There have been umpteen research reports on the usage of artificial intelligence (AI) strategies for modelling various machining processes. One of the well-known AI strategies is that of fuzzy logic (FL) techniques that has been used for prediction of machining performance variables for both the categories of machining processes and controls the machining process. Given the increasing trend of FL in machining, the chapter reviews the application of fuzzy logic in modelling and controlling the machining processes. The work begins with introduction section and then proceeds to discuss the importance role played by FL strategies in the traditional and modern manufacturing processes. The work summarizes some of the major applications of FL-based systems in various machining processes. Limitations, advantages, and the improvements to minimize the limitations are then discussed. The authors of the chapter hope that the review will aid all those researching in the domain of manufacturing sciences and their optimization techniques.
TopIntroduction
The branch of science that makes the machines intelligent is known as Artificial intelligence (AI). A number of tools have been produced by AI that assists in dealing with complex problems involving human intelligence. The branches comprising of AI are Genetic Algorithm (GA), Expert System (ES), Artificial Neural Network (ANN), Particle Swarm Optimization, Simulated Annealing (SA), Ant Colony Optimization (ACO) and Fuzzy Logic (FL). The results obtained from the different AI tools are far more accurate than the conventional non-AI based approaches and have the potential to predict the tool wear and surface roughness and other machining performance variables with close to 95% confidence level (Torabi, Joo, Xiang, Siong, Lianyin, Jie, ... & Tijo, 2009). In recent times, the use of soft computing tools have increased significantly for modeling and controlling of modern and conventional machining processes (Markos, Szalay, Szöllösi, & Raifu, 1993; Tarng & Hwang, 1995). The FL statements based descriptive language forms the basis for a fuzzy model. The modeling of the machining processes had been done using the conventional modelling tools until the advent of AI techniques such as artificial neural network (ANN), genetic algorithm (GA), fuzzy logic (FL) etc. However, there are few challenges with the AI based techniques such as the speed of learning or being trained and the flexibility of the structure. AI based approaches have also been used for monitoring of the machining systems (Hermann, 1990). Abellan-Nebot and Subiron (2010) have presented a comprehensive review of the different AI techniques for monitoring of the machining systems. Several studies have reported on modeling and optimization of the machining parameters, such as the optimization of the cutting conditions using GA (Zain, Haron, & Sharif, 2011; Singh & Rao, 2007), ACO (Vijayakumar, Prabhaharan, Asokan, & Saravanan, 2003; Wu & Yao, 2008; Kadirgama, Noor, & Alla, 2010), SA (Rao & Pawar, 2010b, Raja & Bhaskar, 2010; Zain, Haron, & Sharif, 2010b; Zain, Haron, & Sharif, 2011 c, d). ANN (Zain, Haron, & Sharif, 2010c) and FL (Dong & Wang, 2011; Ren, Balazinski, Baron, & Jemielniak, 2011; Syn, Mokhtar, Feng, & Manurung, 2011) have been widely used for modeling machining performances. Further, there has been a wide scale application of hybrid systems that are a result of two or more than two branches of AI. For instance, the integrated framework of ANN and PSO has been used for the optimization of cutting conditions for pocket-milling under the constraints of feed and speed (Tandon, El-Mounayri, & Kishawy, 2002) and for optimization of cutting conditions for ball end milling (El-Mounayri, Kishawy, & Briceno, 2005). Hybrid system of FL hybridized with GA was used for abrasive water jet machining (Chakrvarthy & Babu, 2000). ANN and GA was used for the optimization of surface roughness and metal removal rate in case of creep feed grinding machining process (Sedighi & Afsahri, 2010), for achieving optimal cutting condition for electric discharge machining (EDM) as has been done by Wang et al. (2003) andRangajanardhaa and Rao (2009) and for AISI 10140 using TiAlN solid carbide tool (Oktem, 2009).Genetically optimized neural network system (GONNS) was obtained using the ANN and GA systems. The proposed GONNS system was used for the selection of proper cutting condition from experimental data (Aykut, Demetgul, & Tansel, 2010).