Philippe Dépincé

User-centered design by genetic algorithms: Application to brass musical instrument optimization

This work presents an implementation of genetic algorithms (GAs) for a user-centered design of products. It describes at first a methodology for a user-centered design, based on the coupling between a subjective study to define desirable features and on objective study to find out the influencing objective variables. It relies on two domains that remain generally distinct: the design with a scientific approach (generally math-based) and the design with a sensory and perceptual approach (subjective). The methodology is presented on a particular product for which the perceived aspects are essential: a musical instrument (trumpet). Two types of study were carried out on a set of trumpets: firstly, a sensory study, which aim is to characterize the perception of the intonation of the instruments by musicians; secondly, an objective study, which consists in an objective description of the instruments by physical measurements (impedance). We correlated the intonation assessments data and the physical measurements, in order to deduct useful objective functions for the design of a new instrument, and to formulate the user-centered design problem as a multi-objective optimization problem. The paper presents next how GAs were implemented to solve the multi-objective optimization problem.

Stagewise Multidisciplinary Design Optimization Formulation for Optimal Design of Expendable Launch Vehicles

DOI: 10.2514/1.52507 Optimal design of launch vehicles is a complex process that gathers a series of disciplines. The classical method used to solve such problems consists in decomposing the problem into the different disciplines and in associating a global optimizer and disciplinary analyzers (multidiscipline feasible method, most used in launch vehicle design). Thispaper presents a new multidisciplinary design optimization method basedon atransverse decomposition of the design process adapted to the multistage launch vehicle architecture. The proposed bilevel method splits up the optimizationprocessintodifferent flightphasesandperformsthedifferentstageoptimizationseithersequentiallyor concurrently. Thus, the proposed approach transforms the global multidisciplinary design optimization problem into the coordination of elementary multidisciplinary design optimization problems and moves the problem complexity from the system level to the subsystem level. Three formulations of this method are proposed and compared with the multidiscipline feasible method on a multistage launch vehicle design problem. The proposed methodallowsthedimensionofthesearchdomainandthenumberofconstraintsatthesystemleveltobereduced.In thatway,thisapproachmakestheuseofheuristicmethodssuchasthegeneticalgorithmsmoreefficientinsolvingthe large-scale highly nonlinear launch vehicle design problem.

Multi-agent environment for process planning elaboration

Presents the development of a CAPP framework based on a multi-agent architecture. It generates nonlinear process plans (NLPP) and will allow interactivity with the process planners. The multi-agent architecture is discussed and detailed.

The Virtual Manufacturing Concept: Scope, Socio-Economic Aspects and Future Trends

The research area “Virtual Manufacturing (VM)” is the use of information technology and computer simulation to model real world manufacturing processes for the purpose of analysing and understanding them. As automation technologies such as CAD/CAM have substantially shortened the time required to design products, Virtual Manufacturing will have a similar effect on the manufacturing phase thanks to the modelling, simulation and optimisation of the product and the processes involved in its fabrication. After a description of Virtual Manufacturing (definitions and scope), we present some socio-economic factors of VM and finaly some “hot topics” for the future are proposed.

Collaborative Optimization Strategy for Multi-Objective Design

The increasing economic competition of all industrial markets and growing complexity of engineering problems lead to a progressive specialization and distribution of expertise, tools and works. On the other hand, engineering products becomes more and more complex and the designer has to face with an increase design variables and design objectives. Besides multi-objective optimization (MOO) and multi-disciplinary design optimization (MDO) are more commonly used as methods to provide optimal solutions for complex design problems. The paper describes an innovative mixing between genetic algorithms (MOGA) and collaborative optimization (CO) as a tool to: 1) increase the convergence rate when a design problem can be broken up regarding design variables, and 2) provide an optimal set of design variables in case of multi-level design problem. This method gives multidisciplinary optimization the advantages AG has brought to multi-objective optimization. The method, tested on test functions, assures high optimization results containing CPU times.Copyright

Compromise Based Evolutionary Multiobjective Optimization Algorithm for Multidisciplinary Optimization

Multidisciplinary Design Optimization deals with engineering problems composed of several sub-problems – called disciplines – that can have antagonist goals and thus require to find compromise solutions. Moreover, the sub-problems are often multiobjective optimization problems. In this case, the compromise solutions between the disciplines are often considered as compromises between all objectives of the problem, which may be not relevant in this context. We propose two alternative definitions of the compromise between disciplines. Their implementations within the well-known NSGA-II algorithm are studied and results are discussed.

Probe Definition for Inspection of Mechanical Parts by C.M.M.

Accessibility of measured points to touch probes is crucial in planning dimensional inspection of mechanical parts by Co-ordinate Measuring Machines (CMM). This paper presents a new method to define the kind of touch probe needed to measure the various points of a mechanical part. The limits of the notion of visibility as it is usually defined for machining purposes are shown. We propose a new definition of visibility, dedicated to measurement. A method allowing the determination of the probe’s characteristics (maximum length of the stylus, sufficient number of styli) is described.

Preparation and Setting of the Production Concerning a Virtual Manufacturing in a Cooperative Work

Abstract Whereas lots of research works have been carried out on mechanical design, a few have been done about manufacturing and more precisely its preparation and the setting of a production. The setting of the production is extremely complex because of all the different agents, systems and heterogeneous activities concerned. The work of agents and operators can be easier thanks to the hardware and software means that is to say the network, modular mounting, database, loading and unloading NC programs.…involved. However these means are not sufficient, so we have to reorganize the whole process managing the new tools of the groupwork especially since those steps are seen in the context of virtual manufacturing.

Path Compensation with Respect to Manufacturing Tolerances

Abstract In this paper, we present our research dealing with the choice of the reference of compensation with respect to a given tolerance and the compensation tool trajectory in the milling process. The general purpose of our research is to compensate the tool deflection that occurs during the milling process. We consider it under two aspects : surface prediction generated by the tool deflection and surface prediction generated by the contact points between the cutting tool flute and the workpiece. These two aspects imply two different approaches to carry out the compensation method, called mirror method , which can generate a new tool trajectory. In order to compare these two approaches, we present some practical examples.