本論文以P(VDF-TrFE) 粉末調製之壓電溶液為基礎，設計與製作出一種新型態的聚焦式超聲波換能器，稱為解析式超聲波換能器。本研究利用黃光微影、蝕刻、與金屬蒸鍍、…等技術，在石英聚焦鏡頭上製作出多個獨立的壓電感測單元，可以同時讀取不同位置的聲學反射與背向散射信號。與傳統點聚焦式換能器將所有壓電感測區域內的信號進行積分與輸出單一一個電壓信號相比，新型態之換能器能具有更佳的空間與時間解析能力，有助於還原回波信號所挾帶的資訊，提升其聲學檢測能力。
在實驗方面，本論文以上述自製解析式超聲波換能器搭配多通道擷取系統，對標準裂紋試片進行C-Scan掃描，蒐集大量的背向散射波形資訊，應用於卷積人工智慧模型的訓練，充分篩檢與擷取出最有利於建構正確缺陷資訊的關鍵信號，使其具備判斷裂縫深度的能力。
傳統超聲波影像掃描受限於單一的積分輸出信號，且仰賴檢測人員對二維影像的判斷，容易受到外界因素的干擾。本論文提出利用新型態解析式換能器蒐集大量背向散射信號，並使用人工智慧進行檢測，避免人為因素影響的同時，可以使用更多維度的資訊進行缺陷辨識，提高檢測的精度。

This thesis successfully designed and developed a novel type of focused ultrasonic transducer, called analytical ultrasonic transducer, based on a piezoelectric solution formulated with P(VDF-TrFE) powder. By employing techniques such as photolithography, etching, and electron beam evaporation, multiple independent sensing regions are created on a quartz focusing lens, enabling simultaneous readout of acoustic reflection and backscatter signals from different positions. In contrast to traditional point focused transducers that integrate signals from the whole sensing region, the proposed transducer can spatially and independently analyze acoustic signals, and therefore is able to reconstruct the information carried by the signals.
The analytical ultrasonic transducer is utilized in conjunction with a multi-channel acquisition system for performing C-Scan on standard crack specimens. A significant amount of backscatter waveform information is collected for training a convolutional artificial intelligence model. The model effectively screens and extracts key signals that are most advantageous for constructing accurate defect information, empowering it with the ability to determine crack depth.
Traditional ultrasound imaging inspection is limited by the single integrated output signal and relies on human judgment, making them susceptible to external interference. The proposed analytical transducer can collect a substantial amount of backscatter signals and employs artificial intelligence for inspection. By avoiding human factors and utilizing higher-dimensional information for recognition, the accuracy of detection is enhanced.

[1] D. E. Bray and D. McBride, Nondestructive Testing Techniques, Wiley, 1992.
[2] H. Kawai, “The Piezoelectricity of Poly (vinylidene Fluoride),’’ Jpn. J. Appl. Phys., vol. 8, pp. 975-976, 1969.
[3] K. Kimura, and H.Ohigashi, “Generation of very high‐frequency ultrasonic waves using thin films of vinylidene fluoride‐trifluoroethylene copolymer,’’ J. Appl. Phys., vol. 61, no. 10, pp. 4749-4754, 1987.
[4] L. F. Brown, and A. M. Fowler, “High vinylidene-fluoride content P(VDF-TrFE) films for ultrasound transducers,’’ Proc. 1998 IEEE Ultrason., vol. 1, pp. 607-610, 1998.
[5] L. F. Brown, “P (VDF-TrFE) films for high-frequency ultrasonic transducers,’’ Proc. SPIE Medical Imaging 1999, vol. 3664, pp. 31-34, 1999.
[6] E. Fukada, “History and recent progress in piezoelectric polymers,’’ IEEE Trans. UFFC, vol. 47, no. 6, pp. 1277-1290, 2000.
[7] S. Smolorz, and W. Grill, “Focusing PVDF transducers for acoustic microscopy,’’ Rev. Prog. Quant. Nondestr. Eval., vol. 7, no. 4, pp. 195-201, 1996.
[8] M. Robert, G. Molingou, K. Snook, J. Cannata, and K. Shung, “Fabrication of focused poly (vinylidene fluoride-trifluoroethylene) P (VDF-TrFE) copolymer 40–50 MHz ultrasound transducers on curved surfaces,’’ J. Appl. Phys., vol, 96, no. 1, pp. 252-256, 2004.
[9] R. S. Dahiya, M. Valle, G. Metta, L. Lorenzelli, and S. Pedrotti, “Deposition, processing and characterization of P (VDF-TrFE) thin films for sensing applications,’’ SENSORS, 2008 IEEE, pp. 490-493, 2008.
