Vincent Y. Blouin

Bi-objective optimization design of functionally gradient materials

In this paper, a procedure for bi-objective optimization design of functionally gradient materials (FGM) is presented. Different microstructures formed by two primary materials are evaluated by a micromechanical analysis method. Macroscopically, FGMs are optimally designed by using these microstructures. Instead of using conventional simply assumed power law material distribution functions, a generic material distribution function is used. The bi-objective FGM optimization design procedure is highlighted by a flywheel example. A parametric formulation is used for both the geometric representation and the optimization procedure.

Final Shape of Precision Molded Optics: Part I—Computational Approach, Material Definitions and the Effect of Lens Shape

Coupled thermomechanical finite element models were developed in ABAQUS to simulate the precision glass lens molding process, including the stages of heating, soaking, pressing, cooling and release. The aim of the models was the prediction of the deviation of the final lens profile from that of the mold, which was accomplished to within one-half of a micron. The molding glass was modeled as viscoelastic in shear and volume using an n-term, prony series; temperature dependence of the material behavior was taken into account using the assumption of thermal rheological simplicity (TRS); structural relaxation as described by the Tool-Narayanaswamy-Moynihan (TNM)-model was used to account for temperature history dependent expansion and contraction, and the molds were modeled as elastic taking into account both mechanical and thermal strain. In Part I of this two-part series, the computational approach and material definitions are presented. Furthermore, in preparation for the sensitivity analysis presented in Part II, this study includes both a bi-convex lens and a steep meniscus lens, which reveals a fundamental difference in how the deviation evolves for these different lens geometries. This study, therefore, motivates the inclusion of both lens types in the validations and sensitivity analysis of Part II. It is shown that the deviation of the steep meniscus lens is more sensitive to the mechanical behavior of the glass, due to the strain response of the newly formed lens that occurs when the pressing force is removed.

Design for Manufacturing of 3D Heterogeneous Objects With Processing Time Consideration

Rapid prototyping (RP) technology, such as laser engineering net shaping, can be used to fabricate heterogeneous objects with various levels of gradient variations in material composition. These objects are engineered to achieve a potentially enhanced functional performance. Past research on the design of such objects has focused on representation, modeling, and desired functional performance. However, the inherent constraints in RP processes, such as system capability and processing dwell time, lead to heterogeneous objects that may not meet the designer’s original intent. To overcome this situation, the research presented in this paper focuses on the identification and implementation of manufacturing concerns into the design process. Previous work on a 2D disk brake rotor design has shown that processing dwell time is one of the critical factors that affect manufacturability. This paper focuses on incorporating the processing time into the optimization design for manufacturing of 3D heterogeneous objects. A node-based finite element modeling technique is used for the representation and analysis. The multicriteria design problem corresponds to finding the nodal material compositions with minimized structural weight, maximized structural stiffness, and minimized extra processing time used to deposit the multimaterial subjected to stress constraints. The optimizer used in this research is a self-adaptive, real-valued evolutionary strategy, which is well suited for this type of multimodal problem. A 3D I-beam made of two materials, aluminum for lightweight and steel for better strength characteristics, is used to illustrate the trade-off between manufacturability and functionality.

FINAL SHAPE OF PRECISION MOLDED OPTICS: PART II—VALIDATION AND SENSITIVITY TO MATERIAL PROPERTIES AND PROCESS PARAMETERS

In Part I of this study a coupled thermo-mechanical finite element model for the simulation of the entire precision glass lens molding process was presented. That study addressed the material definitions for the molding glass, L-BAL35, computational convergence, and how the final deviation of the lens shape from the mold shape is achieved for both a bi-convex lens and a steep meniscus lens. In the current study, after validating the computational approach for both lens types, an extensive sensitivity analysis is performed to quantify the importance of several material and process parameters that affect deviation for both lens shapes. Such a computational mechanics approach has the potential to replace the current trial-and-error, iterative process of mold profile design to produce glass optics of required geometry, provided all the input parameters are known to sufficient accuracy. Some of the critical contributors to deviation include structural relaxation of the glass, thermal expansion of the molds, TRS and viscoelastic behavior of the glass and friction between glass and mold. The results indicate, for example, the degree of accuracy to which key material properties should be determined to support such modeling. In addition to providing extensive sensitivity results, this computational model also helps lens molders/machine designers to understand the evolution of lens profile deviation for different lens shapes during the course of the process.

Multi-scenario Multi-objective Optimization with Applications in Engineering Design

The notion of multi-scenario multi-objective optimization is proposed as a methodological framework for handling engineering design and other decision problems represented as a collection of multi-criteria optimization problems. Three specific research issues are discussed in this context, namely, the modelling of decision makers preferences, the development of a concept of optimality, and the development of solution approaches to finding a preferred feasible solution for the overall problem. Two models of preferences that generalize the classical Pareto preference and two solution approaches to a class of multi-scenario multi-objective optimization problems are presented. Illustrative examples are included.

Modeling Relative Importance of Design Criteria With a Modified Pareto Preference

Engineering design problems are studied within a multicriteria optimization and decision-making framework. A methodology is developed that modifies the traditional Pareto preference to model designers preferences reflected in the relative importance of criteria. The intent is to reduce the number of candidate designs to facilitate the selection of a preferred design. The versatility of this preference model allows it to be incorporated into the problem solution process either a priori, a posteriori, or iteratively, each offering different advantages. In the a priori approach, all Pareto efficient designs that do not satisfy the designers preferences are never computed. In the a posteriori approach, a set of Pareto efficient designs is computed and then easily reduced based on the designers preferences. Finally, the iterative approach offers the ability to adjust the designers preferences by exploring their impact on the reduction of the Pareto efficient design set. The methodology is based on the concepts of convex cones and allowable tradeoff values between criteria. The theoretical foundation of the preference model is presented in the context of engineering design and the methodology is illustrated using a bi-criteria structural design problem and a tri-criteria vehicle dynamics design problem.

