Paolo Guarneri

Proceedings of the ASME Design Engineering Technical Conference

The meta-material design of the shear layer of a non-pneumatic wheel was completed using topology optimization. In order to reduce the hysteretic rolling loss, an elastic material is used and the shear layer microstructure is defined to achieve high compliance comparable to that offered by the elastomeric materials. To simulate the meta-material properties of the shear layer, the volume averaging analysis, instead of more popular homogenization methods, is used as the relative size of the shear layer places realistic manufacturing constraints on the size of unit cells used to generate the meta-material. In this design scenario the properties predicted by the homogenization methods are not accurate since the homogenization scaling assumptions are violated. A number of optimal designs are shown to have meta-material properties similar to those of the linear elastic properties of elastomers, making them good meta-material candidates for the shear layer of the non-pneumatic wheel.Copyright

Equitable Multi-Objective Optimization Applied to the Design of a Hybrid Electric Vehicle Battery

This work considers the impact of thermal behavior in battery design. The cell performance worsens when the operating temperature falls outside of the ideal range, and evenness of cell temperatures is sought to prevent cell electrical unbalance which may lead to performance fading and premature failure. The heat transfer between the cells and the coolant depends on the cell packaging and layout. A multi-objective optimization model is posed whose Pareto efficient designs minimize cell temperature deviations while maintaining evenness of temperature distribution. The special characteristics of the battery design problem (comparable objectives, anonymity and Pigou–Dalton principle of transfers) make it suitable for the application of the equitability preference, which is a refinement of the Pareto optimality that has not been used in engineering design. The proposed approach based on equitability is applied to compute the spacing of the cylindrical cells in a battery module that yields an optimal thermal behavior.

Bi-level Approach to Vehicle Component Layout With Shape Morphing

Engineering research into packing problems has been widely undertaken in recent years. The use of component shape morphing in layout design has, however, received little attention. Shape morphing is required for fitting a component of sufficient size in a limited space while optimizing the overall performance objectives of the vehicle and improving design efficiency. To morph components that can have arbitrary shapes in layout design, a mass-spring physical model-based morphing method is proposed and implemented. Vehicle layout design with shape morphing is a multi-objective, multilevel problem with a large number of design variables. To solve this large scale problem, decomposition is adopted. At the system level, the overall performance objectives are optimized with respect to locations and orientations of components. At the component level, deformable objects are morphed to fit in the available space. A vehicle underhood layout design problem is demonstrated to illustrate the proposed approach.

Optimizing the Shear Beam of a Non-Pneumatic Wheel for Low Rolling Resistance

The design requirements of a low rolling loss non-pneumatic wheel are determined through a systematic optimization approach. In order to reduce the rolling resistance, linear elastic materials are considered instead of elastomers. To achieve an adequate compliance level, a metamaterial needs to be designed. The required metamaterial properties are determined as a result of an optimization where the metamaterial tensor components as well as the geometric dimensions are the design variables. This way the metamaterial can be designed such that the overall behavior of the non pneumatic wheel achieves the best performance in terms of compliance and contact patch pressure distribution. The resulting constitutive metamaterial properties of the shear layer can be used as prescribed constitutive properties to tailor the periodic mesostructure of a material by means of topology optimization.Copyright

Network target coordination for optimal design of decomposed systems with consensus optimization

The complexity of managing multidisciplinary engineering systems offers an unprecedented opportunity to investigate decomposition methods, which separate a system into a number of smaller subsystems that can be designed in multiple physical locations and coordinate the design of the subsystems to collaboratively achieve the original system design. This paper studies a network target coordination model for optimizing subsystems that are distributed as multiple agents in a network. To solve these coupled subsystems concurrently, we consider the “consensus optimization” approach by incorporating subgradient algorithms so that the master problem or auxiliary design variables required by most distributed coordination methods are not needed. The method allows each agent to conduct its optimization by locally solving for coupling variables with the information obtained from other agents in the network in an iteratively improving process. The convergence results of a geometric programming problem that satisfies the convexity assumption is provided. Moreover, two non-convex examples are tested to investigate the convergence characteristics of the proposed methods.Copyright

Genetic Algorithms Applied to Affordance Based Design

Affordance based design (ABD) theory has been presented in several papers and has interested several researchers in the field of design. One criticism of ABD is that the number of affordances identified can be very large, and therefore, the approach may not be amenable for automation. This paper presents a computer based implementation of a process to improve design using affordances.The Affordance Structure Matrix (ASM) design tool is used to identify relevant relationships between the design parameters of an artifact and the affordances (positive and negative) identified by the user. This initial work investigates multiple existing solutions. A user assigns values to critical affordances that are listed in an ASM by visualizing the possible solutions. The parameter values that describe the architecture of the artifact are encoded and fed to a computer code in addition to the multiple affordance values. A Genetic Algorithm is then used to find an optimal combination of design parameters based on the multiple criteria, the affordances, generating new and better concepts.The approach is applied to the redesign of a steering wheel. Multiple variants of steering wheels available in the literature are presented to the user. After twenty generations of the genetic algorithm, using an additive weighting method, an optimal solution is found and presented.Copyright

Using non-simply connected unit cells in the multiscale analysis and design of meta-materials

A method to discretize materials into lattices using non-simply connected, rectangular unit cells such that the corners of the unit cells do not necessarily meet is proposed as an alternative to discretizing with non-square, parallelogram unit cells. When analyzing non-simply connected lattices, the methods commonly used to evaluate linear elastic properties, homogenization and volume averaging, are shown to require only minor changes in their nite element implementations. The linear elastic properties of honeycomb structures, decomposed into non-simply connected lattices, are shown to demonstrate the same convergence characteristics illustrated in the literature for periodic materials with simply connected unit cells. Additionally, these mechanical properties agree with those predicted by established analytical formulae. Finally, topology optimization examples are used to demonstrate the potential application of the proposed methods to meta-material design. Notably, an auxetic honeycomb structure is obtained as a local minimum to a simple optimization problem.