DSpace Repository :: Browsing by Author "Drignei, Dorin"

Browsing by Author "Drignei, Dorin"

Now showing 1 - 9 of 9

A Simulation-Based Fatigue Life Estimation Method for Nonlinear Systems under Non-Gaussian Loads
(2023-01-01) Mande, Onkar K; Mourelatos, Zissimos P.; Gu, Randy J; Monroe, Ryan; Drignei, Dorin
In the fields of durability and stochastic structural dynamics, it is customary to focus on linear structures subjected to Gaussian excitations. However, real-world engineering systems often exhibit nonlinear behavior and are exposed to non-Gaussian loads. Calculating fatigue life for such nonlinear systems under non-Gaussian loading presents many challenges such as complex nonlinear dynamics, multifaceted statistical characteristics, and time-dependent effects resulting in a very high computational effort. To overcome these hurdles, this research uses non-Gaussian Karhunen-Loeve expansion (NG-KLE) to not only predict the expected fatigue life but also obtain the Probability Density Function (PDF) of fatigue life. It integrates a sub-domain-based technique to significantly reduce the computational demands while preserving accuracy, by efficiently obtaining long time trajectories of random processes. This development is very useful for excitation signals that far exceed the process correlation length. The NG-KLE method serves as the main tool for characterizing the excitation process by estimating its non-Gaussian marginal distribution and autocorrelation function. A Karhunen-Loeve (KL) expansion is executed only for the first subdomain, and then extended to subsequent subdomains by establishing correlations between the KL expansion coefficients of adjacent subdomains. This innovative approach is adapted to non-Gaussian (NG) excitation, allowing for efficient characterization of both the input and output random processes using NG-KLE, enabling the generation of very long synthetic output random stress process samples. The fatigue life corresponding to each output stress trajectory contributes to the estimation of the PDF of fatigue life. The proposed generalized fatigue life estimation approach accommodates both Gaussian and non-Gaussian processes for both narrow and wide band signals. To demonstrate its effectiveness, we use a duffing oscillator system and a practical example involving a truck assembly modeled by the Finite Element Method (FEM).
A Vibro-acoustic CAE Approach for Active Noise Control Prediction
(2024-01-01) Abbas, Ahmad A; Mourelatos, Zissimos P; Latcha, Michael; Yang, Lianxiang; Drignei, Dorin; Sturla, Francisco
This research focuses on a comprehensive analysis and prediction of Active Noise Cancellation (ANC) system performance in vehicles, with particular emphasis on the structural and acoustic aspects. While acknowledging the significance of electronic components and control algorithms in ANC systems, this study focuses on predicting the ANC performance using a full vibro-acoustic vehicle model. Additionally, the integration of noise management supplier control systems with CAE ANC models is explored. The development of a predictive CAE methodology is demonstrated using a road noise cancellation example. The process involves several steps including the structural behavior of the Trim Body in White (TBIW) structure, the development of a vehicle cavity model, the derivation of accurate speaker models, the assessment of speaker integration with vehicle doors, and the development of Transfer Paths (TP) from speakers to microphones. A novel methodology is presented to quantify the door stiffness requirements for optimal speaker ANC performance, incorporating substructuring methods and physical testing. The accuracy of the developed CAE models is validated using physical testing of circular and oval-shaped speakers integrated into vehicle doors and the calculation of transfer paths between each door speaker and microphone locations is demonstrated. The significance of microphone and speaker locations relative to driver or passenger ear positions highlights their influence on ANC performance. Finally, a controller is developed to test the CAE model and illustrate its functionality using a supplier’s controller for sound management. Overall, this research establishes a reliable vibro-acoustic CAE ANC model capable of predicting ANC system performance accurately by integrating a full vehicle vibro-acoustic model with an ANC controller. Such a predictive capability enables optimization and enhancement of vehicle performance for noise cancellation, unlocking its full potential in mitigating vehicle noise.
