N. H. Huu, Duc Tran Cong Toan, T. Huu, H. T. T. Thu
{"title":"填充、填充样式和周壳数对FDM法铸造模的影响","authors":"N. H. Huu, Duc Tran Cong Toan, T. Huu, H. T. T. Thu","doi":"10.1109/GTSD.2018.8595703","DOIUrl":null,"url":null,"abstract":"Rapid prototyping has been the up-and-coming technology to replace traditional manufacturing processes for several decades. However, the cost of producing a rapid prototyping machine that’s able to create a model using metal is too high to apply to low volume production. Therefore, one of the focal points of rapid prototyping is producing patterns for investment casting processes, called rapid casting. Fused deposition modelling (FDM) is a rapid prototyping process which forms a part by continuously melts plastic filaments such as ABS and PLA layer by layer. FDM is a method frequently used in rapid casting. There are two main problems with rapid casting, the cracking of mould shells during the burnout process and patterns incompletely burns out, leaving residual ash and releases corrosive by-products which affect the shells. In this study, we use three parameters to determine their effects on the amount of residual ash the patterns leave behind: Number of perimeters shells, infill percentage and infill pattern. ABS patterns fabricated using FDM method are used because ABS has considerably higher burnout temperature amd is used more frequently in industrial processes, previous studies have also shown that semi-hollow ABS plastic patterns have better dimensional accuracy and surface roughness compared to PLA plastic patterns. After that, we trained an Artificial neural network (ANN) and let it predict the outcomes of the same experiments and compared the results to that of Taguchi’s method predictions using the mean absolute percentage error (MAPE). The results indicate that the predicted outputs of the ANN are very close to the expected output with high reliability (95.218%). ANN method also has a lower MAPE than Taguchi’s method, highlighting that ANN method is a reliable alternative method to accurately predict the relationship between the chosen parameters and the results.","PeriodicalId":344653,"journal":{"name":"2018 4th International Conference on Green Technology and Sustainable Development (GTSD)","volume":"26 1","pages":"0"},"PeriodicalIF":0.0000,"publicationDate":"2018-11-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":"2","resultStr":"{\"title\":\"Effects of Infill, Infill Patterns and Number of Perimeter Shells on Casting Patterns Fabricated Using FDM Method\",\"authors\":\"N. H. Huu, Duc Tran Cong Toan, T. Huu, H. T. T. Thu\",\"doi\":\"10.1109/GTSD.2018.8595703\",\"DOIUrl\":null,\"url\":null,\"abstract\":\"Rapid prototyping has been the up-and-coming technology to replace traditional manufacturing processes for several decades. However, the cost of producing a rapid prototyping machine that’s able to create a model using metal is too high to apply to low volume production. Therefore, one of the focal points of rapid prototyping is producing patterns for investment casting processes, called rapid casting. Fused deposition modelling (FDM) is a rapid prototyping process which forms a part by continuously melts plastic filaments such as ABS and PLA layer by layer. FDM is a method frequently used in rapid casting. There are two main problems with rapid casting, the cracking of mould shells during the burnout process and patterns incompletely burns out, leaving residual ash and releases corrosive by-products which affect the shells. In this study, we use three parameters to determine their effects on the amount of residual ash the patterns leave behind: Number of perimeters shells, infill percentage and infill pattern. ABS patterns fabricated using FDM method are used because ABS has considerably higher burnout temperature amd is used more frequently in industrial processes, previous studies have also shown that semi-hollow ABS plastic patterns have better dimensional accuracy and surface roughness compared to PLA plastic patterns. After that, we trained an Artificial neural network (ANN) and let it predict the outcomes of the same experiments and compared the results to that of Taguchi’s method predictions using the mean absolute percentage error (MAPE). The results indicate that the predicted outputs of the ANN are very close to the expected output with high reliability (95.218%). ANN method also has a lower MAPE than Taguchi’s method, highlighting that ANN method is a reliable alternative method to accurately predict the relationship between the chosen parameters and the results.\",\"PeriodicalId\":344653,\"journal\":{\"name\":\"2018 4th International Conference on Green Technology and Sustainable Development (GTSD)\",\"volume\":\"26 1\",\"pages\":\"0\"},\"PeriodicalIF\":0.0000,\"publicationDate\":\"2018-11-01\",\"publicationTypes\":\"Journal Article\",\"fieldsOfStudy\":null,\"isOpenAccess\":false,\"openAccessPdf\":\"\",\"citationCount\":\"2\",\"resultStr\":null,\"platform\":\"Semanticscholar\",\"paperid\":null,\"PeriodicalName\":\"2018 4th International Conference on Green Technology and Sustainable Development (GTSD)\",\"FirstCategoryId\":\"1085\",\"ListUrlMain\":\"https://doi.org/10.1109/GTSD.2018.8595703\",\"RegionNum\":0,\"RegionCategory\":null,\"ArticlePicture\":[],\"TitleCN\":null,\"AbstractTextCN\":null,\"PMCID\":null,\"EPubDate\":\"\",\"PubModel\":\"\",\"JCR\":\"\",\"JCRName\":\"\",\"Score\":null,\"Total\":0}","platform":"Semanticscholar","paperid":null,"PeriodicalName":"2018 4th International Conference on Green Technology and Sustainable Development (GTSD)","FirstCategoryId":"1085","ListUrlMain":"https://doi.org/10.1109/GTSD.2018.8595703","RegionNum":0,"RegionCategory":null,"ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":null,"EPubDate":"","PubModel":"","JCR":"","JCRName":"","Score":null,"Total":0}
Effects of Infill, Infill Patterns and Number of Perimeter Shells on Casting Patterns Fabricated Using FDM Method
Rapid prototyping has been the up-and-coming technology to replace traditional manufacturing processes for several decades. However, the cost of producing a rapid prototyping machine that’s able to create a model using metal is too high to apply to low volume production. Therefore, one of the focal points of rapid prototyping is producing patterns for investment casting processes, called rapid casting. Fused deposition modelling (FDM) is a rapid prototyping process which forms a part by continuously melts plastic filaments such as ABS and PLA layer by layer. FDM is a method frequently used in rapid casting. There are two main problems with rapid casting, the cracking of mould shells during the burnout process and patterns incompletely burns out, leaving residual ash and releases corrosive by-products which affect the shells. In this study, we use three parameters to determine their effects on the amount of residual ash the patterns leave behind: Number of perimeters shells, infill percentage and infill pattern. ABS patterns fabricated using FDM method are used because ABS has considerably higher burnout temperature amd is used more frequently in industrial processes, previous studies have also shown that semi-hollow ABS plastic patterns have better dimensional accuracy and surface roughness compared to PLA plastic patterns. After that, we trained an Artificial neural network (ANN) and let it predict the outcomes of the same experiments and compared the results to that of Taguchi’s method predictions using the mean absolute percentage error (MAPE). The results indicate that the predicted outputs of the ANN are very close to the expected output with high reliability (95.218%). ANN method also has a lower MAPE than Taguchi’s method, highlighting that ANN method is a reliable alternative method to accurately predict the relationship between the chosen parameters and the results.