「熱遮罩」是一種熱隱形技術，旨在操縱熱流或隱藏物體之傳遞熱。迄今為止，熱遮罩技術多年的研究皆取得優異的成果，然而為使其應用得以廣泛，仍有許多困境需要突破。首先，實際工程應用上，內部熱源的散熱是必要的，但過往熱遮罩其封閉型設計，導致內部存在熱源時，內部溫度無法達到穩態並且持續升高進而引發元件功能失效；其次，可重構式的結構也是傳統熱遮罩較少考量的地方，過往關於三層熱遮罩的研究大多都是針對在雙層熱遮罩內部添加一層過渡層，以達成隱形之效果，此方法亦需要重新製作熱遮罩。而藉此希望「熱圓頂」的創新概念能夠在三維空間下，利用其半球型開放式結構能夠針對內部熱源穩定及可重構式結構，將能為熱遮罩技術延伸至更加廣泛的實際應用。因此本文提出了以均勻且各向同性之自然材料建構熱圓頂進而達成隱形的效果，並針對內部存在熱源的情況進行模擬，也針對背景材料變化時其可重構式結構之探討，並嘗試了不規則幾何形狀之可行性，結果顯示，內層即使採用低熱導率之非完美絕熱自然材料，也能找到最佳化幾何尺寸。而針對內部熱源的部分，也確保其所散發之熱量在熱圓頂的保護下不會擴散至背景區域，不僅突破過往熱隱形裝置於材料選擇上的困境，也成功提高其於工程應用上之實用性。此外，本文基於深度學習方法設計熱圓頂，達到降低計算時間成本並能夠針對不規則幾何形狀進行最佳化設計，未來於工程應用上能夠藉由此深度學習方法進行逆向工程求解所需之參數，以達成所需熱遮罩的最佳化設計。

The concept of a "thermal cloak" involves thermal invisibility technology that manipulates heat flow to conceal the heat transmission from objects. Despite significant research advancements, several challenges remain for its broader application. One major issue is that traditional thermal cloaks, designed as closed systems, prevent efficient heat dissipation from internal heat sources. This results in an inability to achieve steady-state temperatures and can lead to overheating and potential component failure. Another challenge lies in the lack of focus on reconfigurable structures in traditional thermal cloaks. Previous research on trilayer thermal cloaks typically added a transition layer within bilayer designs to achieve invisibility, which required remanufacturing.
To address these challenges, this study introduces the innovative concept of a "Thermal Dome." With its hemispherical, open structure, the thermal dome allows for stable internal heat management and reconfigurable configurations, expanding the practical applications of thermal cloaking technology. The dome is constructed using uniform and isotropic natural materials to achieve invisibility, and simulations were conducted with internal heat sources to explore reconfigurable structures using different background materials. The study also tested the feasibility of irregular geometric shapes. Results show that, even when using non-ideal insulating natural materials with low thermal conductivity for the inner layer, optimal geometric designs can be achieved. The thermal dome effectively contains heat from internal sources, preventing diffusion into the surrounding environment. This approach overcomes material limitations previously associated with thermal cloaking devices and significantly enhances their practicality in engineering applications.
Additionally, this study employs deep learning method to design the thermal dome, reducing computational time and optimizing designs for irregular geometries. Looking forward, deep learning could be applied to reverse-engineer parameters for optimal thermal cloaking designs in practical engineering scenarios.

參考文獻

[1] Leonhardt, U., Optical conformal mapping. Science, 2006, 312(5781), 1777-1780.
[2] Pendry, J.B., Schurig, D., and Smith, D.R., Controlling electromagnetic fields. Science, 2006, 312(5781), 1780-1782.
[3] Fan, C., Gao, Y., and Huang, J., Shaped graded materials with an apparent negative thermal conductivity. Applied Physics Letters, 2008, 92(25),
[4] Guenneau, S., Amra, C., and Veynante, D., Transformation thermodynamics: cloaking and concentrating heat flux. Optics Express, 2012, 20(7), 8207-8218.
[5] Han, T., Bai, X., Gao, D., Thong, J.T., Li, B., and Qiu, C.-W., Experimental demonstration of a bilayer thermal cloak. Physical Review Letters, 2014, 112(5), 054302.