Contributions to Multivariate Data Science: Assessment and Identification of Multivariate Distributions and Supervised Learning for Groups of Objects
(2024-01-01) Tran, Nguyen Quynh Huong; Khattree, Ravindra; Drignei, Dorin; Li, Li; Roy, Anuradha; So, Hon Yiu
This dissertation considers three critical aspects of modern statistical analyses and machine learning, namely, (i) addressing the challenges posed by assessing the distributional assumptions of a univariate dataset, (ii) constructing graphical tests for the multivariate normality assumption, and (iii) exploring new algorithms for solving group classification problems, especially when the distance-based methods are not applicable. First, we focus on the development and evaluation of graphical statistical tests for univariate datasets, aiming to assess any specific distributional assumption. Examination of normality is given special emphasis. Recognizing the impact of outliers on the normality assumption, this study also incorporates outlier detection methodologies and provides some graphical tools for their identification. T4 plot is introduced as an additional effective tool for this purpose. Examination of multivariate normality is more challenging as in this case nonnormality may exhibit or mask itself in many different ways. In the second part, we thus emphasize graphical assessments of the multivariate normality assumption. This is done via MT3 and MT4 plots based on the derivatives of the cumulant generating function. The final segment of the dissertation shifts the discussion towards machine learning algorithms devised specifically for group classification problems. This involves the exploration of new methodologies that address the challenges inherent in classification and discrimination within complex datasets where other standard methods based on the classification of individual observation may not be very effective. For this we rely on the eigenstructures of the data. In the process, we also address the problem of dimensionality reduction. The problem of selection of copulas can be viewed as a corollary of the group classification problem. This has also been discussed in a separate chapter.
Experimental Implementation of a New Durability / Accelerated Life Testing Time Reduction Method
(2021-11-13) Baseski, Igor; Mourelatos, Zissimos P.; Latcha, Michael; Drignei, Dorin; Wang, Xia
Fatigue can be defined as a cyclic degradation process resulting in a failure at lower stress levels than the ultimate load. Fatigue reliability is defined as the probability that a structure will perform its intended function throughout its lifetime without any fatigue failure. Durability testing aims to predict fatigue damage in order to estimate the remaining useful life (RUL) based on fatigue. The latter is a useful metric in design for life-cycle cost. The objective of this research is to develop a new durability time reduction method to experimentally estimate the fatigue life of a vehicle component or system with accuracy using a short duration test. We assume that the loading random process (e.g. terrain configuration) is stationary and ergodic so that a single time trajectory can quantify the loading statistics. For the single time trajectory of the load process, we measure the corresponding output stress trajectory at a specified location on the structure. The latter is cycle counted using the 4-point rainflow counting algorithm. The cycle counting identifies all signal (stress) peaks and valleys using a peak picking algorithm and uses them to identify the range of all individual fatigue damage cycles and the time they occur based on a chosen fatigue damage model. Using this information (range of each cycle and the time it occurs), we build a synthetic signal exhibiting the same fatigue damage cycles in the sequence they occur in the actual stress signal. The sequence can be important in order to properly account for the cumulative damage accumulation. Finally, based on the fact that the cycle damage is independent of the time it occurs, we compress the synthetic signal so that its Power Spectral Density (PSD) does not exceed an upper limit dictated by the durability equipment. This proposed durability approach achieves therefore, the same cumulative damage with the original signal in a much shorter testing time. We demonstrate the new durability approach with two examples, and validate it experimentally using a commonly used Belgian block terrain excitation on the suspension coil spring of a military HMMWV (High Mobility Multi-purpose Wheeled Vehicle).
