【作者】 高俊；

【导师】 孟昭鹏；

【作者基本信息】 天津大学 ， 计算机系统结构， 2010， 硕士

【摘要】 膏药作为中药的四大剂型之一,有着悠久的历史,在治疗一些疾病有很好的效果。本文是以科技部项目“中药新型经皮给药系统关键技术与评价”为背景研制一套“基于传感器网络技术的贴膏剂智能成型设备”。从而形成具有自主知识产权的专利技术,对中药外用贴剂的生产工艺做出重大革新,有利于对我国传统中药贴剂生产技术的稳步推进和技术标准的建立。进一步提升中药经皮给药系统的科技含量和产品标准,促进中药经皮给药的标准化、工程化和现代化发展。本文研究的中药贴膏粘度检测主要在适合项目要求的前提下解决实时在线检测的问题。本文在成型设备中的中药贴膏粘度在线检测方面进行了研究,从中药膏体粘度的特性、传统测粘度的方法、中药成型设备要求等方面分析,设计一套适合中药粘度在线检测反馈控制的设备。在硬件上,电机、上位机、PLC、传感器等形成一套完整的检测控制方案。在软件上,实现了PLC实现数据采集、PLC与上位机的通讯、计算并控制粘度等流程。最后,用旋转电流法系统进行了数据采集,并对数据建立最小二乘模型和BP神经网络模型,设计了合理的拟合计算过程,并对模型的预测结果进行比较分析。通过比较,BP神经网络模型更符合中药贴膏的精度要求,也方便今后进一步提高精度。并用VC++、MATLAB和数据库实现了BP神经网络的预测算法程序。课题设计的系统可以准确的对中药生产过程中的粘度参数进行实时在线检测,及时反馈到PLC和计算机中进行处理和实时控制,确保膏药的质量,实现生产的自动化。其次,中药粘度的在线检测为粘度的在线检测提供了一种新的方法,可应用在其他的工业领域实时在线检测和控制,从而提高生产效率,保证产品的质量。更多还原

【Abstract】 Plaster as one of the four traditional Chinese medicine has a long history and positive impact on the treatment of some diseases. This article develops a "sensor network technology based on intelligent shaping equipment of plaster.", which is based on Science and Technology Project," Chinese new plaster systems and evaluation of key technologies". This will improve the patented technology with independent intellectual property rights and make significant innovation of traditional Chinese medicine in China. It will improve the progress of production technology and the establishment of technical standards. It will further enhance technology and product standards of plaster system, and promote the standardization and modernization of Chinese medicine.This study mainly solves the problem of real-time online plaster viscosity’s detection according to the requirements of project .We have done lots of research about Chinese plaster viscosity of online detection including viscosity characteristics of plaster, the traditional methods of measuring liquid’s viscosity, the requirements of project, etc. Then we design a set of equipment which can measure the plaster’s viscosity and control with feedback. In terms of hardware, the equipment consists of motor, PC, PLC and sensors and so on .In terms of software, PLC acquires data from sensor, PC computes and send data to PLC by communication, and then PLC control the viscosity according to the default parameter. Finally, we get data by the device, build model by least squares method and BP neural network model, and design the fitting and calculation process. And then the model predictions were compared. By comparison, BP neural network is more consistent with precision of plaster’s viscosity. Besides, it is convenient to improve the plaster’s viscosity accuracy in the future. And BP neural network prediction algorithm program is completed by VC++,MATLAB and database.The system can accurately get the on line detection viscosity and timely feedback to the PLC and PC for processing and real-time controlling to ensure the quality of the plaster and improve the automation of production. Secondly, the online detection of Chinese plaster’s viscosity provides a new method which can be applied in other industries to real-time online monitor and control viscosity to enhance efficiency of production and ensure product’s quality.更多还原