A Multiagent Framework for a Diagnostic and Prognostic System Irtaza Barlas 124 Pages Directed By: Dr. George Vactsevanos The shortcomings of the current diagnostic and prognostic systems stem from the limitations of their frameworks. The framework is typically designed on the passive, open loop, static, and isolated notions of diagnostics, in that the framework does not observe its diagnostic results (open-looped), hence can not improve its performance (static). Its passivity is attributed to the fact that an external event triggers the diagnostic or prognostic action. There is also no effort in place to team-up the diagnostic systems for a collective learning, hence the implementation is isolated. In this research we extend the current approaches of the design and implementation of diagnostic and prognostic systems by presenting a framework based upon Multiagent systems. This research created novel architectures by providing such unique features to the framework, as learning, reasoning, and coordination. As the primary focus of the research the concept of Case-Based Reasoning was exploited to reason in the temporal domain to generate better prognosis, and improve the accuracy of detection as well as prediction. It was shown that the dynamic behavior of the intelligent agent helps it to learn over time, resulting in improved performance. An analysis is presented to show that a coordinated effort to diagnose also makes sense in uncertain situations when there are certain number of systems attempting to communicate certain number of failures, since there can be high probability of finding a shareable experience.

Masters of Science / This study is focused on the Coal Bed Methane resources of Botswana with specific reference to the Central Kalahari basin where prospect license blocks forming the focus of this study are located. The aim of this study is to evaluate the fracture network in the coal seams and the fracture systems in the surrounding coal bearing sedimentary sequences and their contribution to dynamic flow. Coal bed methane sources are dual-porosity media documented on the natural fracture network, seen as micropores (matrix/natural fractures) and macropores (cleat). The coals of this region belong to the Ecca Group’s Morupule Fm (Permian) (70 m), focus of this study and have been preserved in the extensive Karoo basin within the Southern Africa region. Fractures can easily be identified in Acoustic Televiewer logs (ATV) and their orientation and structural character interpreted by rose plots, tadpoles and stick dip plots. In-situ stress fields have been determined from breakout structural evaluation and maintains a general E-W dip direction and N-S strike, thus most fractures are orientated optimally with inferred in-situ stress and enhancing flow potential in pore systems. A qualitative (MID plots & M-N cross-plots) and quantitative description of the fracture system is fundamental to the petrophysical evaluation, and involves the estimation of fracture parameters (fracture porosity, resistivity fracture index and both horizontal and vertical fracture indices).

The nation is challenged with improving the mathematics achievement of its students. No Child Left Behind holds schools, districts, and states accountable for improving student achievement. Because high stakes tests are given at the end of the school year, schools are presented with the challenging task of developing or purchasing reliable assessments that provide accurate information describing how well students understand the skills that will be measured on the end-of-the-year high stakes tests. Curriculum based measurements are used periodically to measure student progress toward meeting objectives during the school year. The problem exists that schools are utilizing limited resources of time and money on a tool with little evidence of effectiveness in increasing mathematics scores on state assessments. The purpose of this study was to determine the relationships between the scores of the three assessments of the Mathematics-Curriculum Based Measures (M-CBM) and the scores from the mathematics MCT2 for sixth-grade students. A correlational research design is used to fulfill the purpose and test the three null hypotheses. Hypothesis 1 for this study states that there is no relationship between the August 2009 M-CBM scores and the May 2010 mathematics MCT2 scores. Hypothesis 2 states that there is no relationship between the December 2009 M-CBM scores and the May 2010 mathematics MCT2 scores. Hypothesis 3 states that there is no relationship between the April 2010 M-CBM scores and the May 2010 mathematics MCT2 scores. To test the three null hypotheses, correlation coefficients were computed using the Pearson r. The results from all three hypotheses indicated that there were moderate positive correlations between scores of the M-CBM and scores of the mathematics portion of the MCT2, with the strongest relationship being between the April M-CBM and the mathematics MCT2 scores. Further analysis was done to determine if the relationship between M-CBM and mathematics MCT2 scores continued to exist when examined by mathematics MCT2 proficiency levels. The results of this analysis indicated that relationships between M-CBM and the mathematics portion of the MCT2 scores only existed for students scoring in the proficient range.