Generation of Internal Combustion Engine Maps and Spark Timing Profiles Using Metamodels
(2022-03-14) Tafreshi, Ali; Mourelatos, Zissimos; Sangeorzan, Brian; Drignei, Dorin; Maisonneuve, Jonathan
With the growth of computing technologies, many leading automotive companies tend to use simulation tools to reduce the number of actual engine testing for evaluating the performance of Internal Combustion (IC) engines. However, a high-fidelity engine model which is very complex and computationally demanding, is needed. In this dissertation, we present efficient and accurate metamodels to predict an engine fuel map and to also obtain the spark timing profile to generate a specified torque curve. Time-dependent Kriging metamodels using Singular Value Decomposition (SVD) and Nonlinear Autoregressive metamodels with Exogenous inputs (NARX) in conjunction with Neural Networks (NN) are developed and used. A sequential process was first developed to generate steady-state engine fuel maps using Kriging accounting for different engine characteristics at different operating conditions. The generated map predicts engine output parameters such as Brake Mean Effective Pressure (BMEP) and fuel flow rate. The Kriging metamodels are created sequentially to ensure acceptable accuracy with a small number of expensive engine simulations. Two optimization problems are solved for full load and part load conditions, respectively. We demonstrate that the estimated fuel map is of high accuracy compared to the actual map. The internal combustion engine is a source of unwanted vehicle vibration produced by engine mount forces which depend on the engine torque profile during a transient tip-in or tip-out maneuver. A methodology was also developed to obtain the desired engine torque profile to minimize the unwanted vibration by controlling a set of engine calibration parameters. A set of design coefficients defining a spark timing profile and the corresponding engine torque profiles are used to construct time-dependent metamodels using SVD and Kriging. The accuracy of the approach is demonstrated using GT-Power engine simulations. In addition, we developed a time-dependent NARX-NN metamodel to predict engine spark timing and cylinder pressure profiles corresponding to a desired torque profile. The NARX-NN metamodel predicts the spark timing accurately using a very small number of engine simulations.
Modeling Extreme Insurance Losses Using Transmutation and Copula
(2023-01-01) Addai, Solomon; Ogunyemi, Theophilus; Perla, Subbaiah; Shillor, Meir; Drignei, Dorin; So, Hon Yiu
In this dissertation, we apply transmutation to the theoretical work in insurance. From our extensive literature search, this seems to be a novel piece of work with regards to the transmutation, we particularly focus on the theoretical application of the exponential, Pareto and Weibull distributions. By shedding light on this unexplored area, our findings contribute valuable insights into the broader domain of insurance studies. We also do some exploratory work with regard to future research pursuit on a combined application of copula and transmutation to insurance data.
Optimal Cut-Points for Diagnostic Variables in Complex Surveys
(2023-01-01) Madi, Samar Adnan; Drignei, Dorin; Brown, Elise; Ogunyemi, Theophilus; Perla, Subbaiah; So, Hon Yiu
The ability to diagnose an individual is crucial in promoting treatment and improved health. However, finding a simple tool to base the diagnosis on can be complicated. This research will focus on developing statistical methodology for accurate diagnostic tests in the context of complex survey data. The proposed method will be illustrated with data from National Health and Nutrition Examination Survey (NHANES) to construct a diagnostic test to predict cardiometabolic disease risk in the US younger population. This research will begin with the exploration of a single diagnostic variable to be used as a diagnostic tool. The first 1-dimensional method explored uses receiver operating characteristic (ROC) curves for survey data as a means of determining an optimal cut-point for the diagnostic variable. This method is shown to be accurate but not conducive to multi-variable diagnostic tools using survey data. Another 1-dimensional method uses logistic regression for survey data to determine an optimal cut-point, using minimizing information criteria such as AIC to select the cut-point. The method is applied to NHANES data but considering a single diagnostic variable is shown to be too simplistic to create a comprehensive diagnostic tool. This method will then be extended to a multi-dimensional case, creating a diagnostic tool based on multiple variables using logistic regression for survey data. This method, although accurate, is shown to be time-consuming and computationally inefficient. A modified method using kriging-based optimization is proposed. Under this method, a more efficient search algorithm of efficient global optimization is explored, using a criterion of expected improvement. This proposed method is more computationally efficient in creating a multi-dimensional diagnostic tool. Application of these methods in a healthcare setting could be beneficial in promoting quick and easy diagnosis.
