Wireless Power Transfer System Testing: Optimization & Quality Assessment Using ML
(2024) EITM01 20241Department of Electrical and Information Technology
- Abstract
- Wireless Power Transfer (WPT) is increasingly utilized across various practical domains, requiring continual development and optimization of supporting technologies. An essential aspect of this progress is the efficient testing of WPT Systems. This thesis explores optimization of the testing process using Machine Learning(ML) and data analysis, applied on historical data represented by test outcomes of different products, known as test reports.
The research focuses on three primary objectives: product fault categorization, test-quality analysis and test failure prioritization. Product fault categorization employs clustering methods to find common groups of failed tests, in the hope it captures underlying product fault. Test-quality... (More) - Wireless Power Transfer (WPT) is increasingly utilized across various practical domains, requiring continual development and optimization of supporting technologies. An essential aspect of this progress is the efficient testing of WPT Systems. This thesis explores optimization of the testing process using Machine Learning(ML) and data analysis, applied on historical data represented by test outcomes of different products, known as test reports.
The research focuses on three primary objectives: product fault categorization, test-quality analysis and test failure prioritization. Product fault categorization employs clustering methods to find common groups of failed tests, in the hope it captures underlying product fault. Test-quality analysis involves examining measured value trends to develop a more nuanced performance metrics and the conversion of the values to useful features for the test failure prioritization. Test failure prioritization employs predictive models and optimal static orders to expedite the running of tests which will fail during product testing.
Critical components in implementing the techniques involve data preparation, feature extraction and innovative new adaptions of existing algorithms to accommodate limitation in data quantity.
The results demonstrate significant improvements in testing efficiency, with theaverage number of tests conducted before detecting a failure decreasing from 51.6 to 8 through the application of ML on historical data. Additionally, the locality of failures across a 0-100 scoring system improved from a average 37.8 to 79.5 when using optimal static ordering. Although clustering and test-quality analysis yielded promising results with discernible value trends, additional validation is necessary to establish their full utility. (Less) - Popular Abstract
- Wireless charging technology, much like the charging of our mobiles devices, is becoming both convenient and increasingly common. As this technology grows more widespread, the demand for products that utilize it continues to rise, leading to more extensive testing requirements. With greater testing needs comes an increase in the time and resources required for product development. This thesis aims to address the issue by introducing advanced computational techniques, such as Machine Learning (ML), to make the testing process quicker and more efficient while providing qualitative insight into the testing process.
Wireless power charging relies on Wireless Power Transfer (WPT) systems, which are typically developed and refined through... (More) - Wireless charging technology, much like the charging of our mobiles devices, is becoming both convenient and increasingly common. As this technology grows more widespread, the demand for products that utilize it continues to rise, leading to more extensive testing requirements. With greater testing needs comes an increase in the time and resources required for product development. This thesis aims to address the issue by introducing advanced computational techniques, such as Machine Learning (ML), to make the testing process quicker and more efficient while providing qualitative insight into the testing process.
Wireless power charging relies on Wireless Power Transfer (WPT) systems, which are typically developed and refined through performance testing using specialized tools. The historical data generated from these tests, captured in detailed test reports, serves as a valuable resource for both data analysis and ML. By applying ML and data analysis techniques to this data, we can enhance and optimize the testing process, making it more efficient while also providing deeper insights into system performance. The study focuses on three key improvements:
• Grouping Tests: By analyzing patterns in the test result from various reports, it is possible to identify common failed tests and reason about practical issues with the products that generated those reports.
• Quality Analysis: By examining trends in the test data, the study identifies good performing products as well as which measurement results typically indicate good performance.
• Test Efficiency: Models and test orders capable of prioritizing tests that are more likely to conclude in failures have been implemented. This means the number of tests that are run until a failure is identified can be reduced, allowing more time to be spent on fixing these issues.
The results demonstrate that ML significantly reduced the average number of tests required to find a failure from 51.6 tests to just 8 on historical data. Additionally, test ordering could be improved from a score of 37.8 to 79.5 on a 0-100 scale where higher scores indicate a more efficient identification of early failures. The test grouping and quality analysis also produces promising results, revealing clear patterns in failure groupings and measurement results. Overall the study demonstrate that these techniques can lead to faster testing and qualitative insights into product performance, ultimately speeding up WPT system development. (Less)
Please use this url to cite or link to this publication:
http://lup.lub.lu.se/student-papers/record/9176338
- author
- Nordqvist, Sebastian LU
- supervisor
- organization
- course
- EITM01 20241
- year
- 2024
- type
- H2 - Master's Degree (Two Years)
- subject
- keywords
- WPT Wireless Power Transfer TPR Test Power Receiver TPT Test Power Transmitterr PUD Product Under Test RX Receiver TX Transmitter WPC Wireless Power Consortium ML Machine Learning AI Artificial Intelligence MDS Multidimensional Scaling FOD Foreign Object Detection
- report number
- LU/LTH-EIT 2024-1022
- language
- English
- id
- 9176338
- date added to LUP
- 2024-10-16 11:04:43
- date last changed
- 2024-10-16 11:04:43
@misc{9176338,
abstract = {{Wireless Power Transfer (WPT) is increasingly utilized across various practical domains, requiring continual development and optimization of supporting technologies. An essential aspect of this progress is the efficient testing of WPT Systems. This thesis explores optimization of the testing process using Machine Learning(ML) and data analysis, applied on historical data represented by test outcomes of different products, known as test reports.
The research focuses on three primary objectives: product fault categorization, test-quality analysis and test failure prioritization. Product fault categorization employs clustering methods to find common groups of failed tests, in the hope it captures underlying product fault. Test-quality analysis involves examining measured value trends to develop a more nuanced performance metrics and the conversion of the values to useful features for the test failure prioritization. Test failure prioritization employs predictive models and optimal static orders to expedite the running of tests which will fail during product testing.
Critical components in implementing the techniques involve data preparation, feature extraction and innovative new adaptions of existing algorithms to accommodate limitation in data quantity.
The results demonstrate significant improvements in testing efficiency, with theaverage number of tests conducted before detecting a failure decreasing from 51.6 to 8 through the application of ML on historical data. Additionally, the locality of failures across a 0-100 scoring system improved from a average 37.8 to 79.5 when using optimal static ordering. Although clustering and test-quality analysis yielded promising results with discernible value trends, additional validation is necessary to establish their full utility.}},
author = {{Nordqvist, Sebastian}},
language = {{eng}},
note = {{Student Paper}},
title = {{Wireless Power Transfer System Testing: Optimization & Quality Assessment Using ML}},
year = {{2024}},
}