Search Results

One of the major challenges in multiobjective, multidisciplinary design optimization (MDO) is the long computational time required in evaluating the new designs' performances. To shorten the cycle time of product design, a data mining-based strategy is developed to improve the efficiency of heuristic optimization algorithms. Based on the historical information of the optimization process, clustering and classification techniques are employed to identify and eliminate the low quality and repetitive designs before operating the time-consuming design evaluations. The proposed method improves design performances within the same computation budget. Two case studies, one mathematical benchmark problem and one vehicle side impact design problem, are conducted as demonstration.

Analytical models (math or computer simulation models) are typically built on the basis of many assumptions and simplifications and hence model prediction could be inaccurate in intended applications. Model validation is thus critical to quantify and improve the degree of accuracy of these models. So far, little work considers model validation for various design configurations so that model prediction is accurate in the intended design space. Furthermore, there is a lack of effective approaches that can be used to quantify model accuracy considering different number of experimental data. To overcome these limitations, objective of this paper is to develop a model validation approach for various design configurations with a reference metric for model accuracy check considering different number of experimental data.

Particle swarm optimization (PSO) is a relatively new stochastic optimization algorithm and has gained much attention in recent years because of its fast convergence speed and strong optimization ability. However, PSO suffers from premature convergence problem for quick losing of diversity. That is to say, if no particle discovers a new superiority position than its previous best location, PSO algorithm will fall into stagnation and output local optimum result. In order to improve the diversity of basic PSO, design of experiment technique is used to initialize the particle swarm in consideration of its space-filling property which guarantees covering the design space comprehensively. And the optimization procedure of PSO is divided into two stages, optimization stage and improving stage. In the optimization stage, the basic PSO initialized by Optimal Latin hypercube technique is conducted.

Pending reductions in light duty vehicle PM emissions standards from 10 to 3 mg/mi and below will push the limits of the gravimetric measurement method. At these levels the PM mass collected approaches the mass of non-particle gaseous species that adsorb onto the filter from exhaust and ambient air. This introduces an intrinsic lower limit to filter based measurement that is independent of improvements achieved in weighing metrology. The statistical variability of back-up filter measurements at these levels makes them an ineffective means for correcting the adsorption artifact. The proposed subtraction of a facility based estimate of the artifact will partially alleviate the mass bias from adsorption, but its impact on weighing variability remains a problem that can reach a significant fraction of the upcoming 3 and future 1 mg/mi standards. This paper proposes an improved PM mass method that combines the gravimetric filter approach with real time aerosol measurement.

Design optimization methods are commonly used for weight reduction subjecting to multiple constraints in automotive industry. One of the major challenges remained is to deal with a large number of design variables for large-scale design optimization problems effectively. In this paper, a new approach based on fuzzy rough set is proposed to address this issue. The concept of rough set theory is to deal with redundant information and seek for a reduced design variable set. The proposed method first exploits fuzzy rough set to screen out the insignificant or redundant design variables with regard to the output functions, then uses the reduced design variable set for design optimization. A vehicle body structure is used to demonstrate the effectiveness of the proposed method and compare with a traditional weighted sensitivity based main effect approach.

ACOUSTOMIZE™ is a new method of acoustic evaluation used for the purpose of understanding and optimizing NVH performance of vehicles. The following paper documents a case study of the ACOUSTOMIZE™ test methodology on a passenger car BIW. This study includes an analysis of noise flow through BIW locations, a comparison of noise sound levels through BIW cavities with and without a sound treatment package and a comparison of the original cavity sealing design package consisting of baffles, tapes and baggies to low density polyurethane NVH Foam. The results of the study show detection of complex BIW pass throughs that the body leakage test (BLT) was not able to find. In addition, the data shows improved noise reduction with the low density polyurethane foam versus the original cavity sealing design package.

A copula-based approach for model bias characterization was previously proposed [18] aiming at improving prediction accuracy compared to other model characterization approaches such as regression and Gaussian Process. This paper proposes an adaptive copula-based approach for model bias identification to enhance the available methodology. The main idea is to use cluster analysis to preprocess data, then apply the copula-based approach using information from each cluster. The final prediction accumulates predictions obtained from each cluster. Two case studies will be used to demonstrate the superiority of the adaptive copula-based approach over its predecessor.

