Thomas Michelitsch

Method for optimizing the cooling design of hot stamping tools

During the past years hot-stamped components have gained considerable importance in the automotive industry. This is due to the advantageous properties involved, such as good crash behavior and high strength. In the production of hot-stamped components the blank is rapidly cooled by the tool. This exerts an important influence on the final properties and the process time. The tool is actively cooled by cooling ducts through which a medium flows. This article will present a newly developed method in which the design of the cooling ducts can be systematically optimized. A test tool is used to show the results of this method. It is easy to realize an optimized tool for the hot stamping process.

On the design of optimisers for surface reconstruction

In many industrial applications the need for an efficient and high-quality reconstruction of free-form surfaces does exist. Surface Reconstruction - the generation of CAD models from physical objects - has become an independent area of research. The supplementary modification and the automated manufacturing of workpieces represent typical fields of application. Small tolerances in the desired properties result in a very high number of scan points needed. Thus, modern approaches have to be capable of processing, analysing and modelling these amounts of data.There are several studies that use evolutionary algorithms (EA) for surface reconstruction tasks. Until now, these studies only describe the general ability of EA to successfully optimise surfaces. Aspects like runtime as well as comparability to other optimisation techniques have not been considered. Since these aspects are of great importance for integration in applicable software tools, in this paper the ability of a state-of-the-art multi-objective EA to be successfully integrated in surface reconstruction software is analysed. Major drawbacks are disclosed and necessary add-on modules are presented.

pMOHypEA: Parallel Evolutionary Multiobjective Optimization using Hypergraphs

University of Dortmund, Chair of Systemanalysis, Germanykarlheinz.schmitt@cs.uni-dortmund.de, WWW home page:http://ls11-www.cs.uni-dortmund.deAbstract. An overview of the state-of-the-art in parallel evolutionarymultiobjective optimization with a special view on the population struc-tures is presented. An introduction to the theory of hypergraphs is given.The idea of hypergraphs is to scale freely between all existing populationstructures from coarse-grained island-models to flne-grained difiusion-models. Implementation details of pMOHypEA, a parallel evolutionarymultiobjective optimization system that uses hypergraphs to structurepopulations, are given. First results of the concept are discussed.

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.

Intelligent process planning methods for the manufacturing of moulds

The layout of temperature control systems for moulds is decisive for the performance and stability of the production process. A design and optimisation approach for temperature control systems is introduced, coping with geometric constraints and complex thermal dependencies and allowing a significant reduction of manufacturing costs. Five-axis milling processes are increasingly used for the production of moulds in order to achieve high surface qualities and low manufacturing times. The CAM-programming required for the milling of free-formed surfaces in this field is complex and error-prone. An approach is shown, which automatically generates five-axis NC-paths from existing error-free three-axis paths.

Fast distance computation between cylinders for the design of mold temperature control systems

Optimization algorithms for the design of mold temperature control systems based on deep-hole bores have to assure that the minimal distances between the bores meet given safety margins. If the bores are geometrically modeled as cylinders, this leads to the necessity of determining the minimal Euclidean distances between cylinders and testing them against the corresponding margins. In this paper a very fast and reliable algorithm for the distance computation between cylinders is introduced, which has been developed due to the run-time requirements of the problem at hand.

Sequential Parameter Optimization of an Evolution Strategy for the design of Mold Temperature Control Systems

Sequential Parameter Optimization (SPO) is a popular model-assisted approach for tuning the parameters of metaheuristics, which is based on models from the Design and Analysis of Computer Experiments (DACE). Despite the indisputable success of SPO, some of the assumptions behind DACE, such as deterministic output and stationary covariance, do not hold for parameter optimization. Thus, an analysis of enhanced covariance kernels for the consideration of noise is performed. Furthermore, the effects of different sequential sampling strategies and an increasing number of replicates of each design on the quality of the models are discussed. To accomplish this, an Evolution Strategy (ES) is tuned on the real-world optimization problem of designing Mold Temperature Control Systems. Based on the results, recommendations for the ES parameters are provided, insights about the robustness of DACE with respect to the violations are made, and recommendations for appropriate combinations of sampling strategies and covariance kernels are derived.

Designing memetic algorithms for real-world applications using self-imposed constraints

Memetic algorithms (MAs) combine the global exploration abilities of evolutionary algorithms with a local search to further improve the solutions. While a neighborhood can be easily defined for discrete individual representations, local search within real-valued domains requires an appropriate choice of the local search method. If the subject of optimization shows discontinuous behavior, a standard hill-climbing routine cannot be successfully applied. Thus, in this paper we present a general approach that defines a quasi-discrete neighborhood for real-valued variables by applying problem-specific self-imposed constraints. Thereby, knowledge about properties of good solutions can be easily integrated into the search process and discontinuous parts can be found. Satisfying results can be obtained faster while all important issues in the design of MAs are preserved.

Collaborative Research Centre 531: Computational Intelligence – Theory and Practice (Sonderforschungsbereich 531: Computational Intelligence – Theorie und Praxis)

Computational Intelligence (CI) is an umbrella term for modern problem solvers such as Evolutionary Algorithms, Neural Networks and Fuzzy Logic. These methods have received increasing attention due to their simplicity, robustness and generality. The Collaborative Research Center “Computational Intelligence“ (SFB 531) is concerned with the theoretical foundations and applications of CI methods. This article focuses on two examples from different research domains within SFB 531. First, exemplary results for the runtime analysis of evolutionary algorithms are summarized and evaluated. Second, applied research on mold temperature control is dealt with. Here it is stressed how the wide variety of CI methods leads to very efficient solutions to the problem and how still substantial improvements can be obtained by hybridization with experts knowledge. Der Begriff Computational Intelligence (CI) umfasst moderne Verfahren wie evolutionäre Algorithmen, neuronale Netze und Fuzzy-Logik, die zur Problemlösung in technischen Systemen angewandt werden. Diese Methoden werden zunehmend populärer, da sie bestechend einfach, robust und allgemein gehalten sind. Der Sonderforschungsbereich „Computational Intelligence“ (SFB 531) befasst sich mit den theoretischen Grundlagen und Anwendungen von CI-Methoden. Im Artikel werden zwei Beispiele aus thematisch unterschiedlichen Projektbereichen herausgehoben. Zunächst werden beispielhafte Ergebnisse bei der Laufzeitanalyse evolutionärer Algorithmen zusammengefasst und bewertet. Anschließend werden Forschungsergebnisse zu Temperierbohrungen im Maschinenbau behandelt. Dabei wird herausgestellt, wie die große Vielfalt an CI-Methoden zu sehr effizienten Lösungen des Anwendungsproblems führt und wie nichtsdestoweniger wesentliche Verbesserungen durch eine Hybridisierung der Ansätze mit Expertenwissen gefunden werden können.

Evolutionary optimized mold temperature control strategies using a multi-polyline approach

During the machining process the tools for pressure and injection molding have to keep an optimal working temperature. This temperature depends on the workpiece material and allows a safe, efficient and precise machining process. The compact and very expensive steel molds are penetrated with deep hole drilling bores that are combined to form mold temperature control circuits. Today the structure of these circuits are designed manually. Here, a new automatic layout system for mold temperature control strategies is introduced which uses a multiobjective fitness function. The circuits are encoded via a polyline approach. The complex optimization problem is solved using a variation of the evolution strategy. The evolutionary approach as well as first results of the system will be discussed.