Robert Greif

Experimental Techniques for Dynamic Characterization of Composite Materials

This research combines theoretical and experimental approaches for dynamic material characterization of composite materials. The samples studied include continuous fiber graphite/epoxy beams with various symmetric lay-up configurations. Included are laminated beams with the following lay-ups: [0[sub 8]/90[sub 8]][sub s], [90[sub 8]/0[sub 8]][sub s], [(45/0/[minus]45)[sub 5]][sub s] and [(0/45/0/[minus]45)[sub 3]/90/0/0[sub 1/2]][sub s]. The resonant dwell technique is used to determine the material damping and the real part of the dynamic flexural modulus of double cantilever beam specimens in the first mode of vibration over the frequency range 25 Hz to 300 Hz. The dynamic properties are determined as a function of the frequency of oscillation at room temperature. In addition, the Metravib Viscoanalyzer, based on off-resonance tests, is also used to provide another source of experimental data for comparison. Although the Viscoanalyzer was originally intended for testing viscoelastic polymers, the present research establishes the limits of applicability for composite materials, with particular emphasis on the three point bending tests. Comparisons and limitations of both techniques are critically discussed.

Application of Nadal Limit for the Prediction of Wheel Climb Derailment

Application of the Nadal Limit to the prediction of wheel climb derailment is presented along with the effect of pertinent geometric and material parameters. Conditions which contribute to this climb include wheelset angle of attack, contact angle, friction and saturation surface properties, and lateral and vertical wheel loads. The Nadal limit is accurate for high angle of attack conditions, as the wheelset rolls forward in quasi-static steady motion leading to a flange climbing scenario. A detailed study is made of the effect of flange contact forces F(tan) and N, the tangential friction force due to creep and the normal force, respectively. Both of these forces vary as a function of lateral load L. It is shown that until a critical value of L/V is reached, climb does not occur with increasing L since F(tan) is saturated and the flange contact point slides down the rail. However, for a certain critical value of L/V (i.e. the Nadal limit) F(tan)tan is about to drop below its saturated value and flange climb (rolling without sliding) up the rail occurs. Additionally, an alternative explanation of climb is given based on a comparison of force resultants in track and contact coordinates. The effects of longitudinal creep force F(long) and angle of attack are also investigated. Using a saturated creep resultant based on both [F(tan), F(long)] produces a climb prediction L/V larger (less conservative) than the Nadal limit. Additionally, for smaller angle of attack the standard Nadal assumption of F(tan)=μN may lead to an overly conservative prediction for the onset of wheel climb. Finally, a useful analogy for investigating conditions for sliding and/or rolling of a wheelset is given from a study of a disk in rigid body mechanics.

A Methodology for the Modeling of Forced Dynamical Systems From Time Series Measurements Using Time-Delay Neural Networks

The main goal of this research was to develop and present a general, efficient, mathematical, and theoretical based methodology to model nonlinear forced-vibrating mechanical systems from time series measurements. A system identification modeling methodology for forced dynamical systems is presented based on a dynamic system theory and a nonlinear time series analysis that employ phase space reconstruction (delay vector embedding) in modeling dynamical systems from time series data using time-delay neural networks. The first part of this work details the modeling methodology, including background on dynamic systems, phase space reconstruction, and neural networks. In the second part of this work, the methodology is evaluated based on its ability to model selected analytical lumped-parameter forced-vibrating dynamic systems, including an example of a linear system predicting lumped mass displacement subjected to a displacement forcing function. The work discusses the application to nonlinear systems, multiple degree of freedom systems, and multiple input systems. The methodology is further evaluated on its ability to model an analytical passenger rail car predicting vertical wheel/rail force using a measured vertical rail profile as the input function. Studying the neural modeling methodology using analytical systems shows the clearest observations from results, providing prospective users of this tool an understanding of the expectations and limitations of the modeling methodology.

A Systems Modeling Methodology for Evaluation of Vehicle Aggressivity in the Automotive Accident Environment

