Nuclear power plants are inherently complex systems. While the technology has been used to generate electrical power for many decades, process monitoring continuously evolves. There is always room for improvement in terms of maximizing the availability by reducing the risks of problems and errors. In this context, automated monitoring systems have become important tools – not least with the rapid progress being made in the field of data analytics thanks to ever increasing amounts of processing power. There are many different types of models that can be utilized for identifying anomalies. Some rely on physical properties and theoretical relations, while others rely more on the patterns of historical data. In this thesis, a data-driven approach using a hierarchical autoencoder framework has been developed for the purposes of anomaly detection at the Swedish nuclear power plant Forsmark. The model is first trained to recognize normal operating conditions. The trained model then creates reference values and calculates the deviations in relation to real data in order to identify any issues. This proof-of-concept has been evaluated and benchmarked against a currently used hybrid model with more physical modeling properties in order to identify benefits and drawbacks. Generally speaking, the created model has performed in line with expectations. The currently used tool is more flexible in its understanding of different plant states and is likely better at determining root causes thanks to its physical modeling properties. However, the created autoencoder framework does bring other advantages. For instance, it allows for a higher time resolution thanks to its relatively low calculation intensity. Additionally, thanks to its purely data-driven characteristics, it offers great opportunities for future reconfiguration and adaptation with different signal selections.

Being widely adopted by the transportation and planning practitioners, the fundamental diagram (FD) is the primary tool used to relate the key macroscopic traffic variables of speed, flow, and density. We empirically analyze the relation between vehicular space-mean speeds and flows given different signal settings and postulate a parsimonious parametric function form of the traditional FD where its function parameters are explicitly modeled as a function of the signal plan factors. We validate the proposed formulation using data from signalized urban road segments in Salt Lake City, Utah, USA. The proposed formulation builds our understanding of how changes to signal settings impact the FDs, and more generally the congestion patterns, of signalized urban segments.

Smart cities embrace cutting-edge technologies to improve transportation efficiency and safety. With the rollout of 5G and an ever-growing network of connected infrastructure sensors, real-time road condition awareness is becoming a reality. However, this progress brings new challenges. The communication and vast amounts of data generated by autonomous vehicles and the connected infrastructure must be navigated. Furthermore, different levels of autonomous driving (ranging from 0 to 5) are rolled out gradually and human-driven vehicles will continue to share the roads with autonomous vehicles for some time. In this work, we apply a data-driven control scheme called Learning Model Predictive Control (LMPC) to three different smart city scenarios of increasing complexity. Given a successful execution of a scenario, LMPC uses the trajectory data from previous executions to improve the performance of subsequent executions while guaranteeing safety and recursive feasibility. Furthermore, the performance from one execution to another is guaranteed to be non-decreasing. For our three smart-city scenarios, we apply a minimum time objective and start with a single vehicle in a two-lane intersection. Then, we add an obstacle on the lane of the ego vehicle, and lastly, we add oncoming traffic. We find that LMPC gives us improved traffic efficiency with shorter travel. However, we find that LMPC may not be suitable for real-time training in smart city scenarios. Thus, we conclude that this approach is suitable for simulator-driven, offline, training on any trajectory data that might be generated from autonomous vehicles and the infrastructure sensors in future smart cities. Over time, this can be used to construct large data sets of optimal trajectories which are available for the connected vehicles in most urban scenarios. / Smarta städer använder modern teknik för att förbättra transporteffektiviteten och säkerheten. Med införandet av 5G och ett allt större nätverk av uppkopplade sensorsystem för infrastruktur blir realtidsmedvetenhet om vägförhållandena en verklighet. Denna utveckling medför nya utmaningar. Kommunikationen mellan autonoma fordon och uppkopplade sensorsystem ger upphov till stora mängder data som måste hanteras. Dessutom kommer fordon med olika autocnominivåer (från 0 till 5) att behöva dela gatorna tillsammans med människostyrda fordon samtidigt under en tid. I detta arbete tillämpar vi en datadriven reglermetod som heter Learning Model Predictive Control (LMPC) på tre olika scenarier i en smart stad med ökande komplexitet. LMPC utnyttjar data från en tidigare lyckad körning av ett visst scenario för att förbättra prestandan på efterföljande körningar samtidigt som säkerheten och rekursiv genomförbarhet garanteras. Vidare garanteras att prestandan från en körning till en annan inte minskar. För våra tre scenarier är målet att minimerar restiden och börjar med ett enda fordon i en tvåfilig korsning. Sedan lägger vi till ett hinder på högra filen och till sist lägger vi till mötande trafik. Vi finner att LMPC ger oss förbättrad trafikeffektivitet med kortare restid. Vi finner dock att LMPC må vara mindre lämplig för realtids scenarier. Således drar vi slutsatsen att denna metod är lämplig för optimering i simulatorer, offline, på data som kan genereras från autonoma fordon och sensorsystemet i infrastrukturen. Så småningom kan vår metod användas för att konstruera stora dataset av optimala trajektorier som är tillgängliga för uppkopplade fordon i framtidens smarta städer.

