An integrated database and expert system for failure mechanism identification: Part II—the system and performance testing

An integrated database and expert system has been developed for assisting the human analyst in identifying the failure mechanism of mechanical components. The system comprises six major modules: database and management system; case maintenance; knowledge acquisition and editing; expert system; explanation and test-recommendation facilities, and user interface. The knowledge acquisition and editing module has been detailed in Part I of a two-part paper [6]. The overall system and the details of other modules are described here. The system has been implemented using the M4 expert system shell, Microsoft Access database software, and Visual Basic. Illustrative examples are used to show the capability of the system. Also reported are the results obtained from four different tests to determine the system performance.     

Hybrid expert system for the failure analysis of mechanical elements

An expert system for the failure analysis of metallic mechanical components is described, with emphasis on the structure and the solution strategy followed by the software. The system treats failure in common mechanical elements and is aimed to guide non-experts through the process of basic failure analysis. The system acts as a “virtual colleague”, providing guidance and experience during failure studies. As such, it allows the users to gradually increase their experience while using the software in distinct applications. The system works interactively through a system of structured questions. The answers provided by the user permit the software to direct the analysis toward its solution. Decisions are taken in a hybrid way, through case based reasoning and rule based reasoning, the former using a database of solved cases, the latter mimicking the way experts treat failure analysis. The paper describes the development method, the logic behind the operation of the system and the interaction between case based reasoning and rule based reasoning modules. Its performance in four real-world failure cases is used to demonstrate its feasibility and usefulness.     

用C++ Builder开发污水处理故障诊断专家数据库系统

根据实际项目的开发，介绍了一个针对污水处理13常运行的故障诊断专家系统的设计和实现。系统运用了人工智能与数据库相结合的技术，利用Borland C Builder的强大数据库技术，成功地开发出了一个人机界面良好、易于用户和专家使用的专家系统。     

A performance evaluation of three inference engines as expert systems for failure mode identification in shafts

This paper aims to present performance evaluation of three different inference engines (rule based reasoning, fuzzy based reasoning and Bayesian based reasoning) for failure mode identification in shafts. This research was done with a focus on the validation cases and results after their use in failure cases from several industries where the three systems were tested under the same conditions.Each system was implemented using the same user interface and knowledge base, with different frameworks and techniques as follows: rule based inference reasoning (prolog, C#), Mamdani-fuzzy based reasoning (C, MATLAB®) and Bayesian based reasoning with a variable elimination algorithm (C, MATLAB®).The best performance was obtained using the Bayesian inference engine. The conditional probabilities give flexibility when evidence is not listed, while the fuzzy and classical IF-THEN systems depend on the rules in the inference engine.The process presented in this paper could be used for validation of any expert system or for comparison with other expert systems (inference engines) when the knowledge base is the same.     

An artificial neural network-based expert system for the appraisal of two-car crash accidents

This paper employs artificial neural network (ANN) to develop an accident appraisal expert system. Two ANN models-- party-based and case-based-- with different hidden neurons are trained and validated by k-fold (k=3) cross validation method. A total of 537 two-car crash accidents (1074 parties involved) are randomly and equally divided into three subsets. For the comparison, a discrimination analysis (DA) model is also calibrated. The results show that the ANN model can achieve a high correctness rate of 85.72% in training and 77.91% in validation and a low Schwarz's Bayesian information criterion (SBC) of -0.82 in training and 0.13 in validation, which indicates that the ANN model is suitable for accident appraisal. Furthermore, in order to measure the importance of each explanatory variable, a general influence (GI) index is computed based on the trained weights of ANN. It is found that the most influential variable is right-of-way, followed by location and alcoholic use. This finding concurs with the prior knowledge in accident appraisal. Thus, for the fair assessment of accident liabilities the correctness of these three key variables is of critical importance to police investigation reports.     

Triple-scale failure estimation for a composite-reinforced structure based on integrated modeling approaches. Part 1: Microscopic scale analysis

The structural reliability of a composite component locally reinforced with a fibrous metal matrix composite is essentially affected by the micro-scale failures. The micro-scale failures such as fiber fracture or matrix damage are directly governed by the internal stress states such as mismatch thermal stress. A proper computational method is needed in order to obtain micro-scale stress data for arbitrary thermo-mechanical loads. In this work a computational scheme of microscale failure analysis is presented for a composite component. Micromechanics-based triple-scale FEM was developed using composite laminate element. The considered composite component was a plasma-facing component of fusion reactors consisting of a tungsten block and a composite cooling tube. The micro-scale stress and strain data were estimated for a fusion-relevant heat flux load. Ductile damage of the matrix was estimated by means of a damage indicator. It was shown that the risk of micro-scale composite failure was bounded below an acceptable level.     