A systems modeling approach is presented for assessment of harm in the automotive accident environment. The methodology is presented in general form and then applied to evaluate vehicle aggressivity in frontal crashes. The methodology consists of parametric simulation of several controlled accident variables, with case results weighted by the relative frequency of each specific event. A hierarchy of models is proposed, consisting of a statistical model to define the accident environment and assign weighting factors for each crash situation case, and vehicle and occupant models for kinematic simulation of crash events. Head and chest injury results obtained from simulation are converted to harm vectors, in terms of probabilistic Abbreviated Injury Scale (AIS) distributions based on previously defined risk analyses. These harm vectors are weighted by each case’s probability as defined by the statistical model, and summed to obtain a total estimate of harm for the accident environment. The methodology is applied to a subset accident environment consisting of singleand two-vehicle frontal collisions among passenger cars and light trucks. The model is validated against recent crash statistics, and is found to accurately reflect trends in distribution of injury severity while slightly underestimating moderate to severe injuries. The model is subsequently exercised for variable sensitivity analyses, wherein the effects of light truck/car population mix are evaluated in terms of their impact on occupant harm within the subset accident environment. 1.0 Introduction This paper presents a systems modeling approach for evaluation of overall safety in the automotive fleet. This methodology stands in contrast to typical approaches, where specific safety issues such as air bags are addressed independently. However, the recent surge in light truck sales in the U.S. has led to the advent of a broader problem: how to evaluate the aggressivity of these large heavy vehicles in twovehicle accidents while also considering their potential safety benefits in single-vehicle crashes. While light truck vehicles do provide added protection to occupants within the vehicle, one recent statistical study reports that light trucks are so aggressive due to both mass and geometry that in head-on crashes between cars and light trucks, deaths in the cars outnumber those in the light trucks by 70% (Joksch, 1998). The systems model methodology applied here features computational vehicle models to represent cars and light trucks, making it suitable for analysis of aggressivity and compatibility among dissimilar vehicles. This paper describes a systems modeling methodology for prediction of passenger injuries across the entire accident environment, considering a variety of metrics including vehicle type, impact speed, occupant size, safety belt usage, and other factors which directly affect overall safety. This approach will allow for evaluation of global effects of small changes to the accident environment, so that proposed automotive safety regulations may be evaluated in terms of their total safety benefit. The methodology has been developed as a generalized tool for assessment of a variety of crashworthiness topics, such as air bags and vehicle design characteristics. The methodology is applied here to study vehicle aggressivity in terms of the relationship between passenger vehicle fleet mix and overall harm. History. Several previous studies have considered a systems approach for investigating vehicle safety. During 1975-78, the Ford Motor Company developed the Safety Systems Optimization Model (Ford Motor Co., 1978), featuring a simulation-optimization program for maximizing a single vehicle’s safety performance in frontal crashes. The same program was substantially modified by the University of Virginia in the early 1980s (White, et al., 1985), to include Figure 1. Fleet Systems Model Methodology new biomechanical transforms and updated accident data as well as multivariate analysis capability. This model utilized approximating functions to estimate relationships between crash variables due to limitations in computational power at the time. Other motor vehicle manufacturers, including Fiat and Volkswagen, have also developed programs for optimizing vehicle design for crashworthiness, with emphasis on single-vehicle as opposed to fleet wide performance. The model presented here differs from these earlier models in several aspects. It predicts total harm over a range of vehicle types rather than a single subject vehicle. While the model estimates injuries over a given set of crashes, it does not include an optimization algorithm for minimization of total harm. The model considers air bags in addition to seat belts, and occupants of varying size. It also incorporates recent accident statistics and more sophisticated biomechanical transforms than earlier approaches. Furthermore, due to improvements in structural modeling techniques and computer efficiency, the model includes parametric simulation of a range of vehicle and occupant crashworthiness models. Governing Equation and Methodology. The methodology is based upon the following governing equation for estimation of total injuries:

Impact behavior of recycled core composite polymeric enclosures

The growing awareness about the impact of non-biodegradable polymeric waste on the environment and the associated cost benefits have led to extensive use of recycled materials. The properties of polymers degrade once they are recycled. This paper presents a comparison of the impact behavior of virgin and recycled polymers as a function of thickness. The thickness is an important parameter in design of polymeric enclosures and as such, the impact energy as a function of thickness needs to be optimized. The ‘DSGZ’ phenomenological constitutive model, developed at Tufts University, uniformly describes the entire range of stress–strain constitutive relationship of polymers under any monotonic loading mode, and is used to predict the plastic failure energies. This paper compares the impact behavior and impact energies of a monolithic virgin PBT (polybutylene-terephthalates) to that of recycled ABS (acrylonitrile–butadiene–styrene)/ASA (acrylic–styrene–acrylonitrile) composite enclosures. The composite enclosures consist of skin layers of ASA at the top and bottom with a center core of recycled ABS. ABAQUS/Explicit and finite element analysis are used to model the enclosures, calculate the plastic failure energies and develop a better understanding of the impact behavior of polymers for enclosure applications.

Effect of Cant Deficiency on Rail Vehicle Performance

Simplified train models are analyzed to assess the relationship of unbalance on carbody acceleration and wheel unloading during steady state curving motion. In this paper a half-car model appropriate for both power cars and tilting coach cars is theoretically analyzed. Models of this type are useful for examining static lean requirements as well as margins of safety at higher cant deficiencies in the Track Safety Standards of the Federal Railroad Administration (FRA). The suspension systems modeled and analyzed include the following types: rigid, flexible, tilting actuation, and combined flexible and tilting actuation suspension. Simplified formulas are derived which can be used as an analysis tool by railroad designers to assess vehicle performance. Parametric results are presented for vertical wheel unloading and lateral carbody acceleration as a function of cant deficiency. Results show that incorporation of tilting systems, better suspension designs and better track quality, are necessary in order to provide an equivalent level of margin of safety for operations at higher cant deficiency. The relationship of these results to limits in the Track Safety Standards is discussed.Copyright

