奶牛个体反刍行为监测技术研究

发布时间：2018-06-04 04:00

本文选题：奶牛 + 反刍监测　；参考：《东北农业大学》2017年硕士论文

【摘要】：畜牧业的每一次飞跃,都伴随着一场技术革新,快速发展的同时,也暴露出当前我国牧场养殖体系的缺口,虽然已经成为第三大奶业生产国家,但我国依旧存在牛源紧缺,牧场管理方式落后,奶制品供给量达不到标准等问题,并且我国目前多数大规模牧场引进国外先进设备对奶牛反刍行为进行实时监测,但其技术垄断,研发成本高昂。为此,本文研究了奶牛个体反刍行为检测设备,实现成本较低,可自动化检测奶牛反刍行为,从而掌握奶牛个体发情及身体健康情况,减少牧场人力物力,提升牧场工作效率,增添牧场收益。首先,对奶牛个体反刍行为检测设备进行总体方案设计。主要包括设计需求、功能分析、总体设计以及工作原理,同时,对设备的抗干扰性进行设计,并从技术可行性以及经济可行性对设备进行可行性分析。随后,对奶牛个体反刍行为检测设备进行硬件系统的详细设计。选用MSP430F149型号单片机作为该设备的主处理器,LM386声音传感器采集奶牛反刍及采食时的声音信号,并为此设备设计信号调理模块,采用场效应管高输入阻抗差动放大器进行声音信号的放大,选用二阶带通有源滤波器对声音信号进行滤波,应用mocro SDHC UHS-l存储卡对信息进行存储。因为奶牛反刍多数发生在夜晚并且不宜受到外界的干扰,所以数据传输模块选用基于Zig Bee的无线传输方式。采用RFID射频技术对奶牛个体身份进行识别。本设备的设计以低功耗为原则,故又对设备的供电方式进行全面分析,最终选取锂亚电池对设备进行供电。最后,对奶牛个体反刍声音识别算法进行研究设计并移植到奶牛个体反刍行为检测设备中进行模拟仿真实验。应用Cool Edit Pro软件对奶牛反刍及采食行为的声音频谱进行特征分析,应用预加重以及加汉明窗方式将声音信号做预处理,并获取声音信号短时平均能量以及短时平均过零率,以此对奶牛反刍声音信号进行时域分析。随后应用LPCC倒谱系数对奶牛反刍声音信号进行频域特征提取,得出奶牛反刍声音频率大约在400Hz-1KHz左右。将此算法移植到奶牛个体反刍行为检测设备中,通过对比发现该算法能够较好的检测奶牛的反刍行为,可靠性较高。应用录音仪器在农场采集连续5天同一头奶牛的反刍及采食的声音,应用本设备对录音仪器进行5轮每轮5天的奶牛反刍次数模拟检测,实验结果表明,该设备性能稳定,可靠性较高,精准度达90%,可以有效的对奶牛反刍行为进行检测。综上所述,本文研究的奶牛个体反刍行为检测设备具备奶牛反刍行为的检测、反刍次数的测量以及数据无线传输的功能。应用嵌入式、单片机、传感器、无线传输等技术,实现了自动化、智能化、信息化的监测奶牛反刍行为,为我国畜牧业的长足发展提供了支持。该设备实现成本低、易操作、精度较高、稳定性较好,对奶牛个体的身体健康状况以及发情情况具有良好的指导性作用。
[Abstract]:Each leap forward in animal husbandry has been accompanied by a technological innovation and rapid development, at the same time, it has also exposed the gap in the current pastureland farming system in China. Although it has become the third largest dairy production country, there is still a shortage of cattle resources in our country. The management of pasture is backward, the supply of dairy products is not up to the standard, and most of the large-scale pastures in our country introduce foreign advanced equipment to monitor the ruminant behavior in real time, but its technology monopoly and high cost of research and development. Therefore, this paper studies the equipment of individual ruminant behavior detection in dairy cattle, which can realize low cost and can automatically detect cow ruminant behavior, so as to master individual estrus and body health of dairy cow, reduce manpower and material resources of pasture, and improve the efficiency of pasture work. Increase pasture income. First of all, the overall scheme design of individual ruminant behavior testing equipment is carried out. It mainly includes design requirements, function analysis, general design and working principle. At the same time, the anti-interference ability of the equipment is designed, and the feasibility of the equipment is analyzed from the technical feasibility and economic feasibility. Then, the detailed design of the hardware system for individual ruminant behavior detection equipment of dairy cattle was carried out. The MSP430F149 singlechip was chosen as the main processor of the equipment, and the LM386 sound sensor was used to collect the sound signals of cow ruminating and feeding, and the signal conditioning module was designed for this equipment. The high input impedance differential amplifier of FET is used to amplify the sound signal, the second order band-pass active filter is used to filter the sound signal, and the mocro SDHC UHS-l memory card is used to store the information. Because most cows ruminant occurs at night and is unsuitable for outside interference, the wireless transmission mode based on Zig Bee is used in the data transmission module. RFID radio frequency technology was used to identify the individual identity of dairy cattle. The design of the equipment is based on the principle of low power consumption, so the power supply mode of the equipment is analyzed comprehensively, and the lithium subbattery is selected to supply the power to the equipment. Finally, the algorithm of individual ruminant recognition is studied and designed and transplanted to the individual ruminant behavior detection equipment to simulate the simulation experiment. The acoustic spectrum of ruminant and feeding behavior of dairy cattle was analyzed by Cool Edit Pro software. The sound signal was preprocessed by pre-weighting and Hanming window, and the short time average energy and zero crossing rate of sound signal were obtained. The time domain analysis of the ruminant sound signal was carried out. LPCC cepstrum coefficient was then used to extract the frequency feature of the ruminant sound signal, and the frequency of ruminant sound was about 400Hz-1KHz. The algorithm is transplanted into the equipment for individual ruminant behavior detection of dairy cows. It is found that the algorithm can detect the ruminant behavior of dairy cows better and has higher reliability. The sound of ruminating and feeding of the same cow was collected by recording instrument in farm for 5 consecutive days. The equipment was used to simulate the ruminant times of dairy cows for 5 rounds and 5 days per round. The experimental results showed that the equipment was stable in performance. The reliability is high and the precision is 90, which can effectively detect the ruminant behavior of dairy cows. To sum up, the equipment of individual ruminant behavior detection in this paper has the functions of testing cow ruminant behavior, measuring ruminant times and wireless data transmission. With the application of embedded technology, single chip computer, sensor and wireless transmission technology, automatic, intelligent and informational monitoring of cow ruminant behavior has been realized, which has provided support for the rapid development of animal husbandry in China. The equipment has the advantages of low cost, easy operation, high precision and good stability. It has a good instructive effect on the individual health and estrus of dairy cow.
【学位授予单位】：东北农业大学
【学位级别】：硕士
【学位授予年份】：2017
【分类号】：S823;TP274

【参考文献】