Triple-scale failure estimation for a composite-reinforced structure based on integrated modeling approaches. Part 2: Meso- and macroscopic scale analysis

In this article, the meso- and macroscopic failure features are discussed considering the identical composite component as in the foregoing article. In the meso-scale failure analysis, the risk of plastic instability of the composite tube was estimated considering the shakedown boundary as a failure criterion. The meso-scale stresses of the composite tube were computed using micromechanical homogenization and compared with the shakedown boundary of the composite obtained from the direct shakedown analysis. The stress states were close to the shakedown boundary indicating no critical danger of plastic failure. In the macro-scale failure analysis, the mechanical influence of the local composite integration was investigated with regard to the brittle failure risk of the neutron-embrittled component. To this end, a probabilistic failure analysis code was applied which was based on the fracture mechanics and the weakest-link failure theory. Various fracture criteria were considered. It was found that the failure risk of the tungsten block was strongly reduced by the composite reinforcement of the tube due to the intensification of compressive stress fields.     

A random critical-core theory of microdamage in interply hybrid composites: I—First failure and hybrid effect

This paper presents a statistical analysis for the tensile first failure and hybrid effect (which is part of the synergistic effect) of unidirectional interply hybrid composites. A shear-lag model is applied to solve the stress redistributions after the breakage of low-elongation fibers. By applying the random critical core model, the probabilistic first failure strain of hybrids is evaluated first, and the hybrid effect for strains is consequently evaluated. Reduction of the stress concentration factor and ineffective length of the low-elongation fibers in hybrids and differences between microdamage mechanisms in hybrids and single fiber composites provide a theoretical explanation of the hybrid effect. The present theoretical results show reasonable agreement with the existing experimental results.     

Uncertainties and quantification of common cause failure rates and probabilities for system analyses

Simultaneous failures of multiple components due to common causes at random times are modelled by constant multiple-failure rates. A procedure is described for quantification of common cause failure (CCF) basic event probabilities for system models using plant-specific and multiple-plant failure-event data. Methodology is presented for estimating CCF-rates from event data contaminated with assessment uncertainties. Generalised impact vectors determine the moments for the rates of individual systems or plants. These moments determine the effective numbers of events and observation times to be input to a Bayesian formalism to obtain plant-specific posterior CCF-rates. The rates are used to determine plant-specific common cause event probabilities for the basic events of explicit fault tree models depending on test intervals, test schedules and repair policies. Three methods are presented to determine these probabilities such that the correct time-average system unavailability can be obtained with single fault tree quantification. Recommended numerical values are given and examples illustrate different aspects of the methodology.     

Quality risk analysis in a cGMP environment: multiple models for comprehensive failure mode identification during the computer system lifecycle

Pharmaceutical quality systems use various inputs to ensure product quality and prevent failures that might have patient consequences. These inputs are generally data from failures that have already occurred, for example process deviations or customer complaints. Risk analysis techniques are well-established in certain other industries and have become of interest to pharmaceutical manufacturers because they allow potential quality failures to be predicted and mitigating action taken in advance of their occurring. Failure mode and effects analysis (FMEA) is one such technique, and in this study it was applied to implement a computerized manufacturing execution system in a pharmaceutical manufacturing environment. After introduction, the system was monitored to detect failures that did occur and these were analyzed to determine why the risk analysis method failed to predict them. Application of FMEA in other industries has identified weaknesses in predicting certain error types, specifically its dependence on other techniques to model risk situations and its poor analysis of non-hardware risks, such as human error, and this was confirmed in this study. Hierarchical holographic modeling (HHM), a technique for identifying risk scenarios in wide-scope analyses, was applied subsequently and identified additional potential failure modes. The technique for human error rate prediction (THERP) has previously been used for the quantitative analysis of human error risk and the event tree from this technique was adapted and identified further human error scenarios. These were input to the FMEA for prioritization and mitigation, thereby strengthening the risk analysis in terms of failure modes considered.     

A hybrid PET‐MRI: An integrated molecular‐genetic imaging system with HRRT‐PET and 7.0‐T MRI

A PET‐MRI fusion system is developed for molecular‐genetic imaging. The purpose of the system is to obtain images of the in‐vivo human brain using two high‐end imaging devices, an advanced PET and an ultrahigh‐field MRI. These are the HRRT‐PET, a high‐resolution research tomograph dedicated to brain imaging on the molecular level, and the 7.0‐T MRI, an ultrahigh field version used for morphological imaging. HRRT‐PET delivers high‐resolution molecular imaging with a resolution down to 2.5 mm FWHM, which is currently the highest spatial resolution available for the observation of the human brain's molecular activities, including enzymes and receptors, which are manipulated genetically, such as reporter genes and probes. The 7.0‐T MRI began to reveal submillimeter resolution images of the cortical as well as deep brain areas, down to 250 μm, which allows us to visualize the fine details of the cortical and brainstem areas, including the line of Gennari in the visual cortex and the corticospinal tracts in the pontine area. The current PET‐MRI fusion imaging system produces the highest quality images of molecular and genetic activities of the human brain in vivo. It is starting to provide many answers to the key questions about the neurological diseases. Some of these start providing answers to many cognitive neuroscience problems with clear molecular and genetic bases. There is great potential in the PET‐MRI for early diagnosis of cancers as well as other neurological diseases, which we were previously unable to diagnose, such as microscopic molecular changes that occur in Parkinson's and Alzheimer's diseases. © 2007 Wiley Periodicals, Inc. Int J Imaging Syst Technol, 17, 252–265, 2007     