Point Estimators and Confidence Intervals Under Sequential Sampling Strategies with Applications
(2024-01-01) Alanazi, Ibtihal Hamoud; Hu, Jun; Drignei, Dorin; Perla, Subbaiah; Yiu So, Hon; Li, Li
Statistical inference is the process of making informed decisions about a larger population by analyzing a smaller group of data collected with some form of sampling. In many statistical inference problems, where some prescribed accuracy is desired, the required sample size often depends on unknown population parameters and thus remains unknown. Then, it is necessary to conduct a sequential sampling procedure, where an experimenter takes one observation at a time successively until a predefined stopping rule is satisfied. This thesis involves sequential sampling procedures dealing with three statistical inference problems. These are (i) bounded variance point estimation (BVPE) of a function of the scale parameter in a gamma distribution with known shape parameter; (ii) fixed-width confidence interval (FWCI) estimation for comparing two independent Bernoulli proportions; and (iii) fixed-accuracy confidence interval (FACI) estimation for the shape parameter of a Weibull distribution based on record data. In the first research problem, given a gamma population with known shape parameter α, we develop a general theory for estimating a function g(·) of the scale parameter β with bounded variance. We begin by defining a sequential sampling procedure with g(·) satisfying some desired condition in proposing the stopping rule, and show the procedure enjoys appealing asymptotic properties. After these general conditions, we substitute g(·) with specific functions including the gamma mean, the gamma variance, the gamma rate parameter, and a gamma survival probability as four possible illustrations. For each illustration, Monte Carlo simulations are carried out to justify the remarkable performance of our proposed sequential sampling procedure. This is further substantiated with a real data study on the weights of newborn babies. In the second research problem, we are interested in the proportions of a common characteristic possessed by two independent dichotomous populations, denoted by P1 and P2. We propose sequential sampling procedures for constructing FWCIs to compare the magnitude of P1 and P2 based on the log transformation and the logit transformation, respectively, which are followed by Monte Carlo simulations. We then implement these sequential sampling procedures to solve a real-world problem of mobile games A/B testing. In the third research problem, we focus on utilizing the record data to estimate the shape parameter of a two-parameter Weibull population, which is widely used in lifetime data analysis. A sequential sampling procedure is developed for constructing a FACI for the Weibull shape parameter β, no matter whether the scale parameter α is known or unknown.
Primary school children's health behaviors, attitudes, and body mass index after a 10-week lifestyle intervention with follow-up
(2018-05-09) Brown, Elise C.; Buchan, Duncan S.; Drignei, Dorin; Wyatt, Frank B.; Kilgore, Lon; Cavana, Jonathan; Baker, Julien S.
Background: Given the current global child obesity epidemic, testing the effectiveness of interventions in reducing obesity and its influencers is paramount. The purpose of this study was to determine immediate and long-term changes in body mass index and psychosocial variables following a 10-week lifestyle intervention. Methods: Seven hundred and seventy participants (8.75 ± 0.98 years of age, 379 boys and 391 girls) took part in the study. Participants had height, weight, and psychosocial questionnaires assessed at pre- and post-control, pre- and post-intervention, and 6-months post-intervention. Participants completed a weekly 10-week intervention consisting of healthy eating and physical activity education, physical activity, parental involvement, and behavior change techniques. Regression models were fit with correlated errors where the correlation occurred only between time points, not between subjects, and the nesting effects of school and area deprivation were controlled. Results: Regression models revealed a significant decrease in body mass index from pre- to post-intervention of 0.8512 kg/m2 (P = 0.0182). No Changes in body mass index occurred from post-intervention to 6-month follow-up (P = 0.5446). The psychosocial variables did not significantly change. Conclusions: This lifestyle intervention may be an effective means for improving body mass index in primary school children in the short-term if the duration of the intervention is increased, but these changes may not be sustained without on-going support.

Browsing by Author "Drignei, Dorin"

Results Per Page

Sort Options