Ryan Fellini

Optimization Approach to Hybrid Electric Propulsion System Design

Abstract Environmentally friendly ground vehicles with range and performance capabilities surpassing those of conventional vehicles require a careful balance among competing goals for fuel efficiency, performance, and emissions. The research objective here is to integrate hybrid electric vehicle simulations with high-fidelity engine modules, to increase the accuracy of predictions, and to allow design studies in the concept evaluation stage. This paper describes a methodology for integrating vehicle and engine simulations. The feed-forward model of the engine is modified to allow its linking with the vehicle model, and an engine component scaling routine is added to facilitate system sizing studies. A design optimization framework is then used to find the best overall engine size, battery pack, and motor combination for minimum fuel consumption within proposed US government performance criteria.

Platform selection under performance bounds in optimal design of product families

Designing a family of product variants that share some components usually requires a compromise in performance relative to the individually optimized variants due to the commonality constraints. Choosing components for sharing may depend on what performance losses can be tolerated. In this article an optimal design problem is formulated to choose product components to be shared without exceeding user-specified bounds on performance. This enables the designer to control tradeoffs and obtain optimal product family designs for maximizing commonality at different levels of acceptable performance. A family of automotive body side frames is used to demonstrate the approach.

PLATFORM SELECTION UNDER PERFORMANCE LOSS CONSTRAINTS IN OPTIMAL DESIGN OF PRODUCT FAMILIES

Designing a family of product variants that share some components usually entails a performance loss relative to the individually optimized variants due to the commonality constraints. Choosing com ...

Optimal design of automotive hybrid powertrain systems

Alternative powertrains for automotive applications aim at improving emissions and fuel economy. Lack of experience with these relatively new technologies makes them ideal applications for computer-based modeling and simulation studies. There is a variety of configurations, control strategies, and design variable choices that can be made. If mathematical models exist, rigorous optimization techniques can be used to explore the design space. This paper provides an overview of a design environment for alternative powertrains that has these characteristics: modularity, allowing a system to be built by combining components; flexibility allowing different levels of fidelity and different existing codes to be used; and, rigor, since it is based an mathematical methods of decision making. A simple application to a hybrid diesel-electric powertrain is included.

Optimal design decisions in product portfolio valuation

Product portfolio valuation is a core business milestone in a firm’s product development process: Determine what will be the final value to the firm derived from allocating assets into an appropriate product mix. Optimal engineering design typically deals with determining the best product based on technological (and, occasionally, cost) requirements. Linking technological with business decisions allows the firm to follow a product valuation process that directly considers not only what assets to invest but also what are the appropriate physical properties of these assets. Thus, optimal designs are determined within a business context that maximizes the firm’s value. The article demonstrates how this integration can be accomplished analytically using a simple example in automotive product development.Copyright

Quantitative platform selection in optimal design of product families, with application to automotive engine design

Product variants with similar architecture but different functional requirements may have common parts. We define a product family to be a set of such products, and refer to the set of common parts as the product platform. Product platforms enable rapid adjustment to changing market needs while keeping development costs and time-cycles low. In many cases, however, the individual product requirements are conflicting when designing a product family. The designer must balance the tradeoff between maximizing commonality and minimizing individual product performance deviations. The design challenge is to select the product platform that will generate family designs with minimum deviation from individual optima. We propose a methodology that combines two previous approaches developed for making commonality decisions. In the first approach optimal values and sensitivity information from the individually optimized variants are used to indicate components that are probable candidates for sharing. In the second approach a relaxed combinatorial problem is formulated to maximize sharing among variants subject to bounds on performance reduction for the individually optimized values. In the combined methodology the first approach is used to identify an initial set of shared components and define the candidate platform to be considered by the second approach. The computational load is reduced significantly and the platform-selection problem is solved in a more robust manner. The proposed methodology is demonstrated on the design of an automotive engine family.

Commonality Decisions in Product Family Design

Product variants with similar architecture but different functional requirements may have common parts or elements. We define a product family to be a set of such products, and refer to the set of common elements as the product platform. Product platforms enable efficient derivation of product variants by keeping development costs and time-cycles low. In many cases, however, the individual product requirements are conflicting when designing a product family. The designer must balance the tradeoff between maximizing commonality and minimizing individual product performance deviations. The design challenge is to select the product platform that will generate family designs with minimum deviation from individual optima.

Case Study for Network-Distributed Collaborative Design and Simulation: Extended Life Optimization for M1 Abrams Tank Road Arm

Abstract A framework has been developed for distributed simulation and design of complex systems, with particular application to complete vehicle systems. An object-oriented architecture has been adopted, enabling the use of distributed and heterogeneous computing resources, custom and commercial legacy software, and different models and optimization tools. The industry standard, Common Object Request Broker Architecture (CORBA), is used to provide network-distribution services. To demonstrate the capabilities of the distributed design framework, this paper presents results of a study conducted to optimize the life of an MI Abrams tank road arm. This study demonstrates the reconfigurability of the design framework, as various computing resources, simulation and analysis tools, and optimization algorithms are employed.

CORBA-Based Object-Oriented Framework for Distributed System Design

Abstract A generic framework for designing large, complicated systems is formulated, implemented, and used in the design of mechanical systems, including a pressure vessel, an automotive hybrid powertrain, and a tracked vehicle. The framework supports simulation-based design, distributed and heterogeneous computing resources, custom and legacy simulation and analysis codes, reconfigurability of the design problem, and security of operation across untrusted networks. The framework also facilitates the implementation of methodologies for system design that employ design model partitioning and coordination, as well as a variety of models and search algorithms. Common Object Request Broker Architecture (CORBA) middleware for distributed, object-oriented applications was selected to develop and implement the framework. Framework components include subsystem model, design model, search engine, design model partitioning, design coordination, and user interface.

Analytical Target Cascading in Product Family Design

Most products are neither designed nor manufactured as one piece. They are decomposed into parts that are developed individually before they are assembled to form the final product. Typically, this partitioning-based development process matches the hierarchical structure of the product-offering organization. Design tasks are assigned to divisions, departments, and teams according to expertise. An example from the automotive industry is depicted in Figure 11-1. Obviously, this decomposition is not complete and serves only as an illustration of the decomposition paradigm.