The helical spring is one of fundamental mechanical elements used in various industrial applications such as valves, suspension mechanisms, shock and vibration absorbers, hand levers, etc. In high speed applications, for instance in the internal combustion engine or in reciprocating compressor valves, helical springs are subjected to dynamic and impact loading, which can result in a phenomenon called “surge”. Hence, proper design and selection of helical springs should consider modeling the dynamic and impact response. In order to correctly characterize the physics of a helical spring and its response to dynamic excitations, a comprehensive model of spring elasticity for various spring coil and wire geometries, spring inertial effects as well as contacts between the windings leading to a non-linear spring force behavior is required. In practical applications, such models are utilized in parametric design and optimization studies.

Design and development of an electrohydraulically actuated gas sampling valve is presented for use in auto engine combustion studies. The valve was developed with particular emphasis on sampling within the vicinity of the wall quench layer, requiring minimum leakage rates to avoid sample contamination and flush seating of the valve-stem to valve-seat to avoid perturbations of the wall layer. Response in the range of 0.4 to 1.0 milliseconds is attainable for variable valve lifts measured between 0.01 to 0.30 mm while using a net sealing force of approximately 750N. Gas leakage rates ranged from 0.05% to 1% of the sample mass flow rate when sampling from estimated distances from the wall of 0.3 mm to 0.03 mm, respectively, at a cylinder pressure of 10 bar. The gas sampling valve is presently coupled to a gas chromatograph to measure concentrations of major species components.

Model validation is a process of determining the degree to which a model is an accurate representation of the real world from the perspective of the intended uses of the model. In reliability based design, the intended use of the model is to identify an optimal design with the minimum cost function while satisfying all reliability constraints. It is pivotal that computational models should be validated before conducting the reliability based design. This paper presents an ensemble approach for model bias prediction in order to correct predictions of computational models. The basic idea is to first characterize the model bias of computational models, then correct the model prediction by adding the characterized model bias. The ensemble approach is composed of two prediction mechanisms: 1) response surface of model bias, and 2) Copula modeling of a series of relationships between design variables and the model bias, between model prediction and the model bias.

The automotive industry is investigating the change of electrical system voltage in a vehicle from the present 14 volt (12V battery) to 42 volt (36V battery) to integrate new electrical and electronic features. These new features require more amperes, thicker wires, large power devices, and eventually higher cost. The existing 14V system is very difficult to sustain so much content because of constraints of performance, efficiency, cost, packaging space, and manufacture-ability. This paper discusses foreseeable needs moving to a higher voltage, and reasons of 42V selection. It explores benefits and drawbacks when the voltage is changed from 14V to 42V in the areas of wire harness, power electronics, smart switching, power supply, etc. Finally, two typical 42/14V dual voltage architectures are presented for a likely 42V transition scenario.

A benchmark study was conducted to assess the capability of an open source CFD based process to accurately simulate the physics of the flow field around various vehicle types. The ICON FOAMpro process was used to simulate the flow field of four baseline geometries of a Truck, CD-Car, B-Car and an SUV. Further studies were carried out to assess the effects of geometry variations on the predicted aerodynamic lift and drag. A Detached-Eddy Simulation (DES) approach was chosen for the benchmarks. In addition to aerodynamic lift and drag values, the results for surface pressure data, surface and wake flow fields were calculated. These results were compared with values obtained using Ford's existing CFD processes.

In this paper, thermal stress analysis for powertrain component is carried out using two in-house developed elasto-viscoplastic models (i.e. Chaboche model and Sehitoglu model) that are implemented into ABAQUS via its user subroutine UMAT. The model parameters are obtained from isothermal cyclic tests performed on standard samples under various combinations of strain rates and temperatures. Models' validity is verified by comparing to independent non-isothermal tests conducted on similar samples. Both models are applied to the numerical analysis of exhaust manifold subject to temperature cycling as a result of vehicle operation. Due to complexity, only four thermal cycles of heating-up and cooling-down are simulated. Results using the two material models are compared in terms of accuracy and computational efficiency. It is found that the implemented Chaboche model is generally more computationally efficient than Sehitoglu model, though they are almost identical in regard to accuracy.

Robustness/Reliability Assessment and Optimization (RRAO) is often computationally expensive because obtaining accurate Uncertainty Quantification (UQ) may require a large number of design samples. This is especially true where computationally expensive high fidelity CAE simulations are involved. Approximation methods such as the Polynomial Chaos Expansion (PCE) and other Response Surface Methods (RSM) have been used to reduce the number of time-consuming design samples needed. However, for certain types of problems require the RRAO, one of the first question to consider is which method can provide an accurate and affordable UQ for a given problem. To answer the question, this paper tests the PCE, RSM and pure sampling based approaches on each of the three selected test problems: the Ursem Waves mathematical function, an automotive muffler optimization problem, and a vehicle restraint system optimization problem.

