以CFD模擬分析禽舍環境並建置物聯網環境監控平台

Abstract

根據行政院農業委員會統計，臺灣總畜牧產業產值於2017年時約為1637億元，僅次於農耕產值，而其中家禽產業佔總畜牧產業產值約30 %，然而近年來由於禽流感的爆發使產值受影響，因此飼養過程的生物安全性顯得相對重要，據行政院農業委員會修正「畜牧場主要設施設置標準」，推動將過去開放式養殖方式改變成非開放式，甚至更嚴謹的密閉式養殖，並導入自動化環境控制，並降低因人員進出而帶入病原之風險。因此，本研究於密閉式禽舍建立物聯網環境監控平台，利用計算流體動力學(Computational Fluid Dynamics, CFD)的模擬結果選擇適當的環境感測器安裝位置，以收集環境資訊，並結合環境控制策略，以提高管理之效益。使用CFD模擬之結果，亦有助於禽舍內部流場分佈之改善。本研究以溫溼度感測器及熱線式風速計進行環境資訊收集以及CFD模擬模型驗證。宜蘭分所鴨舍的實驗結果顯示，流場、溫場及濕度場之平均誤差為10.5 %、1.05 %及1.12 %，CFD模擬結果顯示，宜蘭分所鴨舍裝設三片導流板並且開啟內循環扇可改善內部氣流分佈情形。台南總所孵化室的實驗結果顯示，流場、溫場及濕度場平均誤差為10.5 %、4.03 %及4.29 %。結果也皆與文獻相近，顯示兩者模型皆具預測能力。CFD模擬結果顯示，台南總所孵化室於內部加裝循環扇可改善熱累積現象。物聯網環境監控平台利用CFD模擬選擇STM 32無線感測模組位置並收集環境資訊上傳網路儲存裝置(Network Attached Storage, NAS)，由NAS進行環境控制。實驗日之上午晴朗下午陰天，開啟水簾及抽風機降溫，鴨舍最大溫降可達4.3 ℃；實驗日之上午晴朗下午陰天，開啟抽風機降溫，鴨舍內部最大溫降0.6 ℃；實驗日之上午陰天下午陰雨，開啟細霧及抽風機降溫，最大溫降0.8 ℃；實驗日之上午陰雨下午陰雨，開啟水簾、細霧及抽風機降溫，鴨舍最大溫降1.5 ℃，可增加整棟鴨舍溫度與相對濕度均勻度且與CFD模擬結果相符。此四種環境控制參照空氣濕線圖(Psychrometric chart)亦符合，以本研究所建立的環境監控平台進行自動化管理模式可以降低病原進入禽舍內的風險。

According to the statistics of Council of Agriculture, Taiwan’s livestock production value is 163 billion NTD dollars in 2017, which was next to crop production value. Poultry value counts about 30 % of whole livestock production value. The poultry production has been often affected by bird flu invasion in recent years; as a result, bio-safety of breeding process is apparently important. Based on the law amended by Council of Agriculture, poultry houses must be rebuilt as non-open or more rigorous forms and equipped with automated environmental control system to reduce the risk from virus invasion. This study aimed to establish an IoT environmental monitoring platform, including environmental control strategies to provide effectiveness of management. The CFD (Computational Fluid Dynamics) computer simulations were used to know the better locations for the installation of environmental sensors, and to improve the internal flow distribution in the poultry house.The experiments were conducted using temperature sensor, humidity sensor and a hot-wire anemometer to collect environmental information and to verify the CFD simulation. In Yilan duck house, the results show that the averaged error of flow field, temperature field and humidity field compared to the CFD simulated results were 10.5 %, 1.05 % and 1.12 % respectively. While in hatchery room of Tainan, the average error of flow field and temperature were 10.5 % and 4.03 %. All results of simulations were similar to the literature, which showed that the models had good prediction ability. Heat accumulation in hatchery room of Tainan can be improved by installing a circulation fan.IoT environmental monitoring platform was based on the above CFD simulation results, and used STM 32 wireless sensing module to upload the information required by the poultry house to the NAS cloud server in the network layer via Wi-Fi. With wet pads and suction fans, temperature in duck house was decreased 4.3 ℃ when the weather was sunny in the morning and cloudy in the afternoon. With suction fans, temperature in duck house was decreased 0.6 ℃ when the weather was cloudy in the morning and rainy in the afternoon. With mist spray and suction fans, temperature in duck house was decreased 0.8 ℃ when the weather was rainy all day. With wet pads, mist spray and suction fans, temperature in duck house was decreased 1.5 ℃ when the weather was cloudy all day. All the measurements matched the results of CFD simulations. Four strategies also matched the psychrometric chart. The developed automated environmental control platform can be used to reduce the risk of virus invasion.