Samimi, P. (2008). The application of fuzzy logic in CNC programming [Diploma Thesis, Technische Universität Wien]. reposiTUm. http://hdl.handle.net/20.500.12708/179735
The complex and very elaborate mechanism of the human brain that helps him control phenomena and analyse problems, has been the model for the design and structure of most control systems. The two main human advantages in controlling systems are: the access to a flexible data base (which is the data stored in the brain) and the ability to quickly modify and verify it; and the capability of combining different pieces of information obtained through the senses. Hence, the increasing use of intelligent knowledge-base systems is an effort towards externalising the first capability, and the ongoing research on combining information is an effort towards externalising the second. In achieving this, the application and combination of different intelligent systems, like Neural Networks, Fuzzy Systems and Evolutionary Algorithms for controlling systems have been widely considered. The advantage of these systems is the option of benefiting from the learning capabilities. Each of these intelligent systems normally have modifiable parameters that change with learning. In this thesis, the two learning methods of "off line learning" and "on line learning" have been discussed and analysed. The first method that has been developed by the author [2] uses the numerical data off line to generate the data base for a fuzzy controller. In this method, the data base is generated using the numerical data obtained from controlled dynamics. The second method is the self-organised method in which the fuzzy controller can generate the suitable data base for controlling a system while controlling it. These methods are then used to design the controller that move the axes/controls the primary movers of a Computerised Numerically Controlled (CNC) machine. In addition to implementing the above mentioned algorithms, two further applications for CNC machines are presented, namely: (1) Controlling the metal removal rate; (2) Combination of different information from various sensors. There has been a lot of research on combining information, and many theories have been presented [18] to [21]. Some of these theories are later discussed, and in the end the author presents a method in using the fuzzy logic in combining sensor information.