Using Genetic Algorithms to Identify Initial Operating Conditions for a Transient CVT Model

A Continuously Variable Transmission (CVT) provides a continuum of gear ratios between desired limits. CVT is a promising automotive technology and a sundry of models has been researched to realize the potential benefits of a CVT. CVT, being a highly nonlinear system, has a definite operating regime where it is able to maximize the torque transmission. The numerical model presented in this paper is difficult to solve because of its sensitivity with respect to the initial operating conditions such as initial belt tension, axial forces, and driven preload. The present research focuses on using Genetic Algorithms (GA) to identify these operating conditions and to understand the various dynamic interactions in a metal pushing V-belt CVT. This paper uses continuous Coulomb friction approximation theory to model friction between the belt and the pulleys. The computational scheme, the mathematical models, and the results corresponding to different loading scenarios are discussed.Copyright

Optimal design for additive manufacturing of heterogeneous objects using ultrasonic consolidation

Rapid prototyping (RP) technologies, such as Laser Engineering Net Shaping (LENS®) and Ultrasonic Consolidation (UC), can be used to fabricate heterogeneous objects composed of more than one material, wherein spatially varied microscopic structural details produce continuously or discretely changing mechanical or thermal properties on a macroscopic scale. These objects are engineered to achieve a potentially enhanced functional performance. Past research on the design of such objects has focused on representation, modeling, and desired functional performance. However, the inherent constraints in RP processes, such as system capability, size and shape of raw materials, and processing time, lead to fabricated objects that may not meet the designers original intent. To overcome this situation, the research presented in this paper focuses on developing an approach— Design for Additive Manufacturing (DfAM)—to implement identified manufacturing constraints into the design process. Previous work has applied DfAM to the design of heterogeneous objects fabricated using the LENS® process. Two manufacturing constraints for this process, namely the achievable volume fractions and the processing time, were identified and incorporated into the DfAM. In this paper, the DfAM approach is extended to the design and manufacture of heterogeneous objects for the UC process. Constraints on the possible volume fraction values and on the gradient material direction are two identified manufacturing limitations, which are incorporated into the design process. An element-based finite element (FE) representation is extended to model layered heterogeneous objects. Each element is composed of metal foils of different materials according to specific design parameters. An evolutionary-based optimizer is used for its ability to handle the type of multi-modal problems encountered in the design of heterogeneous objects. The multi-criteria design problem, consisting of finding the optimal material composition along the build direction, that satisfies the functions of minimum weight and structural deformation, is implemented and solved. A three-dimensional I-beam made of two materials—aluminum for lightweight and steel for better strength characteristics—is used to illustrate the DfAM approach and its implementation for the design of heterogeneous objects using the UC process.

Multi-Objective Configuration Optimization With Vehicle Dynamics Applied to Midsize Truck Design

Recent improvements in vehicle propulsion systems, such as hybrid electric and fuel cells, demand new configuration solutions that may be totally different from conventional designs. Packaging of vehicle components is still a new area of research. This paper describes a configuration optimization method based on a multiple objective genetic algorithm. The method is applied to configuration optimization of a mid-size truck, in which two objectives are considered: ground clearance and dynamic behavior. A vehicle packaging model was developed using the commercial CAD software, ACIS, to analyze interference among vehicle components. An eight-degree of freedom model was used to analyze the dynamic behavior of a given configuration for the J-turn maneuver. Parallel computation technology was also incorporated to accelerate the optimization process. The applicability of this method is discussed and exemplified with the design of a mid-size truck for two propulsion systems: conventional diesel and hybrid diesel electric. A set of Pareto solutions is generated in which tradeoff decisions can be made to select a final design.Copyright

Intrinsic Analysis of Decomposition and Coordination Strategies for Complex Design Problems

Optimization has been given increased attention in both academic research and industrial applications. The reason appears to be twofold. First, keen global competition is forcing industries to maximize product performance while utilizing minimum resources in order to gain a competitive edge. Second, the significant advances in computer hardware allow optimization techniques to be deployed on problems of higher complexity. One must realize, however, the persisting insufficiency of current optimization techniques in solving full-scale engineering problems. Despite the technical progress of MDO in developing specific decomposition and coordination (D&C) strategies, it is not clear how and when to apply the various design and analysis tools that are being developed. As a result, the industry has not fully benefited from these research efforts. Three specific types of D&C strategies are Concurrent Sub-Space Optimization (CSSO), Collaborative Optimization (CO), and Analytic Target Cascading (ATC). While these methods have been demonstrated on well-structured and fully defined design problems, there is a lack of information on which is most appropriate for use in real world large-scale complex design problems. As a first step in this direction, we present an intrinsic evaluation of these three strategies. First we review the most common MDO strategies, including CSSO, CO, and ATC. Then we describe the attributes of complexity and assess technical and non-technical issues involved in large-scale complex design problems. Finally, we present the characteristics of the strategies that may be employed to develop comparative metrics and discuss the potential strengths and weaknesses of the three D&C strategies in the context of real-world industrial applications.