為了計算長纖維添加塑膠的複合材料內部的纖維排向，需先建立其內部的3D纖維模型。建立3D纖維模型有很多種方式，本文採用研磨及取像的方式，先將工件表面進行研磨，研磨完畢再將其表面放置於纖維掃描機內進行大面積微觀影像的掃描擷取完成一層影像，接著重複上述動作完成每層影像的擷取，最後再經由AVIZO將每層影像疊合成3D纖維影像。而傳統手動研磨的方式因為無法確保研磨時施加於工件各處的研磨力量均一致，使得研磨完畢的工件表面平面度不佳，造成其後進行取像步驟在一次對焦的情況下取像品質不佳，並且以手動研磨方式無法精確控制材料的移除厚度。基於以上原因，作者開發一部自動偵測研磨厚度及力量控制研磨機試圖改善上述情況，經由機構設計、電控設計、程式設計實際實現出一部以PID的控制策略控制研磨力量及附帶以磁性尺感測器自動偵測研磨厚度的研磨機。實際實現機台後，進行手動研磨及機台自動研磨其研磨表面平面度的比較，並實際放入顯微掃描機內進行掃描取像比較兩者取像的清晰度。實驗結果機台自動研磨的研磨表面平面度較好，但兩者研磨表面均有中間高、周圍低的現象，造成放入取像設備中以僅在掃描起始點位置處對焦的掃瞄方式掃描無法使得工件在整個表面的各位置處的影像皆清晰，文後並針對此現象發生的原因加以討論。

Grinding by hand will cause non-uniform force on a sample, resulting in a grinding surface that is not flat. This will cause some of the scanned images of long-fiber-composite samples scanned with microscope scanner to be unclear with one focus. In addition, grinding thickness can’t be exactly predicted when grinding is done by hand. The other reason is that typically, the diameter of a polish plate is 8 inches, but a bigger one is needed to grind a big sample. The author developed a polisher with automatic thickness detection and force control through a mechanical design, an electronic design, programming, noise processing, and PID control of the grinding force. The movements of the automatic polisher occur in four stages: First, the sample is fixed in the fixture, and then the fixture is rotated. Second, the velocity is set to move the fixture down near the grinding plate. Third, the grinding thickness, grinding force, and grinding plate rotation speeds are set, and then grinding is started. Fourth, grinding is automatically stopped when the grinding thickness equals the setting thickness, and the fixture then moves up at a constant velocity to the point of origin. Finally, grinding experiments are conducted for large, long-fiber-composite samples ground by hand and using the automatic polisher, respectively, to compare the flatness of each grinding surface with a dial indicator measurement and one focus image only focusing on the initial point using a micro scanner. The results show that the grinding surface flatness obtained with the automatic polisher is better than that obtained by hand grinding. On the other hand, the results of the grinding force control in steady state indicated significant oscillation through the setting target force, which is discussed in the following article.

[1] Suresh G. Advani, and Charles L. Tucker III, "The Use of Tensors to Describe and Predict Fiber Orientation in Short Fiber Composites," Journal of Rheology, Vol 31, Issue 8, pp 751-784, 1987.
[2] J. C. Halpin, and J. L. Kardos, "The Halpin-Tsai Equations: A Review," Polymer. Engineering and Science, Vol 16, No.5, pp. 344-352, 1976.
[3] Randy S. Bay, Charles L. Tucker III, "Stereological Measurement and Error Estimates for Three Dimensional Fiber Orientation," Polymer Engineering & Science, Vol 32, Issue 4, pp 240-253, 1992.
[4] Yong Lee, Seok Lee, Jae Youn, Kwansoo Chung, Tae Kang, "Characterization of Fiber Orientation in Short Fiber Reinforced Composites with an Image Processing Technique," Materials Research Innovations, Vol. 6, Issue 2, pp 65-72, 2002.
[5] 匿名。FA工廠自動化用機械標準零件。台灣三柱股份有限公司(2014)。
[6] 洪維恩。C語言教學手冊。旗標科技股份有限公司(2010)。.
[7] 洪維恩。C++教學手冊。旗標科技股份有限公司(2010)。
[8] Stephen Prata. C++Primer Plus中文版(第六版)。(蔡明志譯)。碁峰出版社(2012)。
[9] 匿名。台達交流伺服驅動器ASDA-A2系列型錄。台達電子工業股份有限公司。
[10] 匿名。台達交流伺服系統ASDA-B2系列型錄。台達電子工業股份有限公司。
[11] 匿名。變頻馬達型錄。富田電機股份有限公司(2016)。
[12] 匿名。台達精巧標準型向量控制變頻器MS300系列使用手冊。台達電子工業股份有限公司。
[13] 匿名。台達高機能通訊型伺服驅動器ASDA-A2系列應用技術手冊。台達電子工業股份有限公司。
[14] 匿名。台達高機能通訊型伺服驅動器ASDA-B2系列應用技術手冊。台達電子工業股份有限公司。
[15] 匿名。ICP LinPAC-8x81-Series SDK。泓格科技股份有限公司。
[16] 匿名。PCI-8154 User’s Manual。惠達科技股份有限公司。
[17] Wikipedia, PID Controller, https://en.wikipedia.org/wiki/PID_controller.
[18] P.M. Meshram, and Rohit G. Kanojiya, "Tuning of PID Controller Using Ziegler-Nichols Method for Speed Control of DC Motor," Advance in Engineering, Science and Management, Tamil Nadu, India, 30 Mar.~31 Mar., 2012, pp 117-122.
[19] Alan V. Oppenheim and Ronald W. Schafer, Discrete-Time Signal Processing, 2nd ed. Prentice-Hall Inc, USA, 1992.