Application of Time Series Analysis and Neural Networks to the Modeling and Analysis of Forced Vibrating Mechanical Systems

A theoretical and mathematical based methodogy is discussed that utilizes time series analysis techniques and neural networks to model forced vibrating mechanical systems using measured input-output data. A technique in nonlinear time series analysis known as phase space reconstruction may be used to extend our understanding of the active dynamics recorded in a single time series measurement. Using a recorded output (response) measurement phase space reconstruction parameters are calculated; the embedding dimension is estimated using the method of false nearest neighbor, and the time delay is estimated from the first minimum of the mutual information. The phase space reconstruction characteristics are then used to fully shape the architecture of a time delayed neural network model for the dynamical system. The modeling methodology is applied to several forced vibrating systems common to many fields of engineering. The neural models are then used to analyze new input, demonstrating the usefulness and importance of the methodology.Copyright

A Methodology for the Modeling of Rail Vehicles From Time Series Measurements Using Time-Delay Neural Networks

The main goal of this research is to develop and demonstrate a general, efficient, mathematically and theoretically based methodology to model nonlinear forced vibrating mechanical systems from time series measurements. A system identification modeling methodology for forced dynamical systems is presented based on dynamic system theory and nonlinear time series analysis that employs phase space reconstruction (delay vector embedding) for modeling of dynamical systems from time series data using time-delay neural networks (TDNN). The first part of this work details the modeling methodology including background on dynamic systems, phase space reconstruction, and neural networks. In the second part of this work the methodology is evaluated based on its ability to model selected analytical lumped parameter forced vibrating dynamic systems including an example of a linear system predicting lumped mass displacement using a displacement forcing. function The work discusses the application to nonlinear systems, multi degree-of-freedom systems, and multi-input systems. The methodology is further evaluated on its ability to model an analytical passenger rail vehicle predicting vertical wheel/rail force using vertical rail profile as input. Studying the neural modeling methodology using an analytical systems shows the clearest observations from results which provide prospective users of this tool an understanding of the expectations and limitations of the modeling methodology.Copyright

Methodology for Modeling of Passive Shunt Damping of Systems With Bonded Piezocomposites

An aluminum cantilever beam bonded with 1-3 piezocomposite dampers is modeled by means of ANSYS finite element and SIMULINK simulation softwares. ANSYS currently cannot account for heat dissipation in piezoelectric materials. As such, ANSYS is used to obtain strain energies to be input into the SIMULINK model to investigate the dynamic behavior of the system and calculate the damping ratio. The impact of two different shunting arrangements, a damper in conjunction with a simple resistive electrical circuit in series and parallel, is investigated. In addition, a simply supported beam and a simply supported straight pipe are also analyzed for their wide applications in industry, and as an indication of the utility of this methodology to analyze complex structural configurations. For a typical cantilever beam, energy dissipation and transient analysis are used to calculate the tip displacement as a function of time and the damping ratio. Then using ANSYS, with the parameter BETAD to incorporate damping as a stiffness multiplier, a comparison of the transient results is used to quantify the damping response of aluminum beams with bonded 1-3 piezocomposite dampers. The system loss factor due to the piezoelectric damping is also compared to the inherent loss factor of different beam materials. The results show that circuits in series provides a better damping ratio (0.000581) as compared to circuits in parallel (0.000374). In addition, for different boundary conditions (cantilever, simply supported), the damping ratios (0.000581, 0.000202) and the BETAD values (6.3 E-6, 0.7 E-6), respectively, are functions of the boundary conditions and are not directly related to each other. Finally, damping using 1-3 piezocomposites effectively increases the overall system loss factor by at least 100% to almost 300% as compared to the inherent material damping. In general, this methodology of combining finite element method (ANSYS) and transient modeling tools (SIMULINK) can be used to study damping characteristics of any structural system damped with 1-3 piezocomposites.Copyright

Passive Vibration Damping of Aluminum by 1-3 Piezocomposites

The effective properties of 1-3 piezocomposites are used to examine their passive vibration damping characteristics. An aluminum cantilever beam bonded with 1-3 piezocomposite dampers is modeled by means of “ANSYS” and “SIMULINK” softwares to investigate the dynamic behavior of the system. A method of determining the damping ratio introduced by the piezocomposite damper in conjunction with a simple resistive electrical circuit is established. The effect of volume fraction of the 1-3 piezocomposite on the damping of the system is analyzed. Damping ratio is observed to increase with rising volume fraction. At low volume fractions, the participation of piezoelectric fibers in the load-bearing pattern is to a lesser extent and hence the damping ratio is low. On the contrary as the volume fraction rises, the involvement of piezoelectric fibers increases resulting in higher damping ratios. Given that the inherent material damping in the aluminum beam is 0.0002, the additional damping provided by the bonded piezoelectric strips goes up to a maximum of 0.0042. Finally, the methodology developed in this paper can be used to model any type of vibratory structural system to determine the damping introduced by the piezocomposites.Copyright