TY - JOUR
T1 - Computational tool for the intelligent design of gearboxes of cylindrical gears and welded housing
AU - Franco, Rosendo
AU - Blas, Michael A.
AU - Inafuku, Luis H.
AU - Peinado, Angel A.C.
AU - Soto, Jean C.
AU - Solano, Alberto E.
AU - Fernández, Daniel H.
AU - López, Alexander R.
AU - Montalván, José F.
AU - Yépez, Herbert
AU - Valverde, Quino
N1 - Publisher Copyright:
© Springer Nature Switzerland AG 2019.
PY - 2019
Y1 - 2019
N2 - The present work reports the development of intelligent software that permits to design gearboxes of cylindrical gears and welded housing with a high degree of automation. At the time there is a large amount of CAD software with options to design machine elements, but there is not any that focuses on the complete design of gearboxes. This is because a gearbox is a product of high complexity, in which many variables are involved and, at the same time, there are innumerable solutions for the same need. To overcome these difficulties, the authors have used the Artificial Intelligence technique known as Case Based Reasoning (CBR). The developed software has the characteristic modules of a case-based system: a base of resolved cases, a retrieve module and an adaptation module. The main advantage of this type of systems is that they manage to acquire the experience (knowledge) of the specialists and store it through the resolved cases. This experience is reused in the solution of new problems. During the research process it was necessary to define the main characteristics that describe a gearbox, i.e., its similarity attributes, as well as to establish the similarity function and implement a parametric adaptation technique. Additionally, the software has modules for calculating and selecting the components of a gearbox, to tune the design to meet the requirements established by technical standards or by a user. The implementation was done in Visual Basic language in the form of an add-in, which is automatically loaded into the Autodesk Inventor® environment. From the input data, the developed program, called aiGearboxDesigner, is capable of generating a complete design, 3D models and the corresponding assembly planes and manufacturing drawings.
AB - The present work reports the development of intelligent software that permits to design gearboxes of cylindrical gears and welded housing with a high degree of automation. At the time there is a large amount of CAD software with options to design machine elements, but there is not any that focuses on the complete design of gearboxes. This is because a gearbox is a product of high complexity, in which many variables are involved and, at the same time, there are innumerable solutions for the same need. To overcome these difficulties, the authors have used the Artificial Intelligence technique known as Case Based Reasoning (CBR). The developed software has the characteristic modules of a case-based system: a base of resolved cases, a retrieve module and an adaptation module. The main advantage of this type of systems is that they manage to acquire the experience (knowledge) of the specialists and store it through the resolved cases. This experience is reused in the solution of new problems. During the research process it was necessary to define the main characteristics that describe a gearbox, i.e., its similarity attributes, as well as to establish the similarity function and implement a parametric adaptation technique. Additionally, the software has modules for calculating and selecting the components of a gearbox, to tune the design to meet the requirements established by technical standards or by a user. The implementation was done in Visual Basic language in the form of an add-in, which is automatically loaded into the Autodesk Inventor® environment. From the input data, the developed program, called aiGearboxDesigner, is capable of generating a complete design, 3D models and the corresponding assembly planes and manufacturing drawings.
KW - Case based reasoning
KW - Gearbox
KW - Mechanical design
UR - http://www.scopus.com/inward/record.url?scp=85065965786&partnerID=8YFLogxK
U2 - 10.1007/978-3-030-16423-2_22
DO - 10.1007/978-3-030-16423-2_22
M3 - Article
AN - SCOPUS:85065965786
SN - 2211-0984
VL - 71
SP - 237
EP - 247
JO - Mechanisms and Machine Science
JF - Mechanisms and Machine Science
ER -