Peter Köhler

Concept Development of Design Driven Parts Regarding Multidisciplinary Design Optimization

ABSTRACTThe increasing number of derivates in the automotive sector, combined with the product variation for individualization and technical progress, is presenting a major challenge for product designers. There are many case studies, especially in the concept development, which need to be checked on feasibility. This paper outlines how the necessary efficiency improvement, to deal with the rising amount of projects, can be achieved through the application of a Knowledge Based Engineering System (KBES). A special focus is put on the automated construction of design driven components. Furthermore, possibilities for the integration of expert knowledge into the construction system are demonstrated in order to enable a single user to build a product model according to the methodology of Multidisciplinary Design and Optimization (MDO).An approach for the knowledge based creation of automotive headlamps is presented in detail. The advantages of a reacting construction system consisting of a variety of construct...

An Approach for Bone Pose Estimation via Three External Ellipsoid Pressure Points

The accurate reproduction of bone motion during normal gait using noninvasive external sensors is still an open issue: Using skin markers may lead to large artifacts due to skin sliding, while using newer technologies such as fluoroscopy allows only for short exposure in small regions due to radiation limits, and bone pins used in the past are today prohibited due to the risk of inflammations and pain. This paper presents a simple method for noninvasive bone motion estimation based on palpating prominent bone landmarks via tracked pressure foil planes, where three such landmarks suffice for bone pose estimation. Its mathematical formulation corresponds to determining the pose of a rigid body carrying three ellipsoids when the “pressure points”, i.e. the perpendicular feet of the extremal distance points of the ellipsoids on the three pressure foil planes are given. In a previous paper, we showed that the planar case is akin to the 3PPR manipulator, but yielding instead of two solutions up to 64 complex and (up to now found) 48 real solutions. In this paper we treat the 3D case, which is solved numerically, and validate the concept by experimental measurements. It is shown that the method is numerically stable, yielding an accuracy of \(0.8{^{{\circ }}}\) for flexion/extension and \(1.2{^{{\circ }}}\) for abduction/adduction motion of the lower leg.

Preprocess-Optimization for Polypropylene Laser Sintered Parts

Additive manufacturing delivers the opportunity to manufacture complex geometry with comparable effective effort. Nevertheless, comprehensive information for the sufficient configuration of the process and related parameters are still missing. Here joint researches of the chairs for Manufacturing Technology and Computer Aided Design at the University of Duisburg-Essen were carried out in order to receive detailed information about the influencing factors on part quality for polypropylene laser sintering parts. These experimental results provided the basis for the development of software supported applications for the preprocess optimization.

Improving part quality and process efficiency in layered manufacturing by adaptive slicing

The characteristic of layered manufacturing is the incremental build-up which inevitably leads to the staircase effect. One method to reduce this effect is to optimise part orientation and to apply adaptive slicing procedures. Although layer thickness will be decreased in some regions, process efficiency can be kept to a reasonable level. This paper presents different approaches of adaptive slicing for laser-based applications in consideration of limited range of thickness variation to ensure the mechanical part properties. In contradiction to previous work, the number of applied layer thicknesses is limited due to the associated process parameters. In order to optimise the slicing process and to obtain the best results, this paper presents a practicable approach based on existing work to simulate adaptive slicing. To underline the potential of non-uniform slicing, illustrative examples will show possible improvements in the case of replacing the uniform slicing procedure.

KNOWLEDGE-BASED SUPPORT DURING DESIGN OPTIMIZATION USING FLANGES AS AN EXAMPLE

In today’s interdisciplinary environment, design engineers are facing increasing market demands for faster product cycle times. This implies that product-lifecycle processes need to be performed parallelly and faster. Repetitive processes should be questioned. To support the designer, processes can be automated by using clever linking software, Knowledge-Based Engineering (KBE), and a central knowledge base. This paper demonstrates an approach by giving examples of bolted and flange connections. The result supports the user, by providing simulation templates and optimization algorithms.

Knowledge-based approach for an automatic rheological optimization of injection-molded parts

The motivation for the presented approach results from the absence of an integrated method of supporting the designing of injection-molded parts. Thereby, the focus lies on the rheological optimization, which is ensured by knowledge transfer from the production engineer to the designer, i.e., the mentioned aim is achieved by creating a knowledge base including the expert knowledge. This method is based on an integrated product simulation containing the interconnection of CAD and CAE-systems. The first step in the process is a filling simulation done by a standard injection molding process simulation software. The gate location is considered as fixed and optimal. Nevertheless, it is still possible that a uniform mold filling is not reached. To determine an uneven mold filling the results of the simulation are exported as scalar values to a post processing software tool. Thus, a knowledge based analysis of the results using user defined parameters based on flow length, pressure distribution, etc., is enabled. To enhance the filling, the original part design is modified by adding flow leaders or deflectors. Their positioning is calculated by the post processing software tool and this result is returned to the CAD-system. There the preliminary design is done by an analytical approach. By iterations of the mentioned method, the geometry of the flow leaders or deflectors and thus the mold filling are improved. The presented approach is illustrated exemplarily by an injection-molded part. As a result, the designer is supported methodically designing injection-molded parts by implementing and using expert knowledge. So, the number of iterations and the development time (costs) are reduced.The motivation for the presented approach results from the absence of an integrated method of supporting the designing of injection-molded parts. Thereby, the focus lies on the rheological optimization, which is ensured by knowledge transfer from the production engineer to the designer, i.e., the mentioned aim is achieved by creating a knowledge base including the expert knowledge. This method is based on an integrated product simulation containing the interconnection of CAD and CAE-systems. The first step in the process is a filling simulation done by a standard injection molding process simulation software. The gate location is considered as fixed and optimal. Nevertheless, it is still possible that a uniform mold filling is not reached. To determine an uneven mold filling the results of the simulation are exported as scalar values to a post processing software tool. Thus, a knowledge based analysis of the results using user defined parameters based on flow length, pressure distribution, etc., is enable...

