應用自動清淤、水質監測、及殘餌辨識之 低勞力輪蟲智慧生產系統

Applications of Automated Sludge Removing, Waterbody Monitoring, and Residual Feed Recognition to Smart Rotifer Aquaculture System

盧柏任(Po-Jen Lu)

指導教授 ： 廖國基

國立臺灣大學/生物資源暨農學院/生物機電工程學系/碩士(2024年)

https://doi.org/10.6342/NTU202403901

摘要

輪蟲為水產養殖海水魚種幼苗之重要開口餌料生物，其具備增殖速度快、泳速慢、分布廣泛、魚苗適口性佳、高營養價值、及可高密度培養等優點。本研究基於前人研發之連續式輪蟲生產技術，檢視高投餌量對於輪蟲養殖系統之影響，期間亦檢視其殘餘餌料之變化。接續建置一輪蟲智慧養殖系統，導入殘餘餌料面積辨識與自動化監控技術。殘餘餌料面積辨識系統係基於前人研發之輪蟲自動計數，進一步訓練模型使其辨識殘餘餌料，藉以量化水中殘餘餌料。結合包含氨氮自動檢測單元、自動清淤單元、及自動取樣單元。氨氮自動檢測單元可將檢測資料經由微控制器ESP32，以無線傳輸之方式傳送至雲端資料庫內，提供養殖業者於手機應用程式查閱數據；自動清淤單元可大幅減少清淤所需耗費勞力；自動取樣單元則可遠端針對水體進行取樣，監測輪蟲密度及殘餌面積，以達到自動化監控目的。透過量化殘餘餌料及輪蟲智慧養殖系統之開發與驗證，優化餌料應用並大幅節省養殖人力，減少業者親自抵達場域進行輪蟲系統維護之必要性。

關鍵字

智慧養殖 ； 輪蟲 ； 殘餘餌料 ； 自動化監控 ； 深度學習

並列摘要

Rotifers are common live feeds offered to larvae of marine fish in aquaculture. They are characterized by their rapid reproduction and abilities to thrive in high-density conditions. Their small size and slow swimming velocity make them an ideal prey for larvae of marine fish. In the current study, a previously established continuous culture system is utilized to investigate the effect of high feed concentration on the density of rotifers. Automated residual feed area recognition, monitoring, and controlling technology are subsequently adopted to develop an advanced smart aquaculture system. The residual feed area recognition system is modified based on a previously developed automated rotifer counting system. A deep learning algorithm is further trained and applied to identify and quantify the residual feed in waterbody. Automatic ammonia-nitrogen detection, sludge removal, and sampling units are further developed and integrated into the smart aquaculture system. The data of ammonia-nitrogen concentration is wirelessly transmitted to a cloud database via a microcontroller. The automatic sludge removal unit can significantly reduce the requirement for manual labor. The automatic sampling unit enables aquaculturists to remotely sample and count the density of rotifer and residual feed area. This proposed rotifer culture system is verified to be effective in reducing the feed usage as well as the labor requirements.

並列關鍵字

smart aquaculture ； rotifer ； residual feed ； automated monitoring and control system ； deep learning

參考文獻

1.Davis, D. A., T. J. Derbes, and M. E. Head. 2018. Culture of small zooplankton for the feeding of larval fish. SRAC Publication 701: 1-6.

2.Fu, Y., H. Hada, T. Yamashita, Y. Yoshida, and A. Hino. 1997. Development of continuous culture system for stable mass production of the marine rotifer Brachionus. Hydrobiologia 358: 145-151.

3.Fulks, W., and K. L. Main. 1991. Rotifer and microalgae culture systems. In "1991 proceedings of a US-Asia workshop", 61-71.

4.Hino, A., S. Aoki, and M. Ushiro. 1997. Nitrogen-flow in the rotifer Brachionus rotundiformis and its significance in mass cultures. In "1996 Live Food in Aquaculture", 77-82.

5.Hirata, H., and S. Yamasaki. 1987. Effect of feeding on the respiration rate of the rotifer Brachionus plicatilis. Springer eBooks: 283-288.