洋流發電機於實海域測試之訊號擷取與分析暨污染物型態之智慧視覺辨識

國際趨勢使然，且因應國家能源計劃，而發展洋流發電技術，並推動自主的海洋再生能源開發，來促進臺灣海流發電產業發展，因此洋流發電機團隊基於浮游式黑潮發電渦輪機系統(Floating Kuroshio Turbine, FKT)的設計與分析等關鍵技術，依照一定的比例縮小，來建置基於黑潮來進行發電的發電機原型，並建立電力後處理系統及控制系統規劃設計與建置，同時完成實海域拖曳測試與驗證，長期將以商用為目的。 2021年度在海洋委員會國家海洋院的支持下已經完成20kW洋流發電機組之實海拖曳的測試工作，並驗證其發電機與浮力控制模組的工作效能。本次拖曳作業為延續2020年度以10kW機組為基礎來進行的20kW洋流發電機組之設計與改裝，據此來籌建具備完整功能的浮游式洋流發電渦輪機組，並在最後藉由作業母船來藉根據不同的流速與船速協助進行。在本次的實驗中，本文將探討其中兩個部份，其中一個部份是以圖形化程式LabVIEW來進行人機介面的設計，並結合PCI-6143多功能I/O介面卡BNC2110類比訊號採集設備來監控發電機於水下產生之電壓及電流的數值，並加以分析，藉以驗證發電機之發電特性；另一個部分則是使用C#在WPF框架下設計人機介面，搭配Arduino微控制器結合感測器來監控發電機在水下的類比訊號呈現，並加以記錄與儲存，以利後續研究分析，上述的兩個部分都將依據取樣定理來完成。以往政府對於廢水的檢測通常需經過瓶裝將待測水質採樣並以各項儀器來進行，經常消耗大量的人力與時間成本，然而廢水也常造成環境不可逆的損害，也因此廢水的處理與檢測是各國政府重要的議題之一。為了解決檢測廢水耗時以及不便的問題，本文將開發一套系統，針對三種廢水中常見的泡沫種類來進行視覺辨識，包含牛血清蛋白、介面活性劑以及十二烷基苯磺酸鹽，基於物聯網的架構，在架設的實驗場域中藉由相機模組搭配樹梅派並通過WiFi傳輸來達成遠端監控，接著將接收的影像資訊回傳至本機端，再經由LabVIEW、ImageJ、OpenCV 的影像演算法來分別對擷取的影像執行影像處理的操作，同時透過Canny邊緣偵測及輪廓偵測演算法來計算泡沫之面積與周長，最後使用Watershed演算法來依照其泡沫特徵執行圖像分割，藉以區分其當前的泡沫種類並分析其特性以利研究人員進一步的分析與探討。

關鍵字

再生能源；洋流發電機；類比訊號； LabVIEW ；取樣定理；微控制器；泡沫；視覺辨識； ImageJ ； OpenCV ； Canny ；輪廓偵測； Watershed ；圖像分割

並列摘要

Due to the international trend, and in response to the national energy plan, the development of ocean current power generation technology and the promotion of independent marine renewable energy development to promote the development of Taiwan's ocean current power generation industry, so the ocean current generator team is based on the Floating Kuroshio Turbine (FKT) system design and analysis and other key technologies, according to a certain scale, to build a generator prototype based on the Kuroshio for power generation, and to establish the planning, design and construction of the power post-processing system and control system, and complete the actual sea area. Drag testing and verification will be used for commercial purposes in the long run. In 2011, with the support of the National Oceanographic Institute of the Oceanographic Commission, the test work of the 20kW ocean current generator set for real-sea towing has been completed, and the working efficiency of its generator and buoyancy control module has been verified. This towing operation is a continuation of the design and modification of the 20kW ocean current generator set based on the 10kW generator set in 2020. Based on this, a fully functional floating type ocean current generator set will be built, and finally the operation mother ship will be used to borrow it. Different flow rates and boat speeds assist. In this experiment, this article will discuss two of them, one of which is the design of the human-machine interface based on the graphical program LabVIEW, combined with the PCI-6143 multi-function I/O interface card and the BNC2110 analog signal The acquisition equipment is used to monitor the voltage and current generated by the generator underwater, so as to verify the power generation characteristics of the generator; the other part is to use C# to design the human-machine interface under the WPF framework, with the Arduino microcontroller combined with the sensor To monitor the analog signal of the generator underwater, and record and store it for subsequent research and analysis, the above two parts will be completed according to the sampling theorem. In the past, the government usually needed to sample the water quality to be tested through bottled water and use various instruments, which often consumed a lot of manpower and time costs. However, wastewater often caused irreversible damage to the environment. Therefore, the treatment and detection of wastewater are very important. One of the most important issues for governments. In order to solve the problem of time-consuming and inconvenient detection of wastewater, this paper will develop a system to visually identify three types of foams commonly found in wastewater, including bovine serum albumin(BSA), surfactants and sodium dodecyl benzene sulfonate (SDBS). The architecture of the Internet of Things, in the experimental field, the camera module is used with the tree plum pie and transmitted through WiFi to achieve remote monitoring, and then the received image information is sent back to the local end, then through the image algorithm of LabVIEW, ImageJ, OpenCV to perform image processing operations on the captured images respectively. At the same time, the area and perimeter of the foam are calculated through the Canny edge detection and contour detection algorithm. Finally, the Watershed algorithm is used to determine the characteristics of the foam. Perform image segmentation to distinguish current foam types and analyze their characteristics for further analysis and discussion by researchers.

並列關鍵字

Renewable Energy ； Ocean Current Generators ； Analog Signals ； LabVIEW ； Sampling theorem ； Microcontrollers ； Foam ； Visual Recognition ； ImageJ ； OpenCV ； Canny ； Watershed ； Image segmentation

參考文獻

[1] Talley, L. D. (2011). Descriptive physical oceanography: an introduction. Academic press.

Google Scholar

[2] Wikipedia contributors. "台灣再生能源," 18 May 2021 05:57 UTC; https://zh.wikipedia.org/wiki/%E5%8F%B0%E7%81%A3%E5%86%8D%E7%94%9F%E8%83%BD%E6%BA%90

Google Scholar

[3] Wikipedia contributors. "綠島發電廠," 16 September 2020 14:00 UTC; https://zh.wikipedia.org/wiki/%E7%B6%A0%E5%B3%B6%E7%99%BC%E9%9B%BB%E5%BB%A0.

Google Scholar

[4] 環境資訊中心 contributors. "解決綠島供電問題," 11 April 2019 14:00 UTC;

Google Scholar

https://e-info.org.tw/node/217402.

Google Scholar

主題瀏覽