Different insights for improving part and system reliability obtained from exactly same DFOM “failure numbers”

Techniques for improving the reliability and maintainability of both nonrepairable and repairable items can be suggested by failure data analysis. It is shown that a given set of failure numbers leads to very different improvement strategies when the numbers are the times-between-successive-failures of one or more repairable items, rather than the times-to-failure of nonrepairable items. Since this should have been obvious more than 50 years ago, at the onset of formal reliability engineering activities, several reasons are proffered for the widespread and protracted misinterpretation of even the most basic—and simple!—conceptual and practical differences between nonrepairable and repairable items.     

3D‐based hp‐adaptive first‐order shell finite element for modelling and analysis of complex structures—Part 2: Application to structural analysis

The paper presents a 3D‐based adaptive first‐order shell finite element to be applied to hierarchical modelling and adaptive analysis of complex structures. The main feature of the element is that it is equipped with 3D degrees of freedom, while its mechanical model corresponds to classical first‐order shell theory. Other useful features of the element are its modelling and adaptive capabilities. The element is assigned to hierarchical modelling and hpq‐adaptive analysis of shell parts of complex structures consisting of solid, thick‐ and thin‐shell parts, as well as of transition zones, where h, p and q denote the mesh density parameter and the longitudinal and transverse orders of approximation, respectively. The proposed hp‐adaptive first‐order shell element can be joined with 3D‐based hpq‐adaptive hierarchical shell elements or 3D hpp‐adaptive solid elements by means of the family of 3D‐based hpq/hp‐ or hpp/hp‐adaptive transition elements. The main objective of the first part of our research, presented in the first part of the paper, was to provide non‐standard information on the original parts of the element algorithm. Here we describe the second part of the research, devoted to the methodology and results of the application of the element to various plate and shell problems. The main objective of this part is to verify algorithms of the element and to show its usefulness in modelling and adaptive analysis of shell and plate parts of complex structures. In order to do that, there is a presentation of the results of a comparative analysis of model plate and shell problems using the classical and our elements, and equidistributed and integrated Legendre shape functions. For the plate problem a comparison of the results obtained from the adaptive and non‐adaptive analysis is also included. Additionally, some advantages of the application of our element are shown through a comparative analysis of p‐convergence of the thin plate problem and an adaptive analysis of the exemplary complex structure. Copyright © 2006 John Wiley & Sons, Ltd.     

A refined diffuse cohesive approach for the failure analysis in quasibrittle materials—part II: Application to plain and reinforced concrete structures

This work presents the numerical application of the diffuse cohesive interface model introduced in the Part I paper to the failure analysis of plain and reinforced concrete structures, subjected to complex loading conditions, inducing mixed‐mode fracture initiation and propagation. With the aim of capturing the interaction between concrete and steel reinforcements, the adopted fracture model is incorporated in a novel, more general numerical framework for the nonlinear analysis of reinforced concrete structures. Such a framework includes a newly proposed embedded truss model for the reinforcing bars, allowing them to be crossed by the neighboring propagating cracks. Comparisons with available experimental results are provided, assessing the reliability and the numerical accuracy of the proposed concrete model, with reference to plain specimens subjected to single‐crack propagation as well as to reinforced elements subjected to multiple cracking.     

Extending LaMer's mechanism using open system for increasing forced and controlled growth of Au nano particles: Desired decreasing Fe3O4 nano particles size during simultaneous synthesis in optimized conditions

The most effective parameters were found to obtain Au/Fe₃O₄ nano particles (NPs)-oleylamine composite. Having Au NPs with the controlled maximum mean size under the forced conditions was the main aim of this study. We used the continuous flow rates of oleylamine 75% to produce Au NPs under an open system by extended LaMer mechanisms. This process decreased the mean size of Fe₃O₄ NPs synthesized simultaneously, by classic LaMer mechanism. The Fe₃O₄ NPs production was carried out without continuous adding of any iron reactant, viz. as a closed system. In the absence of gold ions, the mean size of the synthesized Fe₃O₄ NPs using 2.5 ml/min oleylamine was about 35.0 nm at 2.0 ± 0.5 °C after 120 min. This mean size was decreased to 27.2, 21.4, 16.8 and 8.7 nm, when Au NPs were simultaneous prepared using 0.5, 0.75, 1.5 and 2.5 ml/min of oleylamine, respectively, at the same conditions. Surface Plasmon Resonance (SPR) adsorption was used to evaluate Au NPs production at first 30 min, while Small Angle X-ray Scattering (SAXS) method was used to monitor the reaction progression for near-real time analysis of increasing the growth of Au NPs up to 280 min, at the optimum conditions. Changing the properties of Fe₃O₄ NPs during processes was determined by studying Magnetization, Potentiometric titration, Inductive heating and Zeta potential.