An Integrated Experimental and Machine Learning Approach for Machinability Assessment and Tool Life Prediction in Drilling of 14NiCr10 Alloy Using AlTiN-Coated Carbide Tools
IF 2 Q3 COMPUTER SCIENCE, INTERDISCIPLINARY APPLICATIONS
Adik M. Takale, Uday A. Dabade, Manjunath G. Avalappa, Uttam U. Deshpande, Mukesh Kumar
{"title":"An Integrated Experimental and Machine Learning Approach for Machinability Assessment and Tool Life Prediction in Drilling of 14NiCr10 Alloy Using AlTiN-Coated Carbide Tools","authors":"Adik M. Takale, Uday A. Dabade, Manjunath G. Avalappa, Uttam U. Deshpande, Mukesh Kumar","doi":"10.1002/eng2.70397","DOIUrl":null,"url":null,"abstract":"<p>To maximize tool life and process efficiency in high-performance drilling applications, it is necessary to examine the machinability of 14NiCr10 alloy using AlTiN-coated carbide tools with varying cobalt compositions. We propose a method to evaluate tool wear, cutting force, temperature, and the number of holes drilled before resharpening during experimental trials in dry cutting settings. In comparison to normal tools, carbide tools with a higher cobalt content demonstrated better wear resistance, lower thermal load, and a 51% increase in tool life during experiments. To validate the measured mechanical and thermal loads, we carried out simulation tests using Finite Element Analysis (FEA) on temperature distribution, torque, and stress behavior loads. For normal and increased cobalt content tool versions, the values stayed within safe operating limits. We trained the sensor-acquired force and temperature data using the Decision Tree Regressor to create a machine learning-based predictive model, further improving process reliability. With more than 90% of tool life estimates falling within an acceptable error range of ±10%, the model exhibited excellent predictive accuracy. Our method provides a comprehensive hybrid framework for machining high-strength alloys by utilizing a combination of simulation, machine learning, experimental analysis, and enhancing tool performance. Thus, it facilitates predictive maintenance and supports the development of smart manufacturing processes.</p>","PeriodicalId":72922,"journal":{"name":"Engineering reports : open access","volume":"7 9","pages":""},"PeriodicalIF":2.0000,"publicationDate":"2025-09-24","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"https://onlinelibrary.wiley.com/doi/epdf/10.1002/eng2.70397","citationCount":"0","resultStr":null,"platform":"Semanticscholar","paperid":null,"PeriodicalName":"Engineering reports : open access","FirstCategoryId":"1085","ListUrlMain":"https://onlinelibrary.wiley.com/doi/10.1002/eng2.70397","RegionNum":0,"RegionCategory":null,"ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":null,"EPubDate":"","PubModel":"","JCR":"Q3","JCRName":"COMPUTER SCIENCE, INTERDISCIPLINARY APPLICATIONS","Score":null,"Total":0}
引用次数: 0
Abstract
To maximize tool life and process efficiency in high-performance drilling applications, it is necessary to examine the machinability of 14NiCr10 alloy using AlTiN-coated carbide tools with varying cobalt compositions. We propose a method to evaluate tool wear, cutting force, temperature, and the number of holes drilled before resharpening during experimental trials in dry cutting settings. In comparison to normal tools, carbide tools with a higher cobalt content demonstrated better wear resistance, lower thermal load, and a 51% increase in tool life during experiments. To validate the measured mechanical and thermal loads, we carried out simulation tests using Finite Element Analysis (FEA) on temperature distribution, torque, and stress behavior loads. For normal and increased cobalt content tool versions, the values stayed within safe operating limits. We trained the sensor-acquired force and temperature data using the Decision Tree Regressor to create a machine learning-based predictive model, further improving process reliability. With more than 90% of tool life estimates falling within an acceptable error range of ±10%, the model exhibited excellent predictive accuracy. Our method provides a comprehensive hybrid framework for machining high-strength alloys by utilizing a combination of simulation, machine learning, experimental analysis, and enhancing tool performance. Thus, it facilitates predictive maintenance and supports the development of smart manufacturing processes.