畜禽舍與孵化機無線環境監控預警系統之開發

Abstract

台灣位居亞熱帶地區，具得天獨厚的地理環境，常年溫度適中多潮濕，但是地狹人稠，禽畜養殖多以小規模飼養為主，如此之飼養方式較難保有完整生產紀錄、疾病控制也不易，現更面臨缺工、從業人員老齡化、成本增加等問題，無論是疾病防治抑或是產能、良率的提升，實務上皆難以落實。且現行之禽畜舍，環境感測資訊的傳遞仍以人工謄寫或有線傳輸設備為主，佈線更改不易、維修所需要之專業需求也較高。 目前針對禽畜環境所做的研究，多是使用市售之感測套組針對其環境進行環境感測資訊蒐集，其價格偏高且耐用性也不佳。而家禽飼養之源頭－孵化業，許多飼養業者仍是依循經驗法則操作設備，缺乏較精確的飼養管理方法。因此，本研究嘗試開發一禽畜環境適用之無線感測模組，其可依業者使用需求搭配感測器方案，以方便業者能夠進行多感測節點佈建之預算管理，另以鵝禽孵化機為設計對象，建立一套無線環控預警系統，其中以無線傳輸之方式降低場內之佈線需求，增加了相對濕度控制功能，以利提升孵化環境之穩定，並於場內蒐集環境與生長資料，提供業者監控之功能，以提升管理之效益。 本研究之模組與孵化機系統分別於台灣大學人工氣候室及合作鵝場之孵化室進行實驗，以本研究自行開發之模組及系統進行資料蒐集。模組可進行防水溫度、一般溫度、相對濕度、照度、風速、氨氣、二氧化碳，共7因子量測，其性能測試之結果顯示，在高、中、低溫搭配高相對濕度情況下，每日平均資料回傳比率除去最後一日之網路不穩情形，均達到0.9左右，蒐集資料之能力平均，表示模組量測能力之穩定度在可接受之範圍。孵化機監控系統測試結果顯示，26日累積之資料總筆數為35499筆，缺漏筆數為4筆，系統之穩定度為0.99。CFD模擬結果指出孵化機後側靠近加熱器之位置易造成溫度之累積，與環控感測點之溫度存在差異，模擬結果與驗證實驗趨勢一致，建議未來可加入如種蛋蛋溫等條件以強健模型。 最後，本研究因應模組與系統之管理需求，建置資訊檢視平台以便資訊之整合，使用者透過網路便能使用此系統並取得分析後之環境資訊，以便更有系統的地進行生產管理。

Taiwan is located in the subtropical region and has a unique geographical environment. The temperature is moderate and humid throughout the year. The livestock farming is mostly based on small-scale breeding due to the high population. It is difficult to maintain a complete production record and disease control. Furthermore, it faces more problems such as lack of labor, aging of employees, increased costs, disease prevention, and improvement in productivity and yield, all of these are difficult to implement in practice. In the current poultry house, the transmission of environmental sensing information is still mainly based on manual transcription or wired transmission equipment. The wiring changes are not easy, and the professional requirements for maintenance are also high. At present, most of the research on the livestock environment used the commercially available sensing kits to collect environmental information. The price is high and the durability is not good enough; even for the poultry breeding, hatching industry. Many producers still follow the rules of thumb to operate the equipment without precise feeding management methods. Therefore, this study attempts to develop a wireless sensing module suitable for a livestock environment, which can provide different combination of sensors to match the demands of the operator to facilitate the operator to carry out budget management for multi-node deployment. In addition, a wireless environment control early warning system was established for the goose poultry incubator, which reduced the wiring requirements in the field by means of wireless transmission, and increased the relative humidity control to improve the stability of the incubation environment and collected environmental data and growth data. This monitoring system enhanced the management effectiveness. The modules and incubator system of this study were tested respectively in the Phytotron of National Taiwan University and the cooperative incubation room, and the data were collected by the modules and systems developed by the research. The module can be used for 7 factors in total: waterproof temperature, general temperature, relative humidity, illuminance, wind speed, ammonia gas, carbon dioxide. The performance testing results show that in the environment with high, medium or low temperature combined with high relative humidity, the average of daily data return ratio is around 0.9 except for the network instability on the last day. The ability to collect data indicates that the stability of the module measurement capability is within an acceptable range. The test results of the incubator monitoring system showed that the total number of data accumulated on the 26th was 35,499, the number of missing data was 4, and the stability of the system was 0.99. The results of CFD simulation indicate that the position of the rear side of the incubator near the heater was likely to cause temperature accumulation, which was different from the temperature of the sensing position. The trend of the simulation results were consistent with the verification experiment. It is recommended that the conditions such as egg temperature should be added to build a robust model in the future. Finally, in response to the management needs of modules and systems, this study has built an information review platform for information integration. Users can operate the system and obtain analytical environmental information through the network for more systematic production management.