Curriculum-based measurement in writing (CBM-W) has been proposed as a means of screening general education and special education students for writing difficulties and as a tool for monitoring the progress of struggling writers. CBM-W involves the administration of multiple probes of equivalent difficulty over time to monitor student progress toward academic standards (McMaster & Espin, 2007). The purpose of this study was to determine the technical adequacy of multiple CBM-W measures and how well the measures predicted writing performance. Additionally, this study examined how well CBM-W measures predicted student growth in writing over time. This study also extended the work of previous research by including a sample of 5th grade Hispanic students (n = 167), including English language learners (ELLs), former ELLs who were being monitored, ELLs with disabilities, and non-ELLs. Students were given story starters and completed narrative writing samples that were scored using words written (WW), words spelled correctly (WSC), correct word sequences (CWS), and correct minus incorrect word sequences (CIWS). The criterion measure was the Test of Written Language, Fourth Edition (TOWL-4). Results indicated inconsistent alternate-forms reliability for WW and WSC, and high alternate-forms reliability for CWS and CIWS. CWS and CIWS were the best predictors of one-time writing performance on the TOWL-4 for the overall sample, ELLs, former ELLs on first year monitor status, and ELLs with disabilities. CBM-W was not a significant predictor of student growth over time for most scoring procedures. However, a promising finding is that the scoring procedure of CWS was a significant predictor of student growth for the overall sample and for former ELLs on first year monitor status. Limitations, practical implications and future research will be discussed. / text

Because of a greater need to focus on early achievement in mathematics and increased global competition in this area, emphasis is increasingly being placed upon screening and monitoring of math development in early elementary to ensure progress for all students. Although protocols for early literacy assessment exist, research of psychometrically sound and instructionally useful early numeracy assessment methods is still developing. Thus, the current study was undertaken to further develop and examine early numeracy measures. In a two-year investigation, 72 kindergarten and 57 first-grade students completed a set of existing and newly developed early numeracy curriculum-based measurement (EN-CBM) tools. Reliability and validity of these measures were examined both within and across years. The nature of score change across the academic year was also analyzed. Findings were generally similar to those published in previous EN-CBM research. Alternate-form reliability estimates were largely adequate for screening purposes. Although concurrent and predictive validity estimates varied as a function of screening period and the outcome criterion, these estimates were generally moderate to strong. EN-CBM discriminant validity evidence was mixed for the kindergarten measures but more evident for first-grade tools. All measures demonstrated score change that was statistically significant over time with at least moderate to large effect sizes for most measures. Implications for EN-CBM use as screening tools are discussed. Study limitations are presented as are suggestions for future EN-CBM research.