[6] Wang, J., Dai, G., and Huang, J., Thermal metamaterial: fundamental, application, and outlook. Iscience, 2020, 23(10),
[7] Alekseev, G. and Levin, V. An optimization method for the problems of thermal cloaking of material bodies. in Doklady Physics. 2016. Springer.
[8] Zhou, Y., Yang, F., Xu, L., Zhuang, P., Wang, D., Ouyang, X., Li, Y., and Huang, J., Reconfigurable three-dimensional thermal dome. arXiv preprint arXiv:2309.04745, 2023,
[9] Leo, M., Sharma, S., and Maddulety, K., Machine learning in banking risk management: a literature review. Risks, 2019, 7(1), 29.
[10] Gregor, K., Danihelka, I., Graves, A., Rezende, D., and Wierstra, D. Draw: a recurrent neural network for image generation. in International Conference on Machine Learning. 2015. PMLR.
[11] Liu, Z., Xia, X., Hassan, A.E., Lo, D., Xing, Z., and Wang, X. Neural-machine-translation-based commit message generation: how far are we? in Proceedings of the 33rd ACM/IEEE International Conference on Automated Software Engineering. 2018.
[12] Pak, M. and Kim, S. A review of deep learning in image recognition. in 2017 4th international conference on computer applications and information processing technology (CAIPT). 2017. IEEE.
[13] Das, P., Ivkin, N., Bansal, T., Rouesnel, L., Gautier, P., Karnin, Z., Dirac, L., Ramakrishnan, L., Perunicic, A., and Shcherbatyi, I. Amazon sagemaker autopilot: a white box automl solution at scale. in Proceedings of the fourth international workshop on data management for end-to-end machine learning. 2020.
[14] Nashif, S., Raihan, M.R., Islam, M.R., and Imam, M.H., Heart disease detection by using machine learning algorithms and a real-time cardiovascular health monitoring system. World Journal of Engineering and Technology, 2018, 6(4), 854-873.
[15] LeCun, Y., Bengio, Y., and Hinton, G., Deep learning. Nature, 2015, 521(7553), 436-444.
[16] Cummer, S.A., Popa, B.-I., Schurig, D., Smith, D.R., and Pendry, J., Full-wave simulations of electromagnetic cloaking structures. Physical Review E, 2006, 74(3), 036621.
[17] Huang, Y., Feng, Y., and Jiang, T., Electromagnetic cloaking by layered structure of homogeneous isotropic materials. Optics Express, 2007, 15(18), 11133-11141.
[18] Zhang, B., Chen, H., and Wu, B.-I., Limitations of high-order transformation and incident angle on simplified invisibility cloaks. Optics Express, 2008, 16(19), 14655-14660.
[19] Chen, H. and Chan, C.T., Acoustic cloaking in three dimensions using acoustic metamaterials. Applied Physics Letters, 2007, 91(18),
[20] Sanchis, L., García-Chocano, V.M., Llopis-Pontiveros, R., Climente, A., Martínez-Pastor, J., Cervera, F., and Sánchez-Dehesa, J., Three-dimensional axisymmetric cloak based on the cancellation of acoustic scattering from a sphere. Physical Review Letters, 2013, 110(12), 124301.
[21] Chen, T., Weng, C.-N., and Chen, J.-S., Cloak for curvilinearly anisotropic media in conduction. Applied Physics Letters, 2008, 93(11),
[22] Narayana, S. and Sato, Y., Heat flux manipulation with engineered thermal materials. Physical Review Letters, 2012, 108(21), 214303.
[23] Han, T., Yang, P., Li, Y., Lei, D., Li, B., Hippalgaonkar, K., and Qiu, C.W., Full‐parameter omnidirectional thermal metadevices of anisotropic geometry. Advanced Materials, 2018, 30(49), 1804019.
[24] Yang, F.-Y., Hung, F.-S., Yeung, W.-S., and Yang, R.-J., Optimization method for practical design of planar arbitrary-geometry thermal cloaks using natural materials. Physical Review Applied, 2021, 15(2), 024010.
[25] Yeung, W.-S., Yang, F.-Y., and Yang, R.-J., Analysis of a trilayer thermal cloak. AIP Advances, 2020, 10(5),
[26] Shan, Q., Shao, C., Wang, J., and Xia, G., Enhanced thermal invisibility effect in an isotropic thermal cloak with bulk materials. Chinese Physics Letters, 2023, 40(10), 104401.
