thickness measurement has a very small allowable error margin, incorrect measurements can cause plates to fail acceptance tolerance and become unusable.","Answer",{"name":78,"@type":73,"acceptedAnswer":79},"What is the main ML approach proposed in the presentation?",{"text":80,"@type":76},"Train machine learning reconstruction models using good signals to reconstruct degraded wave scans caused by destructive interference from rough material interfaces. The approach focuses on predicting the start of the echo of interest.",{"name":82,"@type":73,"acceptedAnswer":83},"How does the pre-processing and classification workflow work?",{"text":84,"@type":76},"The method exponentially smooths the A-scan data, counts peaks and negative peaks, counts major reflections, and locates reflection centers to confirm expected timing. It then classifies scans into fuel, cladding, or water based on reflection location, using additional Fourier transform steps for fuel and identifying degradation via dips near reflection centers.","https://schema.org",{"og:url":52,"og:type":87,"og:title":13,"og:site_name":59,"og:description":14},"article",{"robots":89,"canonical":52},"index,follow",{"doc_id":7,"site_id":24},{"code":4,"msg":5,"data":92},[93,97,101,105,110,113,118,123,128,131,135],{"id":20,"doc_module":4,"doc_module_name":46,"category_name":94,"show_sort_weight":95,"slug":96},"Story & Novel",90,"story-novel",{"id":47,"doc_module":4,"doc_module_name":46,"category_name":98,"show_sort_weight":99,"slug":100},"Literature",80,"literature",{"id":53,"doc_module":4,"doc_module_name":46,"category_name":102,"show_sort_weight":103,"slug":104},"Exam",70,"exam",{"id":106,"doc_module":4,"doc_module_name":46,"category_name":107,"show_sort_weight":108,"slug":109},5,"Comic",60,"comic",{"id":11,"doc_module":4,"doc_module_name":46,"category_name":12,"show_sort_weight":111,"slug":112},50,"technology",{"id":114,"doc_module":4,"doc_module_name":46,"category_name":115,"show_sort_weight":116,"slug":117},7,"Healthcare",40,"healthcare",{"id":119,"doc_module":4,"doc_module_name":46,"category_name":120,"show_sort_weight":121,"slug":122},8,"Research & Report",30,"research-report",{"id":124,"doc_module":4,"doc_module_name":46,"category_name":125,"show_sort_weight":126,"slug":127},9,"Religion & Spirituality",20,"religion-spirituality",{"id":126,"doc_module":4,"doc_module_name":46,"category_name":129,"show_sort_weight":126,"slug":130},"World Cup","world-cup",{"id":132,"doc_module":4,"doc_module_name":46,"category_name":133,"show_sort_weight":132,"slug":134},10,"Lifestyle","lifestyle",{"id":136,"doc_module":4,"doc_module_name":46,"category_name":137,"show_sort_weight":106,"slug":138},19,"General","general"]