New environmental regulations as well as the increasing industrial competitiveness have set new more demanding rules on the manufacturing industry. In order to abide by those rules not only from the legal point of view but also to be able survive, manufacturing has to be more sustainable from many aspects, especially the economical one. One way to achieve the previous target is an unfortunately often oversighted aspect of the industry sector, the maintenance strategy. Condition based maintenance, CBM, can be used successfully in the industry and accurate estimation of spindle life time can lead to large savings in downtime and cost. CBM requires accurate sensors and equipment in order to get the right indicators whether equipment performance is deteriorating or not. One performance factor when planning a machining process is chatter vibration and one way to avoid this deteriorating phenomenon is to choose cutting parameters that allow stable machining. Various types of sensors are available for vibration and other CBM related measurements. Depending on the situation, the most applicable sensor is selected. The core of this thesis is to investigate the usefulness of measurements with the contactless excitations and response unit in terms of condition based maintenance. In the first part of the thesis, some of the theoretical aspects of maintenance are extensively elaborated upon and later on, the experimental part is presented along with the results’ discussion. The hardware required by the experiments has been provided by KTH and the experiments took place in two of an automotive industry’s production sites. There have been two visits at site A and one at site B, apart from the initial meetings. The measurements have been analyzed with the use of MATLAB. / Nya miljöregler samt ökande industriell konkurrens har satt nya mer krävande regler för tillverkningsindustrin. För att följa dessa regler, inte bara ur rättslig synpunkt utan också för att kunna överleva, behöver tillverkningen ske mer hållbar ur många aspekter, särskilt den ekonomiska. Ett sätt att uppnå målen är via, en tyvärr ofta underskattad metod, underhållsstrategin. Tillståndsbaserat underhåll, CBM, kan användas med framgång inom branschen och korrekt uppskattning av spindellivstid kan leda till stora besparingar genom minskade driftstopp och kostnader. CBM kräver noggranna sensorer och utrustning för att få rätt indikatorer för att avgöra om utrustningens prestanda försämras eller ej. En prestationsavgörande faktor vid planering av bearbetningsprocesser är vibrationer. Ett sätt att undvika dessa försämrade fenomen är att välja skärparametrar som tillåter stabil bearbetning. Olika typer av sensorer finns tillgängliga för vibrations- och andra CBM-relaterade mätningar. Beroende på situation, väljs den mest lämpliga sensorn. Kärnan i denna rapport är att undersöka nyttan av mätningar med en beröringsfri excitations- och mätenhet för tillståndsbaserat underhåll. I de första avsnitten av rapporten redogörs några av de teoretiska aspekterna av underhåll och i de senare är den experimentella delen presenterad, tillsammans med diskussion kring resultat. Hårdvaran som krävs för experimenten har tillhandahållits av KTH och experimenten ägde rum på två produktionsanläggningar hos en fordonstillverkare. Det har varit två besök på plats A och ett besök på plats B, bortsett från inledande möten. Mätningarna har analyserats med hjälp av MATLAB.

Vibrationsmätning är inget nytt fenomen men många företag anser sig ändå skeptiskatill den abstrakta metoden. Samma skepticism finns inte vid ett läkarbesök där doktornlyssnar efter oljud med sitt stetoskop, ändå är likheterna slående. Med dagens teknikär det möjligt att diagnostisera en maskins olika beståndsdelar efter skador genom attanalysera det unika vibrationsmönster de bidrar med. Detta arbete syftar att minska denna barriär med hjälp av en introduktion till de olikabegrepp och val som finns inom vibrationsteknisk analys som hjälpmedel för underhåll.Slutligen presenteras en implementeringsmall där ekonomi och tillgänglighet anses varamest vitala inför framtida beslut. / Even though vibration measurement is not a new phenomenon, many businesses stillremain sceptical towards the somewhat abstract method. Same scepticism does not existwhen you visit the doctor and he uses his stethoscope to listen after abnormalities inthe sound pattern, even though the similarities are striking. With todays technology it ispossible to diagnose a machine’s different parts in search for damages by analysing theunique vibrationpatterns they provide. This study aims to close this gap with the help off an introduction to the different termsand choices that exist in vibration analysis as a tool in preventive maintenance and finallyto present guidelines for implementation where economy and availability is consideredmost pivotal for future decisions.