[27] Jordan, M.I. and Mitchell, T.M., Machine learning: trends, perspectives, and prospects. Science, 2015, 349(6245), 255-260.
[28] Carleo, G., Cirac, I., Cranmer, K., Daudet, L., Schuld, M., Tishby, N., Vogt-Maranto, L., and Zdeborová, L., Machine learning and the physical sciences. Reviews of Modern Physics, 2019, 91(4), 045002.
[29] Mahesh, B., Machine learning algorithms-a review. International Journal of Science and Research (IJSR).[Internet], 2020, 9(1), 381-386.
[30] Ma, W., Liu, Z., Kudyshev, Z.A., Boltasseva, A., Cai, W., and Liu, Y., Deep learning for the design of photonic structures. Nature Photonics, 2021, 15(2), 77-90.
[31] Zhu, C., Bamidele, E.A., Shen, X., Zhu, G., and Li, B., Machine learning aided design and optimization of thermal metamaterials. Chemical Reviews, 2024, 124(7), 4258-4331.
[32] Liu, B., Xu, L., and Huang, J., Thermal transparency with periodic particle distribution: A machine learning approach. Journal of Applied Physics, 2021, 129(6),
[33] Ji, Q., Qi, Y., Liu, C., Meng, S., Liang, J., Kadic, M., and Fang, G., Design of thermal cloaks with isotropic materials based on machine learning. International Journal of Heat and Mass Transfer, 2022, 189, 122716.
[34] Dede, E.M., Schmalenberg, P., Nomura, T., and Ishigaki, M., Design of anisotropic thermal conductivity in multilayer printed circuit boards. IEEE Transactions on Components, Packaging and Manufacturing Technology, 2015, 5(12), 1763-1774.
[35] Wang, Y., Wang, H., Liu, F., Wu, X., Xu, J., Cui, H., Wu, Y., Xue, R., Tian, C., and Zheng, B., Flexible printed circuit board based on graphene/polyimide composites with excellent thermal conductivity and sandwich structure. Composites Part A: Applied Science and Manufacturing, 2020, 138, 106075.
[36] Krizhevsky, A., Sutskever, I., and Hinton, G.E., Imagenet classification with deep convolutional neural networks. Advances in Neural Information Processing Systems, 2012, 25,
[37] Hopfield, J.J. and Tank, D.W., Computing with neural circuits: A model. Science, 1986, 233(4764), 625-633.
[38] Goodfellow, I., Pouget-Abadie, J., Mirza, M., Xu, B., Warde-Farley, D., Ozair, S., Courville, A., and Bengio, Y., Generative adversarial nets. Advances in Neural Information Processing Systems, 2014, 27,
[39] McCulloch, W.S. and Pitts, W., A logical calculus of the ideas immanent in nervous activity. The Bulletin of Mathematical Biophysics, 1943, 5, 115-133.
[40] Werbos, P., Beyond regression: new tools for prediction and analysis in the behavioral sciences. PhD Thesis, Committee on Applied Mathematics, Harvard University, Cambridge, MA, 1974,
[41] Rumelhart, D.E., Hinton, G.E., and Williams, R.J., Learning representations by back-propagating errors. Nature, 1986, 323(6088), 533-536.
[42] Qian, N., On the momentum term in gradient descent learning algorithms. Neural Networks, 1999, 12(1), 145-151.
[43] Dozat, T., Incorporating nesterov momentum into adam. 2016,
[44] Duchi, J., Hazan, E., and Singer, Y., Adaptive subgradient methods for online learning and stochastic optimization. Journal of Machine Learning Research, 2011, 12(7),
[45] Hinton, G., Srivastava, N., and Swersky, K., Neural networks for machine learning lecture 6a overview of mini-batch gradient descent. Cited on, 2012, 14(8), 2.
[46] Kingma, D.P. and Ba, J., Adam: a method for stochastic optimization. arXiv preprint arXiv:1412.6980, 2014,
[47] Multiphysics, C., Introduction to comsol multiphysics®. COMSOL Multiphysics, Burlington, MA, accessed Feb, 1998, 9(2018), 32.
[48] Pryor, R.W., Multiphysics modeling using COMSOL®: a first principles approach. 2009: Jones & Bartlett Publishers.
[49] van Rossum, G., Guido van Rossum. Python (programming language) 1 CPython 13 Python Software Foundation 15, 16.