Fault diagnosis based on adaptive observer for a class of non-linear systems with unknown parameters

In this paper, the fault diagnosis problem for a class of non-linear systems with uncertainty which depends on states, inputs and unknown constant parameters is discussed. Under some geometric conditions, the system is transformed into two different subsystems. One is not affected by actuator faults, so a non-linear adaptive observer can be designed based on the assumption of the strictly positive realness (SPR). The other whose states can be measured is affected by the faults. Actuator fault diagnosis is based on estimations of both the state and the unknown parameters with good accuracy. Discussions on release of SPR requirement and extension to the sensor fault case are also made. Finally, two examples are given in order to illustrate the applicability of the proposed methods for actuator fault diagnosis and sensor fault diagnosis respectively.     

Validation of a new self-report measure of parental attributions.

Attributional theory and empirical evidence suggest that a tendency to make stable, global self-causal attributions for undesirable events is associated with negative outcomes. However, existing self-report measures of parental attributions do not account for the possibility that dysfunctional parent-causal attributions for child misbehavior might be important predictors of poor family functioning. To address these concerns, the authors developed and tested a new measure of both parent-causal and child-responsible attributions for child misbehavior in a sample of 453 community couples. Structural validity, convergent validity, discriminant validity, internal consistency, and temporal stability of the new measure were examined. As expected, confirmatory factor analysis resulted in 2 factors, Child-Responsible (9 items) and Parent-Causal (7 items); the final model was cross-validated in a holdout sample. The final scale demonstrated adequate internal consistency (αs = .81–.90), test–retest reliability (rs = .55–.76), and convergent and discriminant validity. Dysfunctional parent-causal and child-responsible attributions significantly predicted parental emotional problems, ineffective discipline, parent–child physical aggression, and low parenting satisfaction. Associations with parent–child aggression and parenting satisfaction were generally larger than with partner aggression and relationship satisfaction. (PsycINFO Database Record (c) 2010 APA, all rights reserved)     

Forming Nanocrystalline Structures in Metal Particle Impact

Grain refinement by plastic deformation is becoming increasingly popular as a way of producing metals with improved properties, such as higher mechanical strength. Surface treatment techniques in which a metallic substrate is bombarded with metallic particles can generate nanocrystalline layers in the impact zone. Understanding the physical mechanisms underlying this grain refinement is crucial for achieving an improvement of existing experimental processes. In this article, we propose a numerical framework combining finite element (FE) simulations with a dislocation-based material model to predict the evolution of the microstructure under particle impact. A single particle normally impacting on a metallic substrate was simulated at different initial velocities. The simulations were compared with previously reported numerical and experimental data. The results indicate that our model accurately captures the grain refinement in the impact zone for a broad range of velocities. This approach provides valuable information on the formation of nanocrystalline layers in both the substrate and the impacting particle. Its potential applications include processes involving surface treatment by high velocity particles, such as shot peening, surface mechanical attrition treatment, kinetic metallization, cold spray, etc.     

Modeling heterogeneous bacterial populations exposed to antibiotics: The logistic‐dynamics case

In typical in vitro tests for clinical use or development of antibiotics, samples from a bacterial population are exposed to an antibiotic at various concentrations. The resulting data can then be used to build a mathematical model suitable for dosing regimen design or for further development. For bacterial populations that include resistant subpopulations—an issue that has reached alarming proportions—building such a model is challenging. In prior work, we developed a related modeling framework for such heterogeneous bacterial populations following linear dynamics when exposed to an antibiotic. We extend this framework to the case of logistic dynamics, common among strongly resistant bacterial strains. Explicit formulas are developed that can be easily used in parameter estimation and subsequent dosing regimen design under realistic pharmacokinetic conditions. A case study using experimental data from the effect of an antibiotic on a gram‐negative bacterial population exemplifies the usefulness of the proposed approach. © 2015 American Institute of Chemical Engineers AIChE J, 61: 2385–2393, 2015     

Bridging cell multiscale modeling of fatigue crack growth in fcc crystals

The previously developed bridging cell method for modeling coupled continuum/atomistic systems at finite temperature is used to model fatigue crack growth in single crystal nickel under two crystal orientations at different temperatures. The method is expanded to implement a temperature‐dependent embedded atom method potential for finite temperature simulations avoiding time‐scale restrictions associated with small timesteps. Results for the fatigue simulation were compared with respect to deformation behavior, stress distribution, and crack length. Results showed very different crack growth mechanisms between the two crystal orientations as well as reduced resistance to crack growth with increased temperature. Copyright © 2015 John Wiley & Sons, Ltd.     

Issues with Combining Road Elevation Spectral Models and Vehicle Vibration Response to Estimate Vehicle Dynamic Characteristics

The research presented in this paper focuses on issues associated with the development of an experimental technique to estimate the dynamic characteristics of wheeled vehicles (namely, the frequency response function) using only in‐service response data. To validate the approach and eliminate complexities associated with multi‐wheel vehicles, a single‐wheeled prototype vehicle was designed and commissioned. The vertical vibration acceleration of the prototype vehicle's sprung mass was measured during normal operation. The power spectral density function was computed and used to estimate the frequency response function of the vehicle. A number of experiments using various configurations of the single‐wheeled prototype vehicle were undertaken, along with a series of vibration table experiments to provide a comparison with the estimated frequency response functions. The results show that the best estimate of the frequency response function using the vehicle response data provides reasonable agreement with the measured laboratory experiments when the value of the slope of the spectral function is not set to the value suggested by the International Organisation for Standardisation. Another technique was further developed to estimate the value of the pavement spectral slope using only in‐service response data; however, this technique does not yield consistent and accurate estimates. Interestingly, the main resonance of the vehicle is in agreement between the vibration table and response data around the sprung mass of all three vehicle configurations when inspected using linear scales (regardless of the variation in the spectral shape of the excitation), although the additional modes (including the unsprung mass) differs for all vehicles. Copyright © 2014 John Wiley & Sons, Ltd.     

Microbiome of free-living amoebae isolated from drinking water

Free-living amoebae (FLA) are protozoa that can be found in water networks where they prey on bacteria within biofilms. Most bacteria are digested rapidly by phagocytosis, however some are able to survive within amoebae and some are even able to multiply, as it is the case for Legionella pneumophila. These resisting bacteria are a potential health problem as they could also resist to macrophage phagocytosis. Several publications already reported intra-amoebal bacteria but the methods of identification did not allow metagenomic analysis and are partly based on co-culture with one selected amoebal strain. The aim of our study was to conduct a rRNA-targeted metagenomic analysis on amoebae and intra-amoebal bacteria found in drinking water network, to provide the first FLA microbiome in environmental strains. Three sites of a water network were sampled during four months. Culturable FLA were isolated and total DNA was prepared, allowing purification of both amoebal and bacterial DNA. Metagenomic studies were then conducted through 18S or 16S amplicons sequencing. Hartmannella was by far the most represented genus of FLA. Regarding intra-amoebal bacteria, 54 genera were identified, among which 21 were newly described intra-amoebal bacteria, underlying the power of our approach. There were high differences in bacterial diversity between the three sites. Several genera were highly represented and/or found at least in two sites, underlying that these bacteria could be able to multiply within FLA. Our method is therefore useful to identify FLA microbiome and could be applied to other networks to have a more comprehensive view of intra-amoebal diversity.