An attention-based long short-term memory prediction model for working conditions of copper electrolytic plates

计算机科学 人工神经网络 精炼(冶金) 滞后 人工智能 数据挖掘 材料科学 冶金 计算机网络
作者
Hongqiu Zhu,Lei Peng,Can Zhou,Yusi Dai,Tianyu Peng
出处
期刊:Measurement Science and Technology [IOP Publishing]
卷期号:34 (6): 065202-065202 被引量:2
标识
DOI:10.1088/1361-6501/acc11f
摘要

Abstract Copper is an important source of non-ferrous metals, with electrolytic refining being one of the main methods to produce fine copper. In the electrolytic process, plate states seriously affect the output and quality of the copper. Therefore, timely and accurate prediction of the working states of the plates is of great significance to the copper electrolytic refining process. Aiming at the issues associated with traditional plate state detection algorithms of large lag, poor anti-interference ability and low accuracy, a plate state prediction model based on a long short-term memory (LSTM) neural network with an attention mechanism is here proposed in this paper. The average gray values of the plates in infrared imagery are used to characterize the plates’ working states. To address the problems of large fluctuation and the large amount of time series data required in such a study, a double-layer LSTM neural network structure is used to improve the efficiency and accuracy of model training. Meanwhile, in view of the periodicity of the time series data and the possible correlation between adjacent data, a unique attention mechanism is proposed to enable the model to learn this correlation between the adjacent data so as to improve the accuracy of the model prediction. The experimental results show that the accuracy of the proposed model for plate state prediction reaches 95.11%. Compared with commonly used prediction algorithms, the plate state prediction model proposed in this paper demonstrates stronger prediction ability.
最长约 10秒,即可获得该文献文件

科研通智能强力驱动
Strongly Powered by AbleSci AI
科研通是完全免费的文献互助平台,具备全网最快的应助速度,最高的求助完成率。 对每一个文献求助,科研通都将尽心尽力,给求助人一个满意的交代。
实时播报
1秒前
留胡子的之云完成签到,获得积分20
1秒前
xinl518完成签到,获得积分10
1秒前
1秒前
transition发布了新的文献求助50
1秒前
2秒前
2秒前
2秒前
2秒前
3秒前
3秒前
4秒前
4秒前
优美怀蕊完成签到 ,获得积分20
4秒前
七七发布了新的文献求助10
5秒前
Mzhao发布了新的文献求助10
5秒前
elaine完成签到,获得积分20
5秒前
dms完成签到,获得积分10
5秒前
侯mm完成签到,获得积分10
6秒前
zhuyu完成签到,获得积分10
6秒前
霸气的鹭洋完成签到,获得积分10
6秒前
丘比特应助Muy采纳,获得10
6秒前
7秒前
妮妮完成签到,获得积分10
7秒前
不爱吃芒果完成签到,获得积分10
7秒前
7秒前
7秒前
哈利波特完成签到,获得积分10
7秒前
wonderingria完成签到,获得积分10
8秒前
BAEssss发布了新的文献求助10
8秒前
8秒前
9秒前
小蘑菇应助申改改采纳,获得10
9秒前
xiwaiwai发布了新的文献求助10
9秒前
刘gg发布了新的文献求助10
9秒前
Sophist完成签到,获得积分10
9秒前
1233333发布了新的文献求助20
10秒前
10秒前
10秒前
10秒前
高分求助中
【此为提示信息,请勿应助】请按要求发布求助,避免被关 20000
Production Logging: Theoretical and Interpretive Elements 3000
CRC Handbook of Chemistry and Physics 104th edition 1000
Density Functional Theory: A Practical Introduction, 2nd Edition 840
J'AI COMBATTU POUR MAO // ANNA WANG 660
Izeltabart tapatansine - AdisInsight 600
Gay and Lesbian Asia 500
热门求助领域 (近24小时)
化学 材料科学 医学 生物 工程类 有机化学 物理 生物化学 纳米技术 计算机科学 化学工程 内科学 复合材料 物理化学 电极 遗传学 量子力学 基因 冶金 催化作用
热门帖子
关注 科研通微信公众号,转发送积分 3756055
求助须知:如何正确求助?哪些是违规求助? 3299291
关于积分的说明 10109581
捐赠科研通 3013845
什么是DOI,文献DOI怎么找? 1655326
邀请新用户注册赠送积分活动 789704
科研通“疑难数据库(出版商)”最低求助积分说明 753361