2010-11-012024-05-13https://scholars.lib.ntu.edu.tw/handle/123456789/654486摘要：近幾十年以來，臺灣高科技產業快速的發展，使得科技產品應用已深入各行各業。為了提升農業競爭力，臺灣農業已經開始利用農業自動化技術發進行轉型。目前，高度自動化的小農經營模式已成為主要趨勢。然而，在各樣作物栽培管理中，蟲害對作物的影響最為快速且直接。1960年代統計資料顯示，全球作物因田間有害生物所帶來的平均損失為35%，其中蟲害便佔了14%，可見針對害蟲之監測、掌控及預測其時空分布為蟲害管制工作中十分重要的一環。目前農作物蟲害防治大多採用定時噴灑化學農藥的方式，然而大量的農藥使用不僅加速害蟲出現抗藥性，也容易造成農藥殘留的問題。此外，害蟲的數量極易受到溫度、溼度、雨量、風速、照度等環境參數變異的影響。單純的讀取蟲害數無法有效的判定蟲害的分布增生情況與危害程度。其次，人工量測的成本相當高，獲得的環境因素資料也不易同步，往往無法有效率的在同一個時間點上大量擷取多種環境因素資料，造成時間位移效應的記錄誤差。相較之下，自動化管理所含的監測技術是獲得農業資訊較佳的方法。為了取代過去藉由人力操控而在農作地進行單點量測與記錄的方式，自動化的監測系統可以增加量測參數項目，解決耗時、耗力、耗工的問題，並增進田間資訊量測的精準性。因此，開發一套可整合多項田間環境參數量測工作的自動化監測系統是目前相當重要且急迫的工作。 本計畫的主旨為研製一套結合智慧型自動化技術、地理圖資資訊、無線感測器網路技術及無線傳輸技術之植物疫情動態監測網。此系統主要由田間監測網路與後端監控中心兩大部分所組成。田間監測網路負責收集田間感測器節點所感測之環境參數與誘捕之害蟲數量。這些資料由閘道器彙整後，透過無線傳輸技術轉送至後端監控中心。由於田野間的環境複雜，有線網路架設不易，本計畫擬採用GSM行動通訊技術，將資訊回傳至後端監控中心。此外，為避免網路佈署地點、地勢及地形的不同所造成之GSM通訊障礙，本計畫擬建構各監測網間的骨幹網路，透過閘道器間彼此通訊，將各監測網串連，再由具GSM通訊能力之閘道器回傳監測參數，以克服GSM通訊死角之問題，並大量節省通訊成本。本監測網的遠端監控中心具智慧型控管功能，可全面監控田間監測平台的運行狀況，並可採取適當的管理措施，能即時針對所回傳的感測資料進行分析比對，建立一完整自動化監控流程。本計畫將利用ASP.NET技術設計Web控制平台，建構一套植物疫情動態監測網的網站服務系統，使系統監控與管理可透過網路進行。此外，監控中心還包含資料展示與分析平台，可提供感測資料的統計圖表查詢功能，並整合地理圖資呈現蟲害密度狀況，提供使用者一整合式的展現平台，對於疫情的掌控和分析有極大的助益。此動態監測網同時具備植物疫情預警功能，利用統計模型建立蟲害數量增生與環境因子間之關係，達警戒狀態時可即時發佈預警訊息，有效降低損害。本計畫之執行預計以三年分三階段進行，茲分述如下： 第一年： 本年度計畫擬設計一套植物疫情動態監測網的雛型系統與現場佈建測試，主要著重在監測網系統實際佈建與雛型系統可行性研究。主要項目包括(A) 感測器節點改良與製作。(B) 多功能田間閘道器改良與製作。(C) LabVIEW主控端之研製。(D) Web控制平台之研製並結合地理圖資建構資料庫。(E) 監測網現場佈建測試。其中(A) 包含四個子項目，分別為東方果實蠅偵測裝置、斜紋夜盜蛾偵測裝置、感測通訊模組及自給式電源供應模組。(B)以不同需求導向分別開發多功能型田間閘道器、低耗能型田間閘道器及綜合型田間閘道器。(D) 著重於以ASP.NET技術建構Web控制平台與結合地理圖資建構資料庫。 第二年： 本年度將進行監控中心之資料展示與分析平台研製、通信協定設計、整合與實測。本平台將架構於資料庫上，並以人性化網頁介面呈現。資料庫將結合地理圖資資訊顯示蟲害分佈狀況，透過時間、空間、感測資料的結合展示，使得農業相關從業人員能透過網頁等介面，更清楚了解蟲害時空之資訊，有效掌握防疫之關鍵期。此外，由於系統硬體可能會隨著佈建地點週遭地理環境、季節氣候的不同而導致測量的偏誤，因此本平台將提供田間監測平台的感測數據迴歸分析功能，建立數據修正模式。其次，本年度將發展適用於田間的高效能通信協定，主要目的為使田間監測平台穩定運行、降低延遲以及節能。本年度也將同步發展骨幹網路的通信協定，建立閘道器間的最佳通訊模式，同時建構與遠端監控中心的連接橋梁。最後，將系統整合並且進行實際田野測試。 第三年： 在進行植物疫情管理時，若充分了解蟲害發生的情況，就能針對不同的情形採取適當的管理措施。利用本計畫所研發的疫情偵測系統即可有效了解田間蟲害發生的情形。此外，由無線通訊模組所回傳的田間資訊，經過長期累積，可作為預測蟲害發生的依據。本計畫將利用系統鑑別的方法，將所收集到的資訊利用演算法處理後，根據蟲數增生的趨勢，預測蟲害可能發生的時間，達到預警的效果，並進而減少農產品的損失。因此，本年度擬針對害蟲族群進行動態分析，以建立蟲害鑑別模型並研發疫情預警功能。 <br> Abstract: In Taiwan, higher technologies have been applied to many industries for decades. To raise the competitiveness of Taiwan’s agriculture, automation has become a new trend and been widely used in many areas. This project will try to introduce an automatic monitoring system to pest management. Pests often bring huge damages to crops. Based on the results of a world-wide survey conducted in the 1960s, pests account for at least 14 percent of crop damages. Pest monitoring and outbreak prediction are thus critical to pest management. Nowadays, pest control generally involves regularly spraying chemical pesticides, but the frequent use of pesticides may lead to the emergence of drug resistance and pesticide poisoning. Moreover, the number of pests is influenced by various environmental parameters, such as temperature, humidity, rainfall, wind speed, and illumination. It is difficult to determine pest proliferation and the risks brought by pests if solely relying on pest number readings. It is also costly if collecting these data manually. Nor is it easy to acquire the data simultaneously. An automatic monitoring system is therefore a better choice to collect agricultural information, given the fact that such a system can detect a large number of environmental parameters, improve the accuracy of measurements, and solve the problems of high costs, time consuming, and labor intensity. In this regard, it is necessary to develop an integrated, automatic monitoring system to detect a variety of environmental parameters in the field. The purpose of this project is to develop a WSN-based monitoring system for pest population dynamics in agriculture via combining automation, geographic information, wireless sensor networks (WSNs), and GSM wireless communication technologies. This system includes two major parts: a field monitoring network, responsible for collecting environmental parameters and pest density detected by sensor nodes