Petra Kersting

Optimizing NC-tool paths for simultaneous five-axis milling based on multi-population multi-objective evolutionary algorithms

Computer-Numerical-Control based five-axis milling offers new possibilities for improving the machining process. However, this procedure is still difficult to handle, particularly in case of machining complex free-formed surfaces. An optimization approach based on the multi-objective evolutionary algorithm SMS-EMOA (S-metric selection evolutionary multi-objective optimization algorithm) combined with a multi-population approach has been developed and used in order to utilize the potential of the five-axis milling process. After a general introduction to this machining process and the potential of path optimization, the designed multi-population multi-objective evolutionary approach, its integration into the simulation, and its adaptation to the practical example is described.

Modeling regenerative workpiece vibrations in five-axis milling

During milling—especially of thin-walled components—the dynamic behavior of the workpiece-tool-machine-system influences the milling process and particularly the quality of the resulting workpiece surface. This article focuses on the presentation of a simulation concept for predicting regenerative workpiece vibrations, which combines a finite element model for analyzing the dynamic behavior of the workpiece with a time domain simulation for the five-axis milling process. Both concepts, their linking, and the experimental setup for verifying the simulation will be described. A comparison of the simulation results with the data measured in experiments with regard to the vibration frequencies as well as the surface quality will be given.

SIMULATION CONCEPT FOR PREDICTING WORKPIECE VIBRATIONS IN FIVE-AXIS MILLING

The characteristic discontinuous cut of the milling process influences the whole machining process by an increased susceptibility to vibrations of the machine-tool-workpiece system. This can result in undesirable effects on the workpiece surface or in a shorter lifetime of the tool and the spindle. Especially with regard to the machining of thin-walled components, such as turbine blades and thin profiles, the dynamic behavior of the workpiece is of particular interest. In this paper a simulation concept for predicting regenerative workpiece vibrations during the five-axis milling process is presented. This concept combines an accurate and fast simulation of the five-axis machining process including material removal and force calculation with an implemented finite element model for computing workpiece displacements. The simulation results are compared with data from experiments, which were conducted using a milling tool with high stiffness in order to minimize the influence of the milling tool dynamics.

High-feed milling of tailored surfaces for sheet-bulk metal forming tools

AbstractThe increasingly investigated and applied production process sheet-bulk metal forming (SBMF) has novel requirements for the forming tools, e.g., the need of an adaptive material flow at different areas of the tool for an adequate form filling. One new method to realize different, defined tribological conditions are tailored surfaces (TS). During the design of forming tools, it is imperative to have profound knowledge about the tribology between the tool and the workpiece. This article introduces structuring with high-feed milling tools as one possibility for influencing the material flow during forming processes and presents a ring-compression test for the quantification of the tribological conditions, which is adapted for SBMF. On the basis of various machined structures, surface parameters are analyzed to identify a correlation with the friction coefficient to gain knowledge about the mechanisms of TS and to be able to choose structures according to the needs of SBMF processes.

Modelling and simulation of Internal Traverse Grinding: bridging meso- and macro-scale simulations

Abstract In this work, we focus on the computational bridging between the meso- and macro-scale in the context of the hybrid modelling of Internal Traverse Grinding with electro-plated cBN wheels. This grinding process satisfies the manufacturing industry demands for a high rate of material removal along with a high surface quality while minimising the number of manufacturing processes invoked. To overcome the major problem of the present machining process, namely a highly concentrated thermal load which can result in micro-structural damage and dimension errors of the workpiece, a hybrid simulation framework is currently under development. The latter consists of three components. First, a kinematic simulation that models the grinding wheel surface based on experimentally determined measurements is used to calculate the transient penetration history of every grain intersecting with the workpiece. Secondly, an h-adaptive, plane-strain finite element model incorporating elasto-plastic work hardening, thermal softening and ductile damage is used to simulate the proximity of one cBN grain during grinding and to capture the complex thermo-mechanical material response on a meso-scale. For the third component of the framework, the results from the preceding two simulation steps are combined into a macro-scale process model that shall in the future be used to improve manufacturing accuracy and to develop error compensation strategies accordingly. To achieve this objective, a regression analysis scheme is incorporated to approximate the influence of the several cutting mechanisms on the meso-scale and to transfer the homogenisation-based thermo-mechanical results to the macro-scale.

