Application of Convolutional Neural Network to Predict Anisotropic Effective Thermal Conductivity of Semiconductor Package
With increasing complexity of design patterns in semiconductor package substrates caused by demand for high-power semiconductors, it is necessary to be able to predict the thermal properties according to the pattern. Classifying the patterns is important to predict the effective thermal conductivity...
| Main Authors: | , , , , |
|---|---|
| Format: | Article |
| Language: | English |
| Published: |
IEEE
2022-01-01
|
| Series: | IEEE Access |
| Subjects: | |
| Online Access: | https://ieeexplore.ieee.org/document/9774413/ |
| _version_ | 1818203156880818176 |
|---|---|
| author | Tae-Hyun Kim Jeong-Hyeon Park Ki Wook Jung Jaechoon Kim Eun-Ho Lee |
| author_facet | Tae-Hyun Kim Jeong-Hyeon Park Ki Wook Jung Jaechoon Kim Eun-Ho Lee |
| author_sort | Tae-Hyun Kim |
| collection | DOAJ |
| description | With increasing complexity of design patterns in semiconductor package substrates caused by demand for high-power semiconductors, it is necessary to be able to predict the thermal properties according to the pattern. Classifying the patterns is important to predict the effective thermal conductivity (ETC), but it has some difficulties due to the variable setting being labor-intensive and creating human uncertainty. These difficulties are amplified by the complexity of the pattern in the printed circuit board (PCB) substrate. This work presents an automated convolutional neural network (CNN)-based algorithm to infer the anisotropic ETCs of package substrates. This algorithm divides a layer-pattern image of a PCB into local unit-cell images and learns the pattern characteristics of each unit cell to identify the local ETC. The algorithm then builds an ETC map by integrating the local ETCs for the entire layer. The entire process is fully automated to reduce human uncertainty and required workforce. The ETC map from the algorithm was then used in finite element (FE) analysis and compared with three other prediction methods. The proposed algorithm can predict the anisotropic ETCs within 2–3 % errors compared to the reference data while other models lead to at least 16 % error. The FE simulation with the ETC map of the algorithm can reflect the effect of the design pattern on the heat flux and temperature distributions on the package layer, leading to the lowest root mean square error in the temperature distribution compared to other models. |
| first_indexed | 2024-12-12T03:20:52Z |
| format | Article |
| id | doaj.art-ac7bb9d7508040fab698badb519b2e25 |
| institution | Directory Open Access Journal |
| issn | 2169-3536 |
| language | English |
| last_indexed | 2024-12-12T03:20:52Z |
| publishDate | 2022-01-01 |
| publisher | IEEE |
| record_format | Article |
| series | IEEE Access |
| spelling | doaj.art-ac7bb9d7508040fab698badb519b2e252022-12-22T00:40:11ZengIEEEIEEE Access2169-35362022-01-0110519955200710.1109/ACCESS.2022.31748829774413Application of Convolutional Neural Network to Predict Anisotropic Effective Thermal Conductivity of Semiconductor PackageTae-Hyun Kim0https://orcid.org/0000-0002-2238-9032Jeong-Hyeon Park1https://orcid.org/0000-0001-5305-3710Ki Wook Jung2Jaechoon Kim3Eun-Ho Lee4https://orcid.org/0000-0003-1270-7954Department of Smart Fab. Technology, Sungkyunkwan University, Suwon, Republic of KoreaDepartment of Mechanical Engineering, Sungkyunkwan University, Suwon, Republic of KoreaDivision of Test and System Package, Samsung Electronics, Hwaseong, Republic of KoreaDivision of Test and System Package, Samsung Electronics, Hwaseong, Republic of KoreaDepartment of Smart Fab. Technology, Sungkyunkwan University, Suwon, Republic of KoreaWith increasing complexity of design patterns in semiconductor package substrates caused by demand for high-power semiconductors, it is necessary to be able to predict the thermal properties according to the pattern. Classifying the patterns is important to predict the effective thermal conductivity (ETC), but it has some difficulties due to the variable setting being labor-intensive and creating human uncertainty. These difficulties are amplified by the complexity of the pattern in the printed circuit board (PCB) substrate. This work presents an automated convolutional neural network (CNN)-based algorithm to infer the anisotropic ETCs of package substrates. This algorithm divides a layer-pattern image of a PCB into local unit-cell images and learns the pattern characteristics of each unit cell to identify the local ETC. The algorithm then builds an ETC map by integrating the local ETCs for the entire layer. The entire process is fully automated to reduce human uncertainty and required workforce. The ETC map from the algorithm was then used in finite element (FE) analysis and compared with three other prediction methods. The proposed algorithm can predict the anisotropic ETCs within 2–3 % errors compared to the reference data while other models lead to at least 16 % error. The FE simulation with the ETC map of the algorithm can reflect the effect of the design pattern on the heat flux and temperature distributions on the package layer, leading to the lowest root mean square error in the temperature distribution compared to other models.https://ieeexplore.ieee.org/document/9774413/Artificial intelligenceconvolutional neural networkeffective thermal conductivityfinite element methodsemiconductor package |
| spellingShingle | Tae-Hyun Kim Jeong-Hyeon Park Ki Wook Jung Jaechoon Kim Eun-Ho Lee Application of Convolutional Neural Network to Predict Anisotropic Effective Thermal Conductivity of Semiconductor Package IEEE Access Artificial intelligence convolutional neural network effective thermal conductivity finite element method semiconductor package |
| title | Application of Convolutional Neural Network to Predict Anisotropic Effective Thermal Conductivity of Semiconductor Package |
| title_full | Application of Convolutional Neural Network to Predict Anisotropic Effective Thermal Conductivity of Semiconductor Package |
| title_fullStr | Application of Convolutional Neural Network to Predict Anisotropic Effective Thermal Conductivity of Semiconductor Package |
| title_full_unstemmed | Application of Convolutional Neural Network to Predict Anisotropic Effective Thermal Conductivity of Semiconductor Package |
| title_short | Application of Convolutional Neural Network to Predict Anisotropic Effective Thermal Conductivity of Semiconductor Package |
| title_sort | application of convolutional neural network to predict anisotropic effective thermal conductivity of semiconductor package |
| topic | Artificial intelligence convolutional neural network effective thermal conductivity finite element method semiconductor package |
| url | https://ieeexplore.ieee.org/document/9774413/ |
| work_keys_str_mv | AT taehyunkim applicationofconvolutionalneuralnetworktopredictanisotropiceffectivethermalconductivityofsemiconductorpackage AT jeonghyeonpark applicationofconvolutionalneuralnetworktopredictanisotropiceffectivethermalconductivityofsemiconductorpackage AT kiwookjung applicationofconvolutionalneuralnetworktopredictanisotropiceffectivethermalconductivityofsemiconductorpackage AT jaechoonkim applicationofconvolutionalneuralnetworktopredictanisotropiceffectivethermalconductivityofsemiconductorpackage AT eunholee applicationofconvolutionalneuralnetworktopredictanisotropiceffectivethermalconductivityofsemiconductorpackage |