15. Gemeinsames Kolloquium Konstruktionstechnik 2017 : Interdisziplinäre Produktentwicklung

Begriffe wie Industrie 4.0, Digitalisierung, Smart Engineering, Digitaler Zwilling stehen fur aktuelle Trends, in die sich zahlreiche Forschungs- und Entwicklungsaktivitaten einordnen lassen. Sie bedeuten allerdings nicht, dass die Themenfelder, die mit Begriffen wie Knowledge Based Engineering, Product Lifecycle Management, Virtuelles Prototyping usw. verbunden sind, schon ausreichend bearbeitet wurden. Aber „Schlagworter“ sollen ja stets auch helfen, notwendigen Entwicklungen den richtigen Schwung zu verleihen. Das setzt naturlich einen gewissen Entwicklungsstand in verschiedenen Bereichen voraus. Das wird auch in zahlreichen Beitragen dieses Tagungsbandes deutlich. Es ist seit vielen Jahren gute Tradition, dass vor allem junge Wissenschaftlerinnen und Wissenschaftler aus Konstruktionslehrstuhlen verschiedener Hochschulen und Universitaten das Kolloquium nutzen, um ihre Forschungs- und Entwicklungsarbeit zur Diskussion zu stellen. Das erste Gemeinsame Kolloquium Konstruktionstechnik fand 2003 in Rostock statt. Es wurde von den Konstruktionslehrstuhlen der Universitaten Rostock, Dresden und Magdeburg initiiert. Seit 2007 gehoren auch die Konstruktionslehrstuhle der Universitaten Aachen und Bayreuth zu den Mitveranstaltern und seit drei Jahren auch die der Universitaten Duisburg-Essen und Clausthal. Sehr erfreulich ist, dass in diesem Jahr dieses Angebot zum Gedankenaustausch uber aktuelle Problemstellungen auch wieder von anderen Universitaten und Hochschulen sowie von Vertretern aus der industriellen Praxis genutzt wird. Dabei stehen insbesondere Themen im Vordergrund, die sich mit dem qualifizierten Einsatz moderner Informationstechniken in interdisziplinaren Produktentwicklungsprozessen beschaftigen. Ging es vor rund 40 Jahren bei der beginnenden Digitalisierung im Bereich der Konstruktionstechnik noch darum, digitale geometrische Reprasentationen der Konstruktionsobjekte zu erzeugen oder Berechnungssysteme aufzubauen, so wird heute mit dem Begriff der Digitalisierung in diesem Bereich der Wandel hin zu digitalen System- und Prozessmodellen verbunden, die miteinander kommunizieren konnen bzw. informationstechnisch verknupft und abgesichert sind. Drauf aufbauend fokussieren einige Aktivitaten von „Industrie 4.0“ auf die Einbindung der realen Welt. Das stellt auch die Konstruktionswissenschaftler immer wieder vor neue Herausforderungen.

Cost Estimation in Engineer-to-Order Manufacturing

Abstract In Engineer-to-Order (ETO) manufacturing the price of products must be defined during early stages of product design and during the bidding process, thus an overestimation of product development (PD) costs may lead to the loss of orders and an underestimation causes a profit loss. What many ETO systems have in common is that the products have to be developed based on different customer requirements so that each order usually results in a new variant. Furthermore, many customer requirement change-requests may arise in different phases of the PD, which is to be considered properly. Thus it is utmost important for ETO systems to have an accurate cost estimation in first stages of the product design and to be able to determine the cost of customer requirement changes in different phases of PD. This paper aims to present a cost estimation methodology as well as a cost estimation model, which estimate the cost of products by relative comparison of the attributes of new product variants with the attributes of standard product variants. In addition, as a necessity in ETO manufacturing, the cost calculation of customer requirement changes in different phases of PD is integrated in the presented method.

Implicit learning by means of a Knowledge-Based Engineering System

Three-dimensional CAD -models are representing the foundation of modern product development in different kinds of branches. Even medium-sized and small-scale enterprises use powerful CAD-software, as they offer a huge possibility to increase the efficiency in the development process. Hence, learning the handling of such programs is part of many different engineering studies. The depth of education is varying from the learning of the basic functions of solid- and surface- modeling up to advanced techniques of automated product design. However, investigations show that the potential of the provided software is only used rudimentarily. Furthermore, the continually progressing development of those software-systems demands a constant learning process of the users, to ensure that they are capable to exhaust the full potential. Due to the fact, that this causes high monetary and temporary effort, it is not practiced in most companies. Goal of this paper is to state an approach, how the application of modern virtual tools can be used for efficiency enhancement, without demanding specific know-how (concerning the construction methods as well as the particular component) of the user. Therefore, a Knowledge Based Engineering System is introduced, which includes the necessary product- and process-knowledge, which can be accessed by the user according to the respective situation. An experimental study is performed and evaluated to proof the effects of the introduced system.

Design Approach for Spiral Milling Parts Using Knowledge Based Engineering

From the engineering point of view there are still problems regarding geometry modelling that cannot be satisfactorily solved despite efficient CAD systems. This applies in particular to virtual models when they are meant to match with the resulting manufacturing geometry of components as closely as possible. In this paper, relationships between modelling and manufacturing strategies are discussed. The focus is directed to the relationship between the modelling strategy and the manufacturing process by spiral milling with a ball cutter, whereby especially the influence of the tool geometry and the tool paths for the modelling strategy will be investigated. By means of an appropriate example of a design of spiral mandrel die from the field of plastics processing an approach will show how functional design and manufacturing-oriented design can be coupled and automated.