基于卷积神经网络算法的稀土酸度自动滴定技术研究  

Research on Automatic Titration of Rare Earth Acidity Based on Convolutional Neural Network Algorithm

作　　者：曹靖[1] 张帅[1] 陈吉文 Cao Jing;Zhang Shuai;Chen Jiwen(North China University of Technology,Beijing 100144,China)

机构地区：[1]北方工业大学,北京100144

出　　处：《实验与分析》2024年第2期1-5,共5页LABOR PRAXIS

基　　金：北方工业大学“有组织科研”创新项目(110051360023XN278-06)。

摘　　要：如今稀土产业发展迅速,市场需求越来越大,应用的范围也越来越广,需建立一个简便易操作且适用于测定各类稀土酸度的方法。目前传统稀土酸度检测方法存在效率低、准确度低、滴定终点差异大等问题,难以满足实时在线检测的需要。本文提出一种基于卷积神经网络算法的稀土酸度在线分析仪,可以助力在线检测的顺利进行。卷积神经网络算法是通过高清工业摄像头记录样品在滴定过程中的溶液颜色的变化,对溶液进行实时图像特征提取和学习,从而有效、准确地实现化学反应过程中溶液颜色的自动识别,配合步进电机和注射泵等部件实现自动滴定过程。图像识别本质上是对图像信息进行特征提取,而卷积神经网络算法有着传统识别方法不具备的优点,比如能够自行训练、识别速度更快、所需特征更少等。本仪器将自动滴定与卷积神经网络相结合,实现了滴定流程的自动化取样和前处理、滴定过程、终点判定等过程的一体化,且仪器能够同时进行五个样品的滴定试验,很大程度上提高了滴定效率和精度。Nowadays,with the development of rare earth industry,the market demand is getting bigger and bigger,and the scope of application is getting more and more wide.It is necessary to establish a simple and suitable method for the determination of all kinds of rare earth acidity.In view of the current problems of low efficiency,low accuracy and large difference of titration endpoints,it is difficult to meet the needs of real-time online detection.This paper presents an online analyzer of rare earth acidity based on convolutional neural network algorithm.Convolutional neural network algorithm is recorded through the HD industrial camera samples in the process of the solution color change,the solution of real-time image feature extraction and learning,so as to effectively and accurately realize the solution color automation of chemical reaction,with stepper motor and injection pump components to realize automatic titration process.Image recognition is essentially a feature extraction of image information,and the convolutional neural network algorithm has the advantages that the traditional recognition methods do not have,such as self-training,faster recognition speed,and fewer required features.This instrument combines automatic titration with convolutional neural network to realize the integration of automatic sampling and pretreatment of titration process,titration process and end point judgment,and the instrument can carry out the titration test of five samples at the same time,which greatly improves the titration efficiency and accuracy.

关 键 词：卷积神经网络算法 自动滴定 稀土酸度 自动取样 

分 类 号：TP23[自动化与计算机技术—检测技术与自动化装置]