首页|Polytechnic University Torino Researchers Advance Knowledge in Machine Learning (A novel feature engineering approach for predicting melt pool depth during LPBF by machine learning models)

Polytechnic University Torino Researchers Advance Knowledge in Machine Learning (A novel feature engineering approach for predicting melt pool depth during LPBF by machine learning models)

扫码查看
By a News Reporter-Staff News Editor at Robotics & Machine Learning DailyNews Daily News ; Fresh data on artificial intelligence are presented in a new report. According to newsreporting from Torino, Italy, b y NewsRx journalists, research stated, “Melt pool geometry is a deterministicfa ctor affecting the characteristics of metal Additive Manufacturing (AM) componen ts.”The news correspondents obtained a quote from the research from Polytechnic Univ ersity Torino:“The wide array of physical and thermal phenomena involved during the formation of the AM melt pool,along with the great variety of alloy compos itions and AM methods, coupled with the clear influenceof multiple process para meters, make it difficult to predict the melt pool geometry under a given setof conditions. Therefore, using Artificial Intelligence (AI) approaches such as Ma chine Learning (ML)is necessary for accurate predictions. Using a physics-infor med feature selection strategy along with theapplication of atomic features for the first time, this work aims to offer accurately trained models relyingon ex isting high-fidelity data for most common alloys in AM academia and industry, i. e., 316 L stainlesssteel, Ti6Al4V, and AlSi10Mg. Multiple ML algorithms were tr ained, and the results revealed that theaverage R2 and RMSE obtained by the K-f old cross-validation (K = 5) were significantly enhanced whenlaser and material properties, inspired by the analytical models for AM melt pool geometry, were u sedas the model features. Removing the excess features and applying atomic feat ures further enhanced theaccuracy of the models. As a result, R2 for the XGBoos t, CatBoost, and GPR models were 0.907, 0.889,and 0.882, respectively, while th e hold-out cross-validation led to 0.978, 0.976, and 0.945, respectively.”

Polytechnic University TorinoTorinoI talyEuropeCyborgsEmerging TechnologiesEngineeringMachine Learning

2024

Robotics & Machine Learning Daily News

Robotics & Machine Learning Daily News

ISSN:
年,卷(期):2024.(Sep.6)