Marek Balazinski

Tool condition monitoring using artificial intelligence methods

Abstract This paper describes an application of three artificial intelligence (AI) methods to estimate tool wear in lathe turning. The first two are “conventional” AI methods—the feed forward back propagation neural network and the fuzzy decision support system. The third is a new artificial neural network based-fuzzy inference system with moving consequents in if–then rules. Tool wear estimation is based on the measurement of cutting force components. This paper discusses a comparison of usability of these methods in practice.

An Energy Based Analytical Force Model for Orthogonal Cutting of Metal Matrix Composites

Abstract The machining of metal matrix composite (MMC) presents a significant challenge to the industry. The hard and abrasive nature of the reinforcement causes rapid tool wear and high machining cost. Cracking and debonding of the reinforcement particles are the significant damage modes that directly affect the tool performance. This paper presents, an energy based analytical force model that has been developed for orthogonal cutting processes. The total specific energy for deformation has been estimated along with the energy consumed for debonding as a function of volume fraction and material properties. Orthogonal cutting tests were carried out for a range of different feeds on different matrix materials and volume fractions. The results showed good agreement between the predicted and measured cutting forces.

Machinability of Graphitic Metal Matrix Composites as a Function of Reinforcing Particles

Abstract Aluminum Metal Matrix Composites (MMCs) reinforced with ceramic particles have been developed for high wear resistance applications such as cylinder liners and brakes as a replacement for gray cast iron. Ceramic particles in an aluminum matrix improve its wear resistance property, but also cause high abrasive wear on cutting tools, which results in poor tool life and inconsistent part quality. A new family of MMCs (GrA-Ni®) consisting of an aluminum matrix reinforced with nickel-coated graphite particles and SiC or Al2O3 particles was recently developed. This paper presents the results of machining tests conducted to assess the machinability of the new graphitic ceramic reinforced MMCs. It was found that graphitic aluminum MMC reinforced with alumina is easier to machine than those reinforced with both SiC and graphite or SiC particles only. The incorporation of graphite into these composites and the variation of hard particle content improve their machinability.

Type-2 Takagi-Sugeno-Kang Fuzzy Logic Modeling using Subtractive Clustering

In this paper, a subtractive clustering identification algorithm is introduced to model type-2 Takagi-Sugeno-Kang (TSK) fuzzy logic systems (FLS). The type-2 TSK FLS identification algorithm is an extension of the type-1 TSK FLS modeling algorithm proposed in (S. L. Chiu, 1994), (S. L. Chiu, 1997). In the type-2 algorithm, subtractive clustering method is combined with least squares estimation algorithms to pre-identify a type-1 FLS form input/output data. Then using type-2 TSK FLS theory (J. M. Mendel, 2001), expand the type-1 FLS to a type-2 TSK FLS. Minimum error models are obtained through enumerative search of optimum values for spreading percentage of cluster centers and consequence parameters. By doing so, fuzzy modeling of type-2 TSK FLS is found to be more effective than that of type-1 TSK FLS. Experimental results confirm the effectiveness of this method. A comparison of the Type-1 and -2 TSK FLSs is presented and the limitations of this method are discussed

Analytical Modeling of Tool Wear Progression During Turning Particulate Reinforced Metal Matrix Composites

This paper presents an analytical model for the prediction of tool flank wear progression during bar turning of particulate reinforced metal matrix composites. In this paper, a methodology for analytically predicting the wear progression as function of tool/workpiece properties and cutting parameters is presented. According to this approach, the wear volume due to two body and three body abrasion is formulated. Then, the flank wear rate is formulated by considering the tool geometry in 3D turning. Turning tests were carried out for a range of cutting speeds, tool nose radius and volume fraction of particles. The results showed good agreement between predicted and measured tool wear progression.

Cutting tool reliability analysis for variable feed milling of 17-4PH stainless steel

Variable feed machining has recently been proposed as a significant method to improve cutting tool life particularly for hard and diffucult to machine materials. This method, which is easy to apply in industry, has been shown to improve tool life in the order of 40% in certain cases. This paper presents a reliability model for the quantitative study of the effect of feed variation on tool wear and tool life. To better compare processes with two different wear modes, a reliability model taking simultaneously into account both flank and face wear has been developed. With this model, which is based on experiments, the tool life for the constant and variable feed cases was calculated from the reliability function. The mean time to failure, obtained from the reliability function, provides an accurate evaluation for any probabilistic distribution. The proposed method is therefore a general approach that can be used for analyzing cutting tool life under any conditions and for any equipment and material.

Application of fuzzy logic techniques to the selection of cutting parameters in machining processes

Abstract An idea and an implementation of a fuzzy decision support system (FDSS) based on the compositional rule of inference have been introduced in this paper. Taking into account the fact that metal cutting processes are stochastic, nonlinear and ill-defined, the application of FDSS to the choice and modification of cutting parameters has been described. The obtained results seem to be reasonable and show that the employed methods are appropriate for such kinds of decision problems.

Flank Wear Progression During Machining Metal Matrix Composites

The machining of composites present a significant challenge to the industry. The abrasive reinforcements cause rapid tool wear and increases the machining cost. The results from machining metal matrix composites (MMCs) with conventional tools show that the main mechanism of tool wear includes two-body abrasion and three-body abrasion. A more flexible method that can be considered as a cost-saving technique is therefore sought for studying the machinability characteristics of these materials. In the previous paper, a methodology for predicting the tool flank wear progression during bar turning of MMCs was presented (Kishawy, Kannan, and Balazinski, Ann. CIRP, 54/1, pp. 55-59). In the proposed model, the wear volume due to two-body and three-body abrasion mechanisms was formulated. Then, the flank wear rate was quantified by considering the tool geometry in three-dimensional (3D) turning. Our main objective in this paper is to validate the proposed model by conducting extensive bar turning experiments under a wide range of cutting conditions, tool geometries, and composite material compositions. The cutting test results showed good agreement between predicted and measured tool wear progression.

Tool wear monitoring using genetically-generated fuzzy knowledge bases

Fuzzy logic is an AI method that is being implemented in a growing number of different fields. One of these applications is tool wear monitoring. The construction of a fuzzy knowledge base from a set of experimental data by a human expert however, is a time consuming task, and hence, limits the expansion of the use of this AI method. Alternatively, the fuzzy knowledge base can be automatically constructed by a genetic algorithm from the same set of experimental data without requiring any human expert. This paper compares these two fuzzy knowledge base construction methods and the results obtained in a tool wear monitoring application.

Tolerance allocation based on fuzzy logic and simulated annealing

This paper presents a new technique for dealing with tolerance allocation problems, which can be encountered in the design process of mechanical engineering practice. This technique uses a combination of fuzzy logic to evaluate manufacturing difficulties and a simulated annealing algorithm to optimize tolerance allocation over the complete dimensioning chain.