and then sending the data to field gateways, and a remote control center, integrating the acquired data through wireless communication technologies. In this system, all of the field gateways will connect to each other, so they can communicate with each other, avoiding communication barriers caused by terrain obstacles. The remote control center enables a smart control function which monitors all of the field control platforms and analyzes the transmitted sensing data. This project also uses ASP.NET technologies to establish a web control platform, so the entire deployed system can be monitored and managed by the web control platform installed at the remote control center. Furthermore, the remote control center includes a data exhibition and analysis platform which not only allows users to query sensing statistics but also provides pest density and related geographic information. Such an integrated platform will become a critical tool in pest management. In addition, the sensing data and the number of pest will be analyzed by statistical tests, so the relationships between pest proliferation and various environmental factors can be revealed. Based on the statistical results, an alarm model for epidemic outbreaks will be established. An alarm message can be sent out if an outbreak occurs or will occur. This project is a three-year plan, and the details are described as follows. The First Year： In the first year, the project plans to develop the infrastructure of the proposed system and conduct field tests to examine the feasibility of the system. The main tasks of the year are: (A) assembling and upgrading sensor nodes and developing the trapping and detecting device for Oriental Fruit Fly and Spodoptera litura (Fabricius), the sensor communication module, and the self-powered battery module, (B) designing various types of gateways, including multi-purpose, power-efficient and integrated gateways, based on different demands, (C) setting up a LabVIEW controller, (D) creating a web control platform and a geographic information database, built upon ASP.NET technologies, and (E) system deployment and field testing. The Second Year： In the second year, this project will develop a data exhibition and analysis platform, design communication protocols, and conduct system integration and field testing. Combining with a database, this platform will support a user-friendly website and provide pest population distributions and related geographic information. Through the exhibition of the sensing data via the website, agricultural personnel can acquire pest dynamics information and prevent pest outbreaks in advance. Furthermore, some sensing errors may occur due to the influences of environmental, seasonal, and climate factors. The platform will create an adjustment formula based on the testing results using various regression models. In addition, this project will design stable and energy-efficient communication protocols which are capable of reducing transmission delays. It will also develop communication protocols for the backbone network of the entire WSN-based monitoring system, establish an optimal communication model for field gateways, set up bridges between the gateways and the remote control center, and conduct a string of system integration and field testing. The third Year： In the third year, this project will conduct analyses on pest population dynamics and establish a pest outbreak model to provide an alarm to agricultural personnel. The long-term sensing data, collected by the proposed WSN-based monitoring system, will bring a very important insight into pest outbreaks. By analyzing these data, this project will establish a pest outbreak model. An outbreak prediction can be made based on the parameters provided by the model, so the crop damage or loss caused by pest outbreaks can be effectively minimized.誘引劑自動化技術生態環境GSM無線通訊技術東方果實蠅斜紋夜盜蛾無線傳輸無線感測器網路。AttractantsAutomationEcological environmentsGSM wireless communication technologiesOriental Fruit FlySpodoptera litura (Fabricius)Wireless transmissionWireless sensor networks.