工程學

生活中網路商店或夜市常看到業者手工用鋁線在折造型，這些造型必須應用手工方式生產，所以生產效率及重現性低，且這類型的塑性加工在課堂上我們只能觀看一些影片來教學，無法實際進行操作。本研究用單折彎頭來進行金屬線之折線加工，以課堂所學知識及加工技術背景進行開發，主要以3D列印機的Arduino+RAMPS1.4控制器為基礎，自行設計及加工折線機構、進線機構、轉軸機構與螺桿機構開發出小型折線機。利用Python程式語言開發出將圖型座標轉換為NC碼，透過ESP-01S物聯網方式，直接控制所需的機械動作，做出所需的作品，讓使用者只需畫圖即可做出成品，不需學習機器控制語法。本研究可自動化加工生產、具有3D折線功能、折不同軟硬度的材料、操作介面簡便、體積小、成本低。

The WHO reports 2.2 billion people internationally have a form of visual impairment, with Perkins School of Blind adding that 4 to 8 percent (8.8 - 17.6 million people) solely rely on a white cane for navigation. In an interview by Stephen Yin for NPR, visually impaired interviewees claimed that a white cane was ineffective as it failed to detect moving obstacles (ex. bikes), aerial obstacles (ex. falling objects), and it became physically demanding after a prolonged period. This problem can be solved with a headset that integrates LiDAR technology and haptic feedback to provide a real-time assessment of their environment. Theoretically, the device will determine how far an object is from the user and place it into one of three conditionals based on distance (0- 290mm, 310-500mm, 510-1200mm). As the user gets closer to the object, the haptic will vibrate more frequently. The device has 11 LIDAR sensors, beetle processors, and ERM motors so that when the LiDAR detects an object, the device will send a haptic signal in that area. It not only identifies the existence of an object but it tells the user its relative position with a latency period of approximately 2 milliseconds. When testing the device, a simulated walking environment was made. Ten obstacles were included: five below the waist (72”, 28”, 35” and 8.5” tall sticks) and five above the waist (paper suspended 6”, 10”, 48” and 28” from the ceiling). The white cane detected 4.1 obstacles, whereas the device detected 7.3 on average. The LiDAR navigation system is 178% more effective at detecting objects comparatively. Visually impaired individuals no longer must rely on the white cane; rather, using this device, they can detect small, moving, and aerial objects at a much faster, and more accurate speed.

我們設計了一款透過壓電片與摩擦片收集資料的智慧鞋，壓電片嵌入在鞋底，摩擦片安裝在前腳掌，兩者並聯。當人行走時，感測器會被擠壓變形，藉由DAQ（數據採集）收集感測器的電壓輸出，可顯示出正常步行、快走、慢跑和漫步等活動的訊息，利用時變電壓形式的輸出數據，與能夠識別時域信號的CNN深度學習（卷積神經網絡）進行不同類型步態辨識。 實驗結果顯示此方法可以辨檢測這四種步態，其辨識率高達95%。訓練好的CNN可同時辨識個人身份與步態。結果顯示，識別快走時辨識率極高，識別正常步行和漫步時辨識率為90%，識別慢跑時辨識率僅達49%。因此，我們未來預計將提高同時辨識不同受試者與不同步態之辨識率，並透過壓電能量擷取器為藍牙模組供電。

本研究藉由孔洞抽氣控制氣流邊界層，影響尾部渦流以達到減阻效果。研究主要探討的變因有:孔洞大小、抽氣速率，進而發想抽吸設計是否可運用在旋轉圓柱上。實驗與之前不同的是為抽氣孔洞加裝導管，以及創新的實驗旋轉裝置。實驗結果顯示，透過延緩邊界層分離可以有效控制阻力，在雷諾數15000時，可減阻。抽氣速率達22m/s，減阻最大值達23%。此實驗想法可有效達到減阻效果，並且可以使旋轉葉片減少旋轉阻力，在電壓6V時，轉速提升11%。未來期望能應用在風力發電機葉片上，減少旋轉風阻，提升發電效率。

網路上已經有很多人在做遠端的魚缸環境監控，但監控好環境條件，魚也不一定過得好，本實驗則是導入魚活動的參數，直接分析魚的活動力，以最直接的方式去觀測小丑魚的舒適程度。 本實驗研究目的為利用Arduino監測系統，監測裝置，並應用Blynk程式繪製環境數值趨勢變化圖。 本實驗的監控變因分為四個：水溫、pH值、水濁度，利用Blynk程式建立起一個能長時間即時監控及紀錄的系統，並建立起警示系統作為提醒裝置。 本實驗更進一步的結合Pixy Cam的監控系統，改變溫度及光照週期，並結合Raspberry pi的數據處理功能，自動處理龐大的數據，探討此兩變因對小丑魚活動力的影響。 此實驗是個發想，自動追蹤紀錄並處理大數據，此方法在未來可用來延伸研究養殖小丑魚的各種行為模式。

In our experiment, we used traditional Chinese medicine to dye on cheap wood, in addition to avoiding the impact of chemical paint on human body; After dyeing, the color and texture quality of the wood are improved, which makes cheap wood have higher price and improves the value of wood; At the same time, it can reduce the felling of slow growing precious wood, which has the functions of environmental protection, earth love and carbon saving. The test material was pretreated with hydrogen peroxide and surfactant, and the bleaching effect was obvious. After dyed with different Chinese medicinal, soak in strong acid and alkali solution for 15 minutes, which shows that strong acid and acid treatment is not allowed. On the other hand, after 15 minutes of immersion in detergent, the color difference value is less than 2, and the rubbing fastness is above grade 4. In the bacteriostasis experiment, no fungus grew in the first 3 days, and it did not grow in the 12th day. In the anti-termite experiment, the mortality rate on the fifth day was 65% for Lithospermum and 83.8% for Wolfberry, and the other groups had a good effect of total elimination. While plastic products have a great impact on the environment, wood that is dyed or modified with natural colored dye, its environmental value far exceeds the human visual perception.

