Research On Grain Yield Information Collection, Processing And System Integration Technology

This research was carried out at the Key Laboratory for Modern Precision Agriculture System Integration Research of Ministry of Education, the Research Centre for Precision Agriculture of China Agricultural University, and supported by the National Development Project Committee and Beijing Government Project - Hi-tech Industrialization Demonstration, Beijing Precision Agriculture Demonstration Project (Project No.A00300100584).Obtaining crop yield data and generating yield map are very important basis of precision agriculture. In order to acquire accurate yield map, AFS yield monitor system of CASE IH Corporation was analyzed and studied, four seasons' harvest experiments were carried out, yield data analysis methods and results were discussed, and a grain yield mapping system including yield data processing and yield mapping generating software were developed. The main content of research include five aspects as follow:1. Through harvesting winter wheat, corn and soybean using combine harvester equipped with AFS Yield Monitor System, the component of yield monitor system and working principle have been well understood, and four seasons yield data have been acquired. Experiment result show that, for grain flow sensor and moisture sensor, measurement errors can be effectively reduced by calibration; but cutting width sensor is not available at present, so cutting width have to be estimated manually, then coarse error will be introduced.2. The information provided by yield map must be accurate enough for site-specific crop management. In order to improve the reliability of the yield map, method of recognition coarse error, systematic error and random error in yield data have been discussed. Three kinds of main errors of yield data were described include: cutter width setting error, filling time error and delay time error. How these errors occur, how to recognize and process these errors, and how the yield map influenced by these errors are discussed. The study result showed that compensation of the deleted yield data is not necessary; after deleting yield data with these errors, the generated yield map was smoother and clustered better.3. A method of analysis temporal variability of yield data and management zoning map has been discussed. According to the management zone map, there is little potential to improve yield of the yield stable area; but to the yield unstable area, fertilizing, irrigation management and soil properties need to be analyzed carefully, in order to improve the crop yield.4. A yield data processing and yield map generating software system-YMapper was designed and developed. In order to generate available and useful yield maps from yield data gathered by combine harvester yield monitoring system, a local-averaging interpolation method and yield data filter methods were studied. Approaches of coordinates definition, yield data classification and statistics, yield map rendering and plotting were discussed. The test result showed that the system was steady, and could be a tool for yield data processing and yield map generations.5. Design principle of yield monitor system was discussed, and some key techniques were analyzed, mainly include: sensors choosing, total error distributing, display terminaldesigning and developing, yield lag time and yield calculate model building. Primary field test results show that system design method is suitable.