Carbon dioxide (CO2) is a primary source of greenhouse gases. Injection of CO2 from power plants near coalbed reservoirs is a win-win method to reducing emissions of CO2 to the atmosphere. Limited studies have investigated CO2 sequestration and enhanced coalbed methane production in San Juan and Alberta basins, but reservoir modeling is needed to assess the potential of the Black Warrior basin. Alabama ranks 9th nationally in CO2 emissions from power plants; two electricity generation plants are adjacent to the Black Warrior coalbed methane fairway. This research project was a reservoir simulation study designed to evaluate the potential for CO2 sequestration and enhanced coalbed methane (ECBM) recovery in the Blue Creek Field of Black Warrior basin, Alabama. It considered the injection and production rate, the components of injected gas, coal dewatering, permeability anisotropy, various CO2 soak times, completion of multiple reservoir layers and pressure constraints at the injector and producer. The simulation study was based on a 5-spot well pattern 40-ac well spacing. Injection of 100 percent CO2 in coal seams resulted in average volumes of 0.57 Bcf of sequestered CO2 and average volumes of 0.2 Bcf of enhance methane production for the Mary Lee coal zone only, from an 80-acre 5-spot well pattern. For the entire Blue Creek field of the Black Warrior basin, if 100 percent CO2 is injected in the Pratt, Mary Lee and Black Creek coal zones, enhance methane resources recovered are estimated to be 0.3 Tcf, with a potential CO2sequestration capacity of 0.88 Tcf. The methane recovery factor is estimated to be 68.8 percent, if the three coal zones are completed but produced one by one. Approximately 700 wells may be needed in the field. For multi-layers completed wells, the permeability and pressure are important in determining the breakthrough time, methane produced and CO2 injected. Dewatering and soaking do not benefit the CO2 sequestration process but allow higher injection rates. Permeability anisotropy affects CO2 injection and enhanced methane recovery volumes of the field. I recommend a 5-spot pilot project with the maximum well BHP of 1,000 psi at the injector, minimum well BHP of 500 psi at the producer, maximum injection rate of 70 Mscf/D, and production rate of 35 Mscf/D. These technical results, with further economic evaluation, could generate significant projects for CO2 sequestration and enhance coalbed methane production in Blue Creek field, Black Warrior Basin, Alabama.

This research develops an equipment failure prognostics model to predict the equipment’s chance of survival, using LAD. LAD benefits from not relying on any statistical theory, which enables it to overcome the problems concerning the statistical properties of the datasets. Its main advantage is its straightforward process and self-explanatory results. Herein, our main objective is to develop models to calculate equipment’s survival probability at a certain future moment, using LAD. We employ the LAD’s pattern generation procedure. Then, we introduce a guideline to employ generated patterns to estimate the equipment’s survival probability. The models are applied on a condition monitoring dataset. Performance analysis reveals that they provide comprehensible results that are greatly beneficial to maintenance practitioners. Results are compared with PHM’s results. The comparison reveals that the LAD models compare favorably to the PHM. Since they are at their beginning phase, some future directions are presented to improve their performances.

Industry 4.0 push forward the development of concepts such as artificial intelligence, big data, and Industrial Internet of Things, which claims an evolution of the monitoring systems design in terms of the accessibility to the information. In this project, the author describes the design of a condition monitoring system to monitor the state of different components of an extrusion line and propose a system that allows predictive maintenance in industry, specifically an extrusion system. In this framework, Condition Based Maintenance, CBM, the health state of the component is continuously monitored. In some cases, the monitoring can be periodical. The goal is to make repairs in the opportune moment, by receiving data of malfunction, so the efficiency is maxed. The aim is to develop a system to monitor the state of different components of an extrusion line and to propose a system that allows predictive maintenance in industry and that the project can be used as a guideline to a complete condition monitoring system implementation in an industrial environment. To be able to achieve this, some first steps must be accomplished. These are a Taxonomy, or a breakdown of the system into individual elements and how they are related; and carrying out a Failure Modes Effect and Criticality Analysis, known as FMECA. With these studies, the author displays the failure modes that are critical for the operating of the system and thus, which have the most likelihood to occur while having a big impact. From the information extracted the author presents a model based in accelerometer, temperature and MCSA (Motor Current Spectral Analysis) sensors. Furthermore, the data obtained will need to be analysed. With that in mind, the operating frequencies as well as the failure modes frequencies must be studied, which will allow correct identification while analysing the data. This analysis will be done by characterizing data and applying analysing techniques as FFT or Hilbert.