[10] A. Habib, S. Wagle, A. Decharat, and F. Melandsø, “Evaluation of adhesive-free focused high-frequency PVDF copolymer transducers fabricated on spherical cavities,’’ Smart Mater. Struct., vol. 29, no. 4, 045026, 2020.
[11] L. E. Kinsler, A. R. Frey, A. B. Coppens, and J. V. Sanders, Fundamentals of Acoustics, John wiley & sons, 2000.
[12] Y. Gues et al., “A neural network architecture for ultrasonic nondestructive testing,” IEEE 1991 Ultrasonics Symposium, pp. 777-780, 1991.
[13] S. W. Lawson and G. A. Parker, “Automatic detection of defects in industrial ultrasound images using a neural network,” Proc. SPIE, vol. 2786, pp. 37-47, 1996.
[14] S. Sambath, P. Nagaraj, and N. Selvakumar, “Automatic defect classification in ultrasonic NDT using artificial intelligence,’’ J. Nondestruct. Eval., vol. 30, pp. 20-28, 2010.
[15] F. C. Cruz, et al., “Efficient feature selection for neural network based detection of flaws in steel welded joints using ultrasound testing,’’ Ultrasonics, vol. 73, pp. 1-8, 2017.
[16] N. Munir, H. J. Kim, S. J. Song, and S. S. Kang, “Investigation of deep neural network with drop out for ultrasonic flaw classification in weldments,’’ J. Mech. Sci. Technol., vol. 32, pp. 3073-3080, 2018.
[17] K. Manjula, K. Vijayarekha, and B. Venkatraman, “Quality enhancement of ultrasonic TOFD signals from carbon steel weld pad with notches,’’ Ultrasonics, vol. 84, pp. 264-271, 2018.
[18] N. Munir, H. J. Kim, J. Park, S. J. Song, and S. S. Kang, “Convolutional neural network for ultrasonic weldment flaw classification in noisy conditions,’’ Ultrasonics, vol. 94, pp. 74-81, 2019.
[19] I. Virkkunen, T. Koskinen, O. Jessen-Juhler, and J. Rinta-Aho, “Augmented ultrasonic data for machine learning,’’ J. Nondestruct. Eval., vol. 40, pp. 1-11, 2021.
[20] T. Latête, B. Gauthier, and P. Belanger, “Towards using convolutional neural network to locate, identify and size defects in phased array ultrasonic testing’’ Ultrasonics, vol. 115, 106436, 2021.
[21] Y. LeCun, C. Cortes, “The MNIST Database of Handwritten Digits,’’ http://yann. lecun. com/exdb/mnist/, 1998.
[22] P. T. De Boer, D. P. Kroese, S. Mannor, and R. Y. Rubinstein, “A tutorial on the cross-entropy method,’’ Ann. Oper. Res., vol. 134, pp. 19-67, 2005.
[23] D. P. Kingma, J. Ba, “Adam: A method for stochastic optimization,’’ Proceedings of the3^rd International Conference on Learning Representations, San Diego, Califonia, 7-9, 2015.
[24] A. Krizhevsky, I. Sutskever, and G. E. Hinton, “Imagenet classification with deep convolutional neural networks,’’ CACM, vol. 60, no. 6, pp. 84-90, 2017.
[25] C.-T. Kuo, Y.-C. Lee, “An FPGA-Based Ultrasonic Image Scanning System and Non-immersed Ultrasonic Inspection,’’ M.S. Thesis, Department of Mechanical Engineering, National Cheng Kung University, Taiwan, 2014.
[26] Y.-Y. Wei, Y.-C. Lee, “Line-Focused P(VDF-TrFE) Transducer for Measuring Lamb Wave Propagation on Plates with Periodic Structures,’’M.S. Thesis, Department of Mechanical Engineering, National Cheng Kung University, Taiwan, 2022.
[27] C.-S. Lo, Y.-C. Lee, “Fabrication and Application of Analytic Back Scattering Arrayed Transducer,’’ M.S. Thesis, Department of Mechanical Engineering, National Cheng Kung University, Taiwan, 2013.
[28] Y.-C. Chuang, Y.-C. Lee, “Analytic Back Scattering Arrayed Ultrasound Transducer for Non-destructive Evaluation,’’ M.S. Thesis, Department of Mechanical Engineering, National Cheng Kung University, Taiwan, 2016.
[29] C.-H. Chung, Y.-C. Lee, “Focusing P(VDF-TrFE) Transducers for Broad Band Acoustic Wave Measurement and Characterization of Coating Materials,’’ PhD Dissertation, Department of Mechanical Engineering, National Cheng Kung University, Taiwan, 2010.
[30] Y.-T. Chen, Y.-C. Lee, and Y.-C. Wang, “Machine Learning Applied to Ultrasonic Touch,’’ M.S. Thesis, Department of Mechanical Engineering, National Cheng Kung University, Taiwan, 2019.