Innovative Tools to Improve Incremental Bulk Forming Processes

Sheet-bulk metal forming is an innovative process with a high potential to generate load-adapted parts with high precision. Bulk forming processes of sheet metals especially require high process forces, resulting in an intense contact pressure and, thus, in a very high abrasive and adhesive wear. As a method to reduce or avoid these common wear phenomena, even hardened or coated tool surfaces are not sufficient. The objective of this paper is to show an improvement of the tool resistance during an incremental forming process by an adapted tool design and the application of structured tool surfaces combined with coatings. For the tool surface the structure of the scarabaeus beetle serves as the basis for a bionic structure. This structure was manufactured by micromilling. Despite the high hardness of the tool material and the complex geometry of the forming tools, very precise patterns were machined successfully using ball-end milling cutters. The combination of bionic structures with coating techniques like physical vapor deposition (PVD) on plasma nitrided tool surfaces is very promising. In this work, the influence of process parameters (workpiece material, lubrication, tool design, stepwise infeed) on the tool resistance during the forming operation was analyzed experimentally. The results of the optimized forming tools were compared to conventional, unstructured, uncoated, and only plasma nitrided forming tools. The different tools were applied to 2 mm thick metal sheets made of aluminum (AlMg3) and steel (non-alloy quality steel DC04). As a result, the process forces could be reduced by a modified shape and surface of the tools. Thus, the lifetime of the tools can be enhanced.

On the use of problem-specific candidate generators for the hybrid optimization of multi-objective production engineering problems

In the field of production engineering, various complex multi-objective problems are known. In this paper we focus on the design of mold temperature control systems, the reconstruction of digitized surfaces, and the optimization of NC paths for the five-axis milling process. For all these applications, efficient problem-specific algorithms exist that only consider a subset of the desirable objectives. In contrast, modern multi-objective evolutionary algorithms are able to cope with many conflicting objectives, but they require a long runtime due to their general applicability. Therefore, we propose hybrid algorithms for the three applications mentioned. In each case, the problem-specific algorithms are used to determine promising initial solutions for the multi-objective evolutionary approach, whose variation concepts are used to generate diversity in the objective space. We show that the combination of these techniques provides great benefits. Since the final solution is chosen by a decision maker based on this Pareto front approximation, appropriate visualizations of the high-dimensional solutions are presented.

Optimizing of NC tool paths for five-axis milling using evolutionary algorithms on wavelets

Computer aided NC-path generation of five-axis milling using a standard CAM-system does usually not take machine dynamics and kinematics into account. This results in machine movements which are often not smooth enough and lead to a deficient surface quality. In order to reduce undesirable abrupt motion changes, an approach for optimizing the NC-path by using a standard evolution strategy is shown in this paper as well as first results of applying this algorithm to the five-axis milling process.

Heat input modeling and calibration in dry NC-milling processes

Due to friction and material deformation in the shear zone, workpieces in NC-milling processes are subjected to heat input and thermal loading. Ongoing geometric changes as well as time-varying contact and cutting conditions result in an inhomogeneous temperature field that is constantly in flux. Such thermally loaded workpieces often exhibit complex and transient thermomechanical deformations, which may result in erroneous material removal with respect to the desired shape. In order to meet critical manufacturing tolerances, it is therefore necessary to avoid and compensate these effects. Predicting the deformation exhibited by a thermally loaded workpiece is a problem of linear thermoelasticity, which can be solved by use of the finite element (FE) method. A prerequisite to this is the accurate calculation of the temperature field that results within the workpiece material during the course of the milling process. Although the FE method may be used for this as well, the practical application to realistic milling processes is limited due to the required computational resources. This paper presents a fast geometric process simulation for the prediction of cutting forces, heat input and thermal loading in dry NC milling. The temperature field of the workpiece is continuously updated, such that it is possible to determine the temperature of any material point at any point in time of the milling process. Individual models comprising the simulation system are described in detail, along with the experiments that are required to calibrate them. The accuracy of the geometric process simulation is validated by comparison with experimental data for a non-trivial milling process.

Predicting thermal loading in NC milling processes

In dry NC milling, a significant amount of heat is introduced into the workpiece due to friction and material deformation in the shear zone. Time-varying contact conditions, relative tool–workpiece movement and continuous geometric change of the workpiece due to material removal lead to a perpetually changing inhomogeneous temperature distribution within the workpiece. This in turn subjects the workpiece to ongoing complex thermomechanical deformations. Machining such a thermally loaded and deformed workpiece to exact specifications may result in unacceptable shape deviations and thermal errors, which become evident only after dissipation of the introduced heat. This paper presents a hybrid simulation system consisting of a geometric multiscale milling simulation and a finite element method kernel for solving problems of linear thermoelasticity. By combination and back-coupling, the described system is capable of accurately modeling heat input, thermal dispersion, transient thermomechanical deformation and resulting thermal errors as they occur in NC milling processes. A prerequisite to accurately predicting thermomechanical errors is the correct simulation of the temperature field within the workpiece during the milling process. Therefore, this paper is subjected to the precise prediction of the transient temperature distribution inside the workpiece.