Multi-Source Domain Generalization Tool Wear Prediction Based on Wide Convolution Weighted Antagonism

鉴别器 一般化 特征(语言学) 计算机科学 卷积(计算机科学) 反向传播 人工神经网络 模式识别(心理学) 时域 人工智能 卷积神经网络 刀具磨损 均方误差 算法 数学 机械加工 统计 工程类 计算机视觉 哲学 数学分析 探测器 机械工程 电信 语言学
作者
Honghao Fu,Z.C. Li,Xiaoping Xiao,Kai Zhang,Zhou Wen-jun
出处
期刊:Measurement Science and Technology [IOP Publishing]
标识
DOI:10.1088/1361-6501/ad95ad
摘要

Abstract While traditional deep learning models achieve high accuracy in predicting tool wear under consistent working conditions, actual production processes frequently involve varying conditions due to different processing methods. The wear data of different working conditions have a large difference in distribution, so that the wear signal of milling cutter trained in one working condition can only predict the wear value of the working condition, which will cause a large waste of material and manpower for actual production. Therefore, a domain generalization method utilizing Wide Deep Convolutional Neural Networksand weighted antagonistic multi-source domain generalization is introduced for tool wear prediction, and MDG-WDCNNWAL model is constructed. After filtering and de-noising the original data, it is input into WDCNN for automatic feature learning, and the resulting multi-source feature signals are input into the prediction model and discriminator respectively to obtain corresponding loss values. Furthermore, wasserstein distance is used to measure the probability distribution distance of multi-source feature signals as a weighted value combined with the discriminator loss function, and the final loss value is obtained by adding together. Finally, the tool wear prediction model is set up using a backpropagation neural network. Various milling wear experimental data along with the NASA tool wear public dataset were utilized to assess the predictive performance of the trained model. The model achieves a high level of generalization in tool wear prediction, with average RMSE of 0.0599 and 0.2075, as well as average R2 of 0.9037 and 0.9196 on the NASA and self-collecting datasets respectively. To validate its exceptional precision in generalization performance, comparative experiments were conducted with other data-driven methods resulting in over a 50% reduction in average RMSE and more than a 40% increase in average R2.
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
更新
大幅提高文件上传限制,最高150M (2024-4-1)

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
莹莹完成签到 ,获得积分10
刚刚
久久一零完成签到,获得积分10
刚刚
可爱的函函应助加油鸭采纳,获得10
刚刚
ebao完成签到 ,获得积分10
刚刚
zuijiasunyou完成签到,获得积分10
1秒前
1秒前
1秒前
1秒前
priss111应助强健的妙菱采纳,获得30
2秒前
2秒前
3秒前
3秒前
一一应助耿影影采纳,获得10
3秒前
tn完成签到,获得积分10
4秒前
大模型应助lxy采纳,获得10
4秒前
5秒前
cc完成签到,获得积分10
6秒前
7秒前
7秒前
酱肉叉烧包完成签到,获得积分10
7秒前
受伤的冰海完成签到 ,获得积分10
7秒前
9秒前
9秒前
9秒前
神勇从波发布了新的文献求助10
10秒前
11秒前
11秒前
wzy完成签到 ,获得积分10
12秒前
12秒前
熊本熊完成签到,获得积分10
13秒前
小二郎应助王森采纳,获得10
14秒前
wangyu发布了新的文献求助10
14秒前
Tiffany发布了新的文献求助10
15秒前
传奇3应助taotao采纳,获得10
15秒前
15秒前
柯科研发布了新的文献求助10
15秒前
16秒前
Ava应助四十四次日落采纳,获得10
16秒前
闪闪的河马完成签到,获得积分10
17秒前
17秒前
高分求助中
歯科矯正学 第7版(或第5版) 1004
Smart but Scattered: The Revolutionary Executive Skills Approach to Helping Kids Reach Their Potential (第二版) 1000
Semiconductor Process Reliability in Practice 720
PraxisRatgeber: Mantiden: Faszinierende Lauerjäger 700
ACUTE EFFECTS OF MYOFASCIAL RELEASE TECHNIQUE ON FLEXIBILITY AND PAIN: OUTCOME FOR CHRONIC LOW BACK PAIN 500
GROUP-THEORY AND POLARIZATION ALGEBRA 500
Mesopotamian divination texts : conversing with the gods : sources from the first millennium BCE 500
热门求助领域 (近24小时)
化学 医学 生物 材料科学 工程类 有机化学 生物化学 物理 内科学 纳米技术 计算机科学 化学工程 复合材料 基因 遗传学 催化作用 物理化学 免疫学 量子力学 细胞生物学
热门帖子
关注 科研通微信公众号,转发送积分 3227367
求助须知:如何正确求助?哪些是违规求助? 2875446
关于积分的说明 8191043
捐赠科研通 2542695
什么是DOI,文献DOI怎么找? 1372977
科研通“疑难数据库(出版商)”最低求助积分说明 646618
邀请新用户注册赠送积分活动 621040