English

Journal of Chinese Agricultural Mechanization

Journal of Chinese Agricultural Mechanization ›› 2025, Vol. 46 ›› Issue (4): 163-170.DOI: 10.13733/j.jcam.issn.2095-5553.2025.04.024

• Facilities Agriculture and Plant Protection Machinery Engineering • Previous Articles     Next Articles

Design and experiment of vaccine injection machine for spindle shaped fish based on visual technology#br#

Zhang Dongdong, Wang Shuchun, Xu Jinfeng, Sun Guodong, Cai Guangyi, Bao Qihang   

  1. (Institute of Intelligent Manufacturing, Zhejiang Polytechnic University of Mechanical and Electrical Engineering, Hangzhou, 310053, China)
  • Online:2025-04-15 Published:2025-04-18

基于视觉技术的纺锤形鱼类疫苗注射机设计与试验

张冬冬,王树春,徐金峰,孙国栋,蔡广义,包启航   

  1. (浙江机电职业技术大学智能制造学院,杭州市,310053)
  • 基金资助:
    浙江省新苗人才计划项目(2024R456A002)

Abstract: Aiming at the problem that the vaccine injection in domestic fish culture mainly depends on manual injection, a kind of fish vaccine injection machine composed of the flip module, clamping module and injection module was designed. The structure of each module and the overall three dimensional model of the equipment was designed with SolidWorks software. The finite element analysis of clamping action of the virtual fry was carried out based on the ANSYS, the maximum equivalent stress was about 0.14 MPa, and the maximum deformation was about 1.1 mm, which met the requirements of safety clamping. The control system was built based on PLC, and the visual recognition platform of OpenMV which based on Python was used to identify the length of fish fry and the characteristics of fish belly or back through algorithms. According to the design results, the prototype was made and tested. The results indicate that the calculated length of fish fry within the allowable error range can be obtained, and stable recognition of fish belly and back features can be achieved by using visual recognition technology, which will reduce the need of fry delivery and improve the injection efficiency. The physical prototype can successfully inject vaccines into fish fry within the design size range of 80-120 mm, and the success rate of vaccine injection can reach 100% when the fish fry reaches a length of about 100 mm to 120 mm. This study provides a design scheme for the development of fish vaccine injection equipment and provides a basis for the subsequent optimization and testing of the equipment.

Key words: vaccine injection machine, virtual fish, visual recognition, clamping module, prototype testing

摘要: 针对国内鱼类养殖过程中疫苗注射主要依靠人工的问题,设计一种由翻转模块、夹紧模块和注射模块等组成的鱼类疫苗注射机。运用SolidWorks软件对各模块进行结构设计并建立设备的整体三维模型。基于ANSYS软件对鱼苗夹紧模块进行有限元仿真分析,得到虚拟鱼苗的最大等效应力约为0.14 MPa,最大变形量约为1.1 mm,符合安全夹紧要求。搭建以PLC为核心的控制系统,采用基于Python的OpenMV视觉识别平台,通过算法识别鱼苗的长度及鱼腹或鱼背特征。根据设计结果制作实物样机进行试验验证,结果表明:通过视觉识别技术可以获得误差允许范围内的鱼苗计算长度并稳定地识别鱼腹和鱼背特征,降低鱼苗投放要求的同时提升注射效率,实物样机对尺寸为80~120 mm的鱼苗均能成功进行疫苗注射,且当鱼苗体长在100~120 mm时,设备的疫苗注射成功率可达100%。为鱼类疫苗注射设备的研制提供设计方案并为设备后续优化和试验提供基础。

关键词: 疫苗注射机, 虚拟鱼苗, 视觉识别, 夹紧模块, 样机试验

CLC Number: