Hilal Singer , Abdullah C. İlçe , Yunus E. Şenel , Erol Burdurlu
{"title":"基于人工神经网络的预测模型,最大限度减少加工过程中的粉尘排放","authors":"Hilal Singer , Abdullah C. İlçe , Yunus E. Şenel , Erol Burdurlu","doi":"10.1016/j.shaw.2024.06.006","DOIUrl":null,"url":null,"abstract":"<div><h3>Background</h3><p>Dust generated during various wood-related activities, such as cutting, sanding, or processing wood materials, can pose significant health and environmental risks due to its potential to cause respiratory problems and contribute to air pollution. Understanding the factors influencing dust emission is important for devising effective mitigation strategies, ensuring a safer working environment, and minimizing environmental impact. This study focuses on developing an artificial neural network (ANN) model to predict dust emission values in the machining of black poplar (<em>Populus nigra</em> L.), oriental beech (<em>Fagus orientalis</em> L.), and medium-density fiberboards.</p></div><div><h3>Methods</h3><p>The multilayer feed-forward ANN model is developed using a customized application built with MATLAB code. The inputs to the ANN model include material type, cutting width, number of blades, and cutting depth, whereas the output is the dust emission. Model performance is assessed through graphical and statistical comparisons.</p></div><div><h3>Results</h3><p>The results reveal that the developed ANN model can provide adequate predictions for dust emission with an acceptable level of accuracy. Through the implementation of the ANN model, the study predicts intermediate dust emission values for different cutting widths and cutting depths, which are not considered in the experimental work. It is observed that dust emission tends to decrease with reductions in cutting width and cutting depth.</p></div><div><h3>Conclusion</h3><p>This study introduces an alternative approach to optimize machining-process conditions for minimizing dust emissions. The findings of this research will assist industries in obtaining dust emission values without the need for additional experimental activities, thereby reducing experimental time and costs.</p></div>","PeriodicalId":56149,"journal":{"name":"Safety and Health at Work","volume":"15 3","pages":"Pages 317-326"},"PeriodicalIF":3.5000,"publicationDate":"2024-09-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"https://www.sciencedirect.com/science/article/pii/S2093791124000520/pdfft?md5=b917c5a3a9257fc077566a0282125bb8&pid=1-s2.0-S2093791124000520-main.pdf","citationCount":"0","resultStr":"{\"title\":\"Artificial Neural Network–based Prediction Model to Minimize Dust Emission in the Machining Process\",\"authors\":\"Hilal Singer , Abdullah C. İlçe , Yunus E. Şenel , Erol Burdurlu\",\"doi\":\"10.1016/j.shaw.2024.06.006\",\"DOIUrl\":null,\"url\":null,\"abstract\":\"<div><h3>Background</h3><p>Dust generated during various wood-related activities, such as cutting, sanding, or processing wood materials, can pose significant health and environmental risks due to its potential to cause respiratory problems and contribute to air pollution. Understanding the factors influencing dust emission is important for devising effective mitigation strategies, ensuring a safer working environment, and minimizing environmental impact. This study focuses on developing an artificial neural network (ANN) model to predict dust emission values in the machining of black poplar (<em>Populus nigra</em> L.), oriental beech (<em>Fagus orientalis</em> L.), and medium-density fiberboards.</p></div><div><h3>Methods</h3><p>The multilayer feed-forward ANN model is developed using a customized application built with MATLAB code. The inputs to the ANN model include material type, cutting width, number of blades, and cutting depth, whereas the output is the dust emission. Model performance is assessed through graphical and statistical comparisons.</p></div><div><h3>Results</h3><p>The results reveal that the developed ANN model can provide adequate predictions for dust emission with an acceptable level of accuracy. Through the implementation of the ANN model, the study predicts intermediate dust emission values for different cutting widths and cutting depths, which are not considered in the experimental work. It is observed that dust emission tends to decrease with reductions in cutting width and cutting depth.</p></div><div><h3>Conclusion</h3><p>This study introduces an alternative approach to optimize machining-process conditions for minimizing dust emissions. The findings of this research will assist industries in obtaining dust emission values without the need for additional experimental activities, thereby reducing experimental time and costs.</p></div>\",\"PeriodicalId\":56149,\"journal\":{\"name\":\"Safety and Health at Work\",\"volume\":\"15 3\",\"pages\":\"Pages 317-326\"},\"PeriodicalIF\":3.5000,\"publicationDate\":\"2024-09-01\",\"publicationTypes\":\"Journal Article\",\"fieldsOfStudy\":null,\"isOpenAccess\":false,\"openAccessPdf\":\"https://www.sciencedirect.com/science/article/pii/S2093791124000520/pdfft?md5=b917c5a3a9257fc077566a0282125bb8&pid=1-s2.0-S2093791124000520-main.pdf\",\"citationCount\":\"0\",\"resultStr\":null,\"platform\":\"Semanticscholar\",\"paperid\":null,\"PeriodicalName\":\"Safety and Health at Work\",\"FirstCategoryId\":\"3\",\"ListUrlMain\":\"https://www.sciencedirect.com/science/article/pii/S2093791124000520\",\"RegionNum\":3,\"RegionCategory\":\"医学\",\"ArticlePicture\":[],\"TitleCN\":null,\"AbstractTextCN\":null,\"PMCID\":null,\"EPubDate\":\"\",\"PubModel\":\"\",\"JCR\":\"Q1\",\"JCRName\":\"PUBLIC, ENVIRONMENTAL & OCCUPATIONAL HEALTH\",\"Score\":null,\"Total\":0}","platform":"Semanticscholar","paperid":null,"PeriodicalName":"Safety and Health at Work","FirstCategoryId":"3","ListUrlMain":"https://www.sciencedirect.com/science/article/pii/S2093791124000520","RegionNum":3,"RegionCategory":"医学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":null,"EPubDate":"","PubModel":"","JCR":"Q1","JCRName":"PUBLIC, ENVIRONMENTAL & OCCUPATIONAL HEALTH","Score":null,"Total":0}
引用次数: 0
摘要
背景在各种与木材有关的活动(如切割、打磨或加工木质材料)中产生的粉尘可能会引起呼吸道问题并造成空气污染,从而对健康和环境构成重大威胁。了解影响粉尘排放的因素对于制定有效的缓解策略、确保更安全的工作环境以及最大限度地减少对环境的影响非常重要。本研究的重点是开发一个人工神经网络(ANN)模型,用于预测黑杨木(Populus nigra L.)、东方山毛榉(Fagus orientalis L.)和中密度纤维板加工过程中的粉尘排放值。ANN 模型的输入包括材料类型、切割宽度、刀片数量和切割深度,而输出则是粉尘排放量。结果结果表明,所开发的 ANN 模型能够以可接受的准确度对粉尘排放进行充分预测。通过实施 ANN 模型,该研究预测了不同切割宽度和切割深度的中间粉尘排放值,而这些在实验工作中并未考虑。据观察,随着切削宽度和切削深度的减小,粉尘排放量呈下降趋势。 结论 本研究引入了另一种优化加工工艺条件的方法,以最大限度地减少粉尘排放。研究结果将有助于工业界在无需额外实验活动的情况下获得粉尘排放值,从而减少实验时间和成本。
Artificial Neural Network–based Prediction Model to Minimize Dust Emission in the Machining Process
Background
Dust generated during various wood-related activities, such as cutting, sanding, or processing wood materials, can pose significant health and environmental risks due to its potential to cause respiratory problems and contribute to air pollution. Understanding the factors influencing dust emission is important for devising effective mitigation strategies, ensuring a safer working environment, and minimizing environmental impact. This study focuses on developing an artificial neural network (ANN) model to predict dust emission values in the machining of black poplar (Populus nigra L.), oriental beech (Fagus orientalis L.), and medium-density fiberboards.
Methods
The multilayer feed-forward ANN model is developed using a customized application built with MATLAB code. The inputs to the ANN model include material type, cutting width, number of blades, and cutting depth, whereas the output is the dust emission. Model performance is assessed through graphical and statistical comparisons.
Results
The results reveal that the developed ANN model can provide adequate predictions for dust emission with an acceptable level of accuracy. Through the implementation of the ANN model, the study predicts intermediate dust emission values for different cutting widths and cutting depths, which are not considered in the experimental work. It is observed that dust emission tends to decrease with reductions in cutting width and cutting depth.
Conclusion
This study introduces an alternative approach to optimize machining-process conditions for minimizing dust emissions. The findings of this research will assist industries in obtaining dust emission values without the need for additional experimental activities, thereby reducing experimental time and costs.
期刊介绍:
Safety and Health at Work (SH@W) is an international, peer-reviewed, interdisciplinary journal published quarterly in English beginning in 2010. The journal is aimed at providing grounds for the exchange of ideas and data developed through research experience in the broad field of occupational health and safety. Articles may deal with scientific research to improve workers'' health and safety by eliminating occupational accidents and diseases, pursuing a better working life, and creating a safe and comfortable working environment. The journal focuses primarily on original articles across the whole scope of occupational health and safety, but also welcomes up-to-date review papers and short communications and commentaries on urgent issues and case studies on unique epidemiological survey, methods of accident investigation, and analysis. High priority will be given to articles on occupational epidemiology, medicine, hygiene, toxicology, nursing and health services, work safety, ergonomics, work organization, engineering of safety (mechanical, electrical, chemical, and construction), safety management and policy, and studies related to economic evaluation and its social policy and organizational aspects. Its abbreviated title is Saf Health Work.