{"title":"利用遗传算法优化混合学习模型预测 3D 打印部件的表面粗糙度","authors":"Gazi Akgun, Osman Ulkir","doi":"10.1177/08927057241243364","DOIUrl":null,"url":null,"abstract":"The final product of additive manufacturing (AM) or 3D printing critically depends on the surface quality. An experimental study on the 3D printed intake manifold flange using acrylonitrile butadiene styrene (ABS) material was executed by varying the four process parameters. A fused deposition modeling (FDM) based 3D printer was used to fabricate the flanges. The association between the parameters and the surface roughness of printed ABS flanges was investigated. A feed forward neural network (FFNN) model trained on particle swarm optimization (PSO) optimized with a genetic algorithm (GA) was used to estimate the surface roughness. A Box-Behnken design (BBD) with printing parameters at three levels was used, and 25 parts were fabricated. The suggested model demonstrated a coefficient of determination (R<jats:sup>2</jats:sup>) of 0.9865 on test values, mean of root-mean-square-error (RMSE) of 0.1231 after 500 times training for generalization. And also mean of overfitting factor is 0.7110. This means that the suggested system could generalize. Comparing the results from the suggested model and ANN, the suggested hybrid model outperformed ANN in predicting the surface roughness values with no overfitting. This suggests that GA optimized PSO based FFNN may be a more suitable method for estimating product quality in terms of surface roughness.","PeriodicalId":17446,"journal":{"name":"Journal of Thermoplastic Composite Materials","volume":"95 1","pages":""},"PeriodicalIF":3.6000,"publicationDate":"2024-04-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":"0","resultStr":"{\"title\":\"Prediction surface roughness of 3D printed parts using genetic algorithm optimized hybrid learning model\",\"authors\":\"Gazi Akgun, Osman Ulkir\",\"doi\":\"10.1177/08927057241243364\",\"DOIUrl\":null,\"url\":null,\"abstract\":\"The final product of additive manufacturing (AM) or 3D printing critically depends on the surface quality. An experimental study on the 3D printed intake manifold flange using acrylonitrile butadiene styrene (ABS) material was executed by varying the four process parameters. A fused deposition modeling (FDM) based 3D printer was used to fabricate the flanges. The association between the parameters and the surface roughness of printed ABS flanges was investigated. A feed forward neural network (FFNN) model trained on particle swarm optimization (PSO) optimized with a genetic algorithm (GA) was used to estimate the surface roughness. A Box-Behnken design (BBD) with printing parameters at three levels was used, and 25 parts were fabricated. The suggested model demonstrated a coefficient of determination (R<jats:sup>2</jats:sup>) of 0.9865 on test values, mean of root-mean-square-error (RMSE) of 0.1231 after 500 times training for generalization. And also mean of overfitting factor is 0.7110. This means that the suggested system could generalize. Comparing the results from the suggested model and ANN, the suggested hybrid model outperformed ANN in predicting the surface roughness values with no overfitting. This suggests that GA optimized PSO based FFNN may be a more suitable method for estimating product quality in terms of surface roughness.\",\"PeriodicalId\":17446,\"journal\":{\"name\":\"Journal of Thermoplastic Composite Materials\",\"volume\":\"95 1\",\"pages\":\"\"},\"PeriodicalIF\":3.6000,\"publicationDate\":\"2024-04-01\",\"publicationTypes\":\"Journal Article\",\"fieldsOfStudy\":null,\"isOpenAccess\":false,\"openAccessPdf\":\"\",\"citationCount\":\"0\",\"resultStr\":null,\"platform\":\"Semanticscholar\",\"paperid\":null,\"PeriodicalName\":\"Journal of Thermoplastic Composite Materials\",\"FirstCategoryId\":\"88\",\"ListUrlMain\":\"https://doi.org/10.1177/08927057241243364\",\"RegionNum\":4,\"RegionCategory\":\"材料科学\",\"ArticlePicture\":[],\"TitleCN\":null,\"AbstractTextCN\":null,\"PMCID\":null,\"EPubDate\":\"\",\"PubModel\":\"\",\"JCR\":\"Q2\",\"JCRName\":\"MATERIALS SCIENCE, COMPOSITES\",\"Score\":null,\"Total\":0}","platform":"Semanticscholar","paperid":null,"PeriodicalName":"Journal of Thermoplastic Composite Materials","FirstCategoryId":"88","ListUrlMain":"https://doi.org/10.1177/08927057241243364","RegionNum":4,"RegionCategory":"材料科学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":null,"EPubDate":"","PubModel":"","JCR":"Q2","JCRName":"MATERIALS SCIENCE, COMPOSITES","Score":null,"Total":0}
引用次数: 0
摘要
增材制造(AM)或三维打印的最终产品关键取决于表面质量。通过改变四个工艺参数,对使用丙烯腈-丁二烯-苯乙烯(ABS)材料的三维打印进气歧管法兰进行了实验研究。使用基于熔融沉积建模(FDM)的三维打印机制造法兰。研究了参数与打印 ABS 法兰表面粗糙度之间的关系。采用遗传算法(GA)优化的粒子群优化(PSO)训练的前馈神经网络(FFNN)模型用于估算表面粗糙度。采用方框-贝肯设计 (BBD),印刷参数分为三个级别,共制造了 25 个零件。所建议的模型在测试值上的判定系数(R2)为 0.9865,经过 500 次训练后的均方根误差(RMSE)平均值为 0.1231。过拟合系数的平均值为 0.7110。这说明建议的系统可以泛化。比较所建议的模型和 ANN 的结果,所建议的混合模型在预测表面粗糙度值方面优于 ANN,且没有过拟合。这表明,基于 GA 优化 PSO 的 FFNN 可能是一种更适合估计产品质量表面粗糙度的方法。
Prediction surface roughness of 3D printed parts using genetic algorithm optimized hybrid learning model
The final product of additive manufacturing (AM) or 3D printing critically depends on the surface quality. An experimental study on the 3D printed intake manifold flange using acrylonitrile butadiene styrene (ABS) material was executed by varying the four process parameters. A fused deposition modeling (FDM) based 3D printer was used to fabricate the flanges. The association between the parameters and the surface roughness of printed ABS flanges was investigated. A feed forward neural network (FFNN) model trained on particle swarm optimization (PSO) optimized with a genetic algorithm (GA) was used to estimate the surface roughness. A Box-Behnken design (BBD) with printing parameters at three levels was used, and 25 parts were fabricated. The suggested model demonstrated a coefficient of determination (R2) of 0.9865 on test values, mean of root-mean-square-error (RMSE) of 0.1231 after 500 times training for generalization. And also mean of overfitting factor is 0.7110. This means that the suggested system could generalize. Comparing the results from the suggested model and ANN, the suggested hybrid model outperformed ANN in predicting the surface roughness values with no overfitting. This suggests that GA optimized PSO based FFNN may be a more suitable method for estimating product quality in terms of surface roughness.
期刊介绍:
The Journal of Thermoplastic Composite Materials is a fully peer-reviewed international journal that publishes original research and review articles on polymers, nanocomposites, and particulate-, discontinuous-, and continuous-fiber-reinforced materials in the areas of processing, materials science, mechanics, durability, design, non destructive evaluation and manufacturing science. This journal is a member of the Committee on Publication Ethics (COPE).