【作者】 王薇；

【导师】 张书慧；

【作者基本信息】 吉林大学 ， 农业电气化与自动化， 2011， 硕士

【摘要】 作物产量是农民经济收益的重要标志,是评估农业生产效率的重要因素,还是指导科学的田间管理的重要依据。玉米是世界上分布最广泛的粮食作物之一,它在中国的播种面积很大。中国年玉米产量仅次于美国占世界第二位,因此,提高玉米产量是增加农民收入和满足国民需要的关键。实时获取准确的玉米小区产量信息和据此得到的小区产量分布图有利于提高田间精准管理能力和管理决策措施的有效性,是提高玉米产量的关键技术。但中国绝大部分地区玉米在收获时的含水率≥30%,不能进行玉米直接脱粒作业,所以不能直接得到玉米产量数据和玉米产量图。为了实时准确的获取玉米小区产量信息和小区产量分布图,及时了解地块产量差异及其分布,本文以ARM920T-S3C2410微处理器为核心,开发了一套玉米智能测产成图系统。本文结合国家“863”计划“玉米精准作业系统研究与应用”之子项目“玉米变量施肥喷药和智能测产技术研究”(合同号：2006AA10A309-6)和吉林大学研究生创新计划“基于GPS/PFA的穗状玉米智能测产系统研究与示范”(20091017)进行玉米智能测产成图系统研制与开发。研究的主要工作与结果如下：1)建立玉米产量模型。为了建立适合于穗状玉米智能测产的玉米产量模型,本文组建了玉米收获试验装置,确定了收获机进行玉米田间收获的升运器轴速。在2009年10月吉林农业大学试验田,使用玉米测产装置,在玉米收获机行进速度为2km/h、升运器轴速为550r/min的条件下进行了玉米收获试验。通过研究该测产装置测得的玉米穗产量与实际玉米穗产量之间的关系从而获得了预测玉米穗产量关系式,结合该关系式和玉米穗粒转换关系式进而得到玉米产量模型。2)构建玉米智能测产成图系统开发平台。本文选择ARM920T-S3C2410微处理器作为玉米智能测产成图系统硬件平台核心,构建了田间玉米实时测产成图系统硬件平台运行环境；选择WinCE操作系统作为玉米智能测产成图系统软件平台,并自主定制和运行了WinCE操作系统。3)开发玉米智能测产成图系统。本文基于玉米智能测产成图系统平台基础上,采用模块化程序设计原则,使用EVC4.0开发工具和C++语言开发设计了玉米智能测产成图系统。系统主要由数据输入／输出模块、玉米产量转换模块、田块轮廓图生成模块、网格划分模块、玉米产量分布图生成模块、图例生成模块、玉米产量直方图生成模块、地图操作模块组成。4)玉米智能测产成图系统应用。本文将玉米智能测产成图系统嵌入到穗状玉米产量监测器中,在2010年10月吉林省吉林农业大学试验田进行了玉米实时测产成图试验。试验结果表明：本系统能够根据实时读取的玉米穗产量信息和田块参数信息,计算得到玉米产量信息,绘制田块网格图、玉米产量分布图和玉米产量直方图。本文研究的创新点如下：1)建立了适合于穗状玉米智能测产的玉米产量模型,在玉米智能测产成图系统中使用该玉米产量模型能够实时将玉米穗产量转换为玉米产量。2)以ARM920T-S3C2410微处理器为核心开发了一套玉米智能测产成图系统。将该系统嵌入到穗状玉米产量监测器中,玉米收获作业时能够实现实时读取、处理原始数据信息,通过玉米产量模型实时获取玉米产量,实时绘制田块网格图、玉米产量栅格图、玉米产量图例、玉米产量直方图等功能。更多还原

【Abstract】 Crop yield is an important symbol of the economic benefits of farmers, an important factor of assessing agricultural productivity, and an important basis of guiding the scientific field management. Corn is one of the world’s most widely-distributed crops, which is planted in most of areas in China. In China, corn yearly yield accounts for the second place in the world, only after the United States. Therefore, increasing corn yield is the key to improve the income of farmers and meet the national demands. According to the accurate, real-time information of corn yield of grid, the yield distribution map can be accomplished, which can be used to improve the ability of fields precision management and the effectiveness of decision-making measurement. That is the key technology of improving the corn yield. However, in most areas in China, the corn moisture content at the time of harvest is more than 30%, which can not be directly threshed, so corn yield and corn yield maps can not be obtained directly. In order to obtain the accurate, real-time information on corn yield in every grid, draw the yield map and timely understand the difference and distribution of corn yield between each grid, this paper developed a corn yield mapping system based on ARM920T-S3C2410 microprocessor in this paper.This paper has done some research on the development and practice of corn yield mapping system under the program of "Study on Variable Rate Fertilization, Spaying and Intelligent Yield Monitor Technology for Corn"(2006AA10A309-6) which was granted by the states 863 Program of "Research and Application of Precision Operating System for Corn" and under the program of "Study and Demonstration on Intelligent Yield Monitor System of Corn Ear Based on GPS/ PFA"(20091017)which belongs to Jilin University Graduate Innovation Plan. The research results are as follows:1) Establish the corn yield model. In this paper, the set of equipment for corn-harvest practice was built to establish the model used to detect the corn yield intelligently, by which the elevator speed was tested and selected while harvesting in the field. In October,2009, corn-harvest experiment under the conditions of harvester speed 2km/h and elevator speed 550r/min, was carried out in the experimental fields of Jilin Agricultural University in Changchun, using the equipment for corn-harvest practice. In the experiment, the corn ear yield was detected and recorded. By studying the relation between the recorded corn ear yield and the actual corn ear yield, the formula can be determined to predict the corn ear yield. Then, based on that formula and the corn ear-kernel transform formula, the corn yield model can be established.2) Build up the platform for developing yield mapping system. The ARM920T-S3C2410 microprocessor is chosen as the core of the yield mapping system, which builds up the hardware environment.And the customized WinCE operating system was chosen to build up the software development enrollment for yield mapping system.3) Design and develop the yield mapping system. After building up the hardware and software platform, the yield mapping system was completed by using EVC4.0 developing tool and C++language based on the modular program developing principle. The yield mapping system mainly included data input/output module, the yield conversion module, meshing module, the yield grid map gridding module, the legend editor module, yield data analysis module, the map operation modules and other components.4) Apply the corn yield mapping system. This corn yield mapping system is embedded in the yield monitor system in our lab and used for the corn harvest test in the experimental fields in Changchun Agricultural University in October,2010. The results showed that the yield mapping system can read the real-time information of corn ear yield and plots parameter information, calculate the corn yield,and draw the grid map, yield distribution map and yield histogram.The innovations of this paper are as follows:1) The corn yield model is established to predict the corn yield intelligently, in which the corn ear yield can be transferred into the corn yield in real-time.2) The corn yield mapping system was developed based on the ARM920T-S3C2410 microprocessor. It can be embedded in the yield monitor system and can accomplish the functions like reading and processing the real-time original data, recording the real-time corn yield, gridding the field, drawing the grid map, yield distribution map and yield histogram.更多还原