此研究選擇以袋鼠為仿生對象，希望應用袋鼠高速移動的特點，製作仿生動物，並觀測其跳耀動作，提升動作的流暢度。 首先對澳洲袋鼠在跳躍時的動作，進行動態分析，取得跳躍時其最佳腿部彎曲動作。經歷二次的外觀與整體結構更改，以及數十次的細微尺寸與外觀修飾的調整，完成了此次使用的仿生袋鼠。 此仿生袋鼠使用18公分長的小腿為基準，設計跳躍動作影響便因。首先更改個部位的馬達扭力進行跳躍距離和高度的紀錄，測驗出距離最遠，高度最高的數據，並依同樣的變因條件，進行腿部長度的變更，同樣求得最佳數據再進行下一實驗，以此求得在各變因下最佳的條件。 本研究包含生物觀察、機械繪圖、動作分析，而希望此實驗數據在未來能應用於跳躍型的外骨骼。

Low-field (LF) MRI scanners have the power to revolutionize medical imaging by provid- 27 ing a portable and cheaper alternative to high-field MRI scanners. However, such scanners are usu- 28 ally significantly noisier and lower quality than their high-field counterparts. This prevents them 29 from appealing to global markets. The aim of this paper is to improve the SNR and overall image quality of low-field MRI scans (called super-resolution) to improve diagnostic capability and, as a result, make it more accessible. To address this issue, we propose a Nested U-Net neural network architecture super-resolution algorithm that outperforms previously suggested super-resolution deep learning methods with an average PSNR of 78.83 ± 0.01 and SSIM of 0.9551 ± 0.01. Our ANOVA paired t-test and Post-Hoc Tukey test demonstrate significance with a p-value < 0.0001 and no other network demonstrating significance higher than 0.1. We tested our network on artificial noisy downsampled synthetic data from 1500 T1 weighted MRI images through the dataset called the T1- mix. Four board-certified radiologists scored 25 images (100 image ratings total) on the Likert scale (1-5) assessing overall image quality, anatomical structure, and diagnostic confidence across our architecture and other published works (SR DenseNet, Generator Block, SRCNN, etc.). Our algo- rithm outperformed all other works with the highest MOS, 4.4 ± 0.3. We also introduce a new type of loss function called natural log mean squared error (NLMSE), outperforming MSE, MAE, and MSLE on this specific SR task. Additionally, we ran inference on actual Hyperfine scan images with successful qualitative results using a Generator RRDB block. In conclusion, we present a more ac- curate deep learning method for single image super-resolution applied to low-field MRI via a 45 Nested U-Net architecture.

Lower limb amputations affect about 28.9 million people worldwide, influencing normal human functions, we are developing a conscious brain mind-controlled Cybonthitic cyborg bionic-leg to provide a professional solution for this problem, which is classified as restricted knee movement, short-term solution, limited pressure bearing, unspecific analog reading of EMG; Because the output voltage measured in nano-volts, resulting in unspecific knee movement. The functionality of these modern gadgets is still limited due to a lack of neuromuscular control (i.e. For movement creation, control relies on human efferent neural signals to peripheral muscles). Electromyographic (EMG) or myoelectric signals are neuromuscular control signals that can be recorded from muscles for our engineering goals. We worked on a sophisticated prosthetic knee design with a 100-degree angle of motion. We also used a specific type of coiled spring to absorb abrupt or unexpected motion force. In addition, we amplified the EMG output from (Nano-Voltage) to (Milli-Voltage) using customized instrumentation amplifiers (operational amplifiers). We used a full-wave rectifier to convert AC to DC, as a consequence of these procedures, sine-wave output voltage measures in millivolts, and the spring constant indicates the most force for every 1cm. Von mises Stress analysis shows bearing as 3000N is the maximum load for the design. Detecting the edge of a stairwell using the first derivative. The benefit of a system that controls the prosthetic limb is activated by the patient’s own EMG impulses, rather than sensors linked to the body.

The surgical mask has been our daily companion since the outbreak of the Corona pandemic. The nonwovens (outer layers, not the filter membrane) from which the surgical mask is constructed consist of very long and thin polypropylene fibers. This leads to the question of whether microplastics are released during breathing through the surgical mask, which could enter the respiratory tract or the lungs. This would have a negative impact on our health, depending on the size of the detached fiber fragments - the smaller the worse because they can enter much deeper in our respiratory tract. In order to investigate the question of whether fiber fragments are released during breathing through a surgical mask, a filtration device was built. The filters were examined under an optical microscope after filtration. If fiber fragments would detach from the surgical mask, they would be found on the filter. Different surgical masks were tested, those that were not worn at all to surgical masks that were worn all day. It was found that fiber fragments were coming off the surgical masks. There were different fiber fragment types. Some fiber fragments were still undamaged (exhibited nice fractures), while others were frayed. Clump-like fragments occurred, but also smaller fine fiber fragments. All these different fiber fragments had a certain size, so that they could be called microplastics. The remarkable result of the whole study is that there is a direct correlation between the wearing time of the surgical mask and the number of detaching fiber fragments. In the case of the unworn surgical masks, 10 times fewer fiber fragments occurred during filtration than in the case of the surgical masks that were worn all day.