Authors

Abstract

In traditional Chinese Baijiu (white wine) brewing factory, most processes were usually done manually. In order to promote industrial automation in liquor brewing industry and alleviate highly repetitive manual labor, an industrial robot operating system equipped with machine vision was designed to accomplish caldron feeding process during the distillation of Chinese Baijiu. Firstly, Modified D-H method was used to establish robot kinematics model. Shovel actuators and peripheral devices were designed to spread and transport grains during caldron feeding process. After that, an online detection system based on machine vision and master computer were designed to control key parameters of caldron feeding process, including grains height, caldron pressure, robot speed and waiting time, etc. Finally, multiple brewing experiments were carried out on crushed grains and spent grains. The experimental results showed that robot system maximum working space was 3.15 m, which could feed 3 caldrons at the same time. During Chinese Baijiu distillation, air pressure average error in caldron was 8.28 %. High proportion of first-class liquor and second-class liquor was obtained, which met engineering production requirements.

Abstract in Chinese

在中国传统的白酒酿造厂中，大部分工艺流程通常都是由人工完成。为了促进白酒酿造行业的工业自动化发展，减轻高重复性的人工劳作。本研究设计了一个配备机器视觉的工业机器人操作系统来完成中国白酒蒸馏过程中的上甑过程。首先采用改进的D-H方法对机器人进行数学建模，设计了铲式执行机构和外围设备用于上甑过程中大曲的铺料和运输。然后还开发了基于机器视觉的在线检测系统和上位机软件对上甑过程中的关键参数：料层高度，甑内压力，铺料速度和等待时间等进行控制。最后对大渣和丢糟两种原料进行多次酿造实验。实验结果表明：智能上甑机器人系统最大工作空间为3.15m，可同时对3个甑桶进行上甑。白酒蒸馏过程中甑桶内气压平均误差为8.28%。产出一级酒和二级酒的比例较高，符合工程生产要求。