The computer-aided engineering (CAE) automation study requires a large disk space and a premium processor. If all finite element (FE) models run locally, it may crash the local machine, and if the FE model runs on high-performance computing (HPC), transferring data from the server to the local machine to do the optimization may cause latency issues. This automation study provides a unique road map to optimize the design by working efficiently using the initial setup on the local machine, running an analysis of a large number of FE models on HPC, and performing optimization on the server. CAE Automation process has been demonstrated using a case study on a driveline component, crush spacer. Crush spacer is a very critical engineering design because, first, it provides the minimum required preload to the bearing inner races to keep them in position and, second, it endures a number of duty cycles.

Corrosion tendency is one of the major inhibitors for increased use of magnesium alloys in automotive structural applications. Moreover, systematic or standardized methods for evaluation of both general and galvanic corrosion of magnesium alloys, either as individual components or eventually as entire subassemblies, remains elusive, and receives little attention from professional and standardization bodies. This work reports outcomes from an effort underway within the U.S. Automotive Materials Partnership - ‘USAMP’ (Chrysler, Ford and GM) directed toward enabling technologies and knowledge base for the design and fabrication of magnesium-intensive subassemblies intended for automotive “front end” applications. In particular, subassemblies consisting of three different grades of magnesium (die cast, sheet and extrusion) and receiving a typical corrosion protective coating were subjected to cyclic corrosion tests as employed by each OEM in the consortium.

There are benefits of using ultra thin wall (UTW) substrates (i.e., 900/2, 400/4, etc) in lowering cost and emission level. However, the more fragile mechanical characteristics of the UTW present a challenge to design and manufacture of robust catalytic converters. This paper describes a method of canning trial, where a combined Design of Experiment / Monte-Carlo analysis method was used, to develop and validate a canning method for ultra thin wall substrates. Canning trials were conducted in two stages-- Prototype Canning Trial and Production Canning Trial. In Prototype Canning Trial, the root cause of substrate failure was identified and a model for predicting substrate failure was established. Key factors affecting scrap rate and gap capability were identified and predictions were performed on scrap rate and gap capability with the allowed variations in the key factors. The results provided guidelines in designing production line and process control.

Simulation based design optimization has become the common practice in automotive product development. Increasing computer models are developed to simulate various dynamic systems. Before applying these models for product development, model validation needs to be conducted to assess their validity. In model validation, for the purpose of obtaining results successfully, it is vital to select or develop appropriate metrics for specific applications. For dynamic systems, one of the key obstacles of model validation is that most of the responses are functional, such as time history curves. This calls for the development of a metric that can evaluate the differences in terms of phase shift, magnitude and shape, which requires information from both time and frequency domain. And by representing time histories in frequency domain, more intuitive information can be obtained, such as magnitude-frequency and phase-frequency characteristics.

A cold turbocharger test facility was designed and developed at The Ohio State University to measure the performance characteristics under steady state operating conditions, investigate unsteady surge, and acquire acoustic data. A specific turbocharger is used for a thermodynamic analysis to determine the capabilities and limitations of the facility, as well as for the design and construction of the screw compressor, flow control, oil, and compression systems. Two different compression system geometries were incorporated. One system allows compressor performance measurements left of the surge line, while the other incorporates a variable-volume plenum. At the full plenum volume and a specific impeller tip speed, the temporal variation of the compressor inlet and outlet and the plenum pressures as well as the turbocharger speed is presented for stable, mild surge, and deep surge operating points.

A new diesel engine, called the 6.7L Power Stroke® V-8 Turbo Diesel and code named "Scorpion" was designed and developed by Ford Motor Company for the full-size pickup truck and light commercial vehicle markets. A high pressure Exhaust Gas Recirculation (EGR) layout in combination with a Variable Geometry Turbine (VGT) is used to deliver cooled EGR for in-cylinder NOx reduction. The cylinder-to-cylinder variation of EGR and swirl ratio is tightly controlled by the careful design of the EGR mixer and intake system flow path to reduce variability of cylinder-out PM and NOx emissions. 3D-CFD studies were used to quickly screen several EGR mixer designs based on mixing efficiency and pressure drop considerations. To optimize the intake system, 1D-3D co-simulation methodology with AVL-FIRE and AVL-BOOST has been used to assess the cylinder-to-cylinder EGR distribution and dynamic swirl.