Process Modeling Optimization in Additive Manufacturing Using Artificial Neural Networks
Jafarian, Hesam (2018)
2018
Automation Engineering
Teknisten tieteiden tiedekunta - Faculty of Engineering Sciences
This publication is copyrighted. You may download, display and print it for Your own personal use. Commercial use is prohibited.
Hyväksymispäivämäärä
2018-12-05
Julkaisun pysyvä osoite on
https://urn.fi/URN:NBN:fi:tty-201811222756
https://urn.fi/URN:NBN:fi:tty-201811222756
Tiivistelmä
The need for production has roots in human life and its history. This date back to primitive days of human life, where he or she had to apply surrounding materials in order to manufacture the tools necessary for survival and durability against any insecurity. This was legitimizing the use of any means in order to obtain the tools and reach the goals at any cost. However, with human development primarily within the knowledge and understanding domain and also with the desire of humanity for best, expectations have risen. This was the time not only the cost mattered but also the simplicity of design, massive production, and diversity, less waste, autonomy, and implementation within a shorter time gained a higher momentum.
On the other hand, the conventional manufacturing method was based on subtractive manufacturing with cutting and eliminating the unwanted sections or parts of an object. The disadvantage of such a method is that it requires a complicated production process design and is accompanied by waste. However, with the rise of additive manufacturing and three-dimensional printing equipment back in the 1980s, it became possible to build parts which could have almost any shape or geometry. Moreover, this also empowered the possibility of using digital and 3D models built by computer-aided design software.
Simultaneously, on the other side, the foundation and application of artificial intelligence were maturing. This was due to the demand for machines to assist human beings in the domain of knowledge reasoning, learning, and planning. These were the pillars for making machines autonomous and to benefit from such features.
Accordingly, this research work studies and overviews the applications and techniques of machine learning and artificial intelligence in the domain of additive manufacturing. It aims to determine the interaction of influential parameters on the process and to find the best solutions for improving the quality and mechanical features of manufactured parts.
Moreover, this research tends to enable the experts to grasp a better understanding of AM process during manufacturing and additionally intends to infuse the experts' knowledge in additive manufacturing field utilizing the artificial neural network and finally generate a model with the ability of prediction and selection for promising results.
On the other hand, the conventional manufacturing method was based on subtractive manufacturing with cutting and eliminating the unwanted sections or parts of an object. The disadvantage of such a method is that it requires a complicated production process design and is accompanied by waste. However, with the rise of additive manufacturing and three-dimensional printing equipment back in the 1980s, it became possible to build parts which could have almost any shape or geometry. Moreover, this also empowered the possibility of using digital and 3D models built by computer-aided design software.
Simultaneously, on the other side, the foundation and application of artificial intelligence were maturing. This was due to the demand for machines to assist human beings in the domain of knowledge reasoning, learning, and planning. These were the pillars for making machines autonomous and to benefit from such features.
Accordingly, this research work studies and overviews the applications and techniques of machine learning and artificial intelligence in the domain of additive manufacturing. It aims to determine the interaction of influential parameters on the process and to find the best solutions for improving the quality and mechanical features of manufactured parts.
Moreover, this research tends to enable the experts to grasp a better understanding of AM process during manufacturing and additionally intends to infuse the experts' knowledge in additive manufacturing field utilizing the artificial neural network and finally generate a model with the ability of prediction and selection for promising results.