subject: Group Technology in High Speed Milling Machine [print this page] Group Technology in High Speed Milling Machine
Introduction
High- speed milling is one of fast emerging cutting process in the country now. It has brought tremendous revolution, especially in machining tough, hard and difficult to machine alloys. It has high potential for increased metal removal rates, high surface finish, excellent dimensional accuracy, burr- free edge and producing stress free components. The performance of high speed machining is controlled by several parameters like cutting speed, feed, and depth of cut, cutting temperature, cutting tool materials, geometry of cutting tool, cutting power and work piece condition. This article presents the development of model using neural network to predict the effect of the input parameters on the output process variable. A large number of experiments will be carried out on different materials. Being trained by experimental data initially sneered by design of experiments using Response Surface Methodology, the cutting parameters for the process are optimized using Group Technology.
Description
The use of high speed machining technology becomes more prevalent in recent years as every manufacturer tries to bring down the product life cycle. important applications of high speed milling include the manufacture of dies and molds, numerous steel and AI parts for automobiles, and thin- walled, ribbed, Al components for aerospace. The implementation of high speed machining presents a challenge since various factors limit the extent to which cutting speeds may be advantageously increased. These factors include the cutting tool material, geometry of cutters, the cutting power, the cutting parameters, metallurgical condition of the work piece and the surface finish achieved.
Complete understanding of the technology requires a model to describe the complex and non-linear high speed milling process. Neural networks are highly flexible modeling tools with an ability to learn the mapping between input variables and output. The application of neural networks in modern manufacturing process has been reported in several references. Therefore, neural networks are considered in this paper to model the high speed milling process, optimization of activities in computer-integrated manufacturing (CIM) and process planning is one of the foremost targets of intelligent manufacturing systems (IMS).
The solution to the optimization problems, which include real- valued variables, can be obtained using numerous methods. However, each method has its own pitfalls. Some models can produce accurate solutions by rigorous computation, which is not economic in terms of the computational time and cost. Sometimes, the solutions from these models may not be optimal. Some other models may develop solutions far from the optimum in a fast manner.
Therefore, a compromise between the high accuracy of rigorous solution and low accuracy of an oversimplified solution should be made. This middle course may be achieved using Group Technology (GT), which are easy to implement and powerful to search large solution space. GT's find several applications in process planning, scheduling, design and machining parameters optimization problems.
Conclusion
This article described the modeling and optimization of high- speed milling using neural network and GT approaches. These approaches are used in an attempt to determine the optimal combinations of control parameters of high speed milling. The selection of cutting speed for all different materials is vital, as the use of same range of cutting speeds will show drastic results. Sometimes, it was impossible to start or continue machining at higher cutting speeds for materials like Ti alloys and Stainless steels. To achieve a better surface finish, high cutting speeds with moderate feed rates and high positive rake geometry is suggested. However, the design and development of neural networks is time consuming.