In this project machine learning (ML) interatomic potentials are trained and used in molecular dynamics (MD) simulations to predict the physical properties of total energy, mean squared displacement (MSD) and specific heat capacity for systems of bulk Aluminium and Silicon. The interatomic potentials investigated are potentials trained using the ML models kernel ridge regression (KRR) and moment tensor potentials (MTPs). The simulations using these ML potentials are then compared with results obtained from ab-initio simulations using the gold standard method of density functional theory (DFT), as implemented in the Vienna ab-intio simulation package (VASP). The results show that the MTP simulations reach comparable accuracy compared to the DFT simulations for the properties total energy and MSD for Aluminium, with errors in the orders of magnitudes of meV and 10-5 Å2. Specific heat capacity is not reasonably replicated for Aluminium. The MTP simulations do not reasonably replicate the studied properties for the system of Silicon. The KRR models are implemented in the most direct way, and do not yield reasonably low errors even when trained on all available 10000 time steps of DFT training data. On the other hand, the MTPs require only to be trained on approximately 100 time steps to replicate the physical properties of Aluminium with accuracy comparable to DFT. After being trained on 100 time steps, the trained MTPs achieve mean absolute errors in the orders of magnitudes for the energy per atom and force magnitude predictions of 10-3 and 10-1 respectively for Aluminium, and 10-3 and 10-2 respectively for Silicon. At the same time, the MTP simulations require less core hours to simulate the same amount of time steps as the DFT simulations. In conclusion, MTPs could very likely play a role in accelerating both materials simulations themselves and subsequently the emergence of the data-driven materials design and informatics paradigm.

When boring machines are used, large loads are exerted on their structure. The load cycles cause material fatigue on the boring machine structure. If the material fatigue can be estimated in real-time, maintenance can be planned more efficiently and the effect of different types of usage can be evaluated. Because of the many advantages of knowing the material fatigue, the goal of this thesis is to develop a model to predict the strain of a boring machine structure and then derive an estimate of the material fatigue caused by the strain. To do this several approaches using machine learning techniques are evaluated. The input signals were selected using both coherence analysis and mutual information. It was found that linear models outperform the tested non-linear model structures, and that non-linear mechanical connections cause difficulties. The signals to be modelled contained high frequency components that were not present in the available input signals. The results show that given favorable sensor positions, an estimate of the material fatigue can be made with sufficient accuracy when using a noise model and noise realization to cover the non-existent high frequency components.

This dissertation develops systems engineering and machine learning models for aviation maintenance support. With the constant increase in demand for air travel, aviation organizations compete to maintain airworthy aircraft to ensure the safety of passengers. Given the importance of aircraft safety, the aviation sector constantly needs technologies to enhance the maintenance experience, ensure system safety, and limit aircraft downtime. Based on the current literature, the aviation maintenance sector still relies on outdated technologies to maintain aircraft maintenance documentation, including paper-based technical orders. Aviation maintenance documentation contains a mixture of structured and unstructured technical text, mainly inputted by operators, making them prone to error, misunderstanding communication, and inconsistency. This dissertation intends to develop decision support models based on systems engineering and artificial intelligence models that can automate the maintenance documentation system, extract useful information from maintenance work orders, and predict the aircraft's top degrader signals based on textual data. The first chapter of this dissertation introduces the significant setbacks of the aviation industry and provides a working ground for the following chapters. The dissertation's second chapter develops a system engineering framework using model-based systems engineering (MBSE) methodology to model the aviation maintenance process using the systems engineering language (SysML). The outcome of this framework is the design of an automated maintenance system model that can be used to automate maintenance documentation, making it less prone to error. The third chapter of the dissertation uses textual data in maintenance work orders to develop a hybrid approach that uses natural language processing (NLP) and transformer models to predict the readiness of a legacy aircraft. The model was tested using a real-life case study of the EA-6B military aircraft. The fourth chapter of this dissertation develops an ensemble transformer model based on three different transformer models. The ensemble model leverages the benefits of three different transformer architectures and is used to classify events based on an aviation log-based dataset. This dissertation's final and fifth chapter summarizes key findings, proposes future work directions, and provides the dissertation's limitations.

No description available.

<p>Predictive modeling in healthcare involves the development of data-driven and computational models which can predict what will happen, be it for a single individual or for an entire system. The adoption of predictive models can guide various stakeholders’ decision-making in the healthcare sector, and consequently improve individual outcomes and the cost-effectiveness of care. With the rapid development in healthcare of big data and the Internet of Things technologies, research in healthcare decision-making has grown in both importance and complexity. One of the complexities facing those who would build predictive models is heterogeneity of patient populations, clinical practices, and intervention outcomes, as well as from diverse health systems. There are many sub-domains in healthcare for which predictive modeling is useful such as disease risk modeling, clinical intelligence, pharmacovigilance, precision medicine, hospitalization process optimization, digital health, and preventive care. In my dissertation, I focus on predictive modeling for applications that fit into three broad and important domains of healthcare, namely clinical practice, public health, and healthcare system. In this dissertation, I present three papers that present a collection of predictive modeling studies to address the challenge of modeling heterogeneity in health care. The first paper presents a decision-tree model to address clinicians’ need to decide among various liver cirrhosis diagnosis strategies. The second paper presents a micro-simulation model to assess the impact on cardiovascular disease (CVD) to help decision makers at government agencies develop cost-effective food policies to prevent cardiovascular diseases, a public-health domain application. The third paper compares a set of data-driven prediction models, the best performing of which is paired together with interpretable machine learning to facilitate the coordination of optimization for hospital-discharged patients choosing skilled nursing facilities. This collection of studies addresses important modeling challenges in specific healthcare domains, and also broadly contribute to research in medical decision-making, public health policy and healthcare systems.</p>

The use of technology in assisting educators to use student data in well-devised ways to enhance the instruction received by students is gaining headway and the support of federal dollars across the nation. Since research has not provided insight as to whether or not reading coaches are using data technology tools with teachers, this mixed methods study sought to examine what behavioral intentions reading coaches have in using data technology tools with teachers, what variables may influence their behavioral intentions, and what trends may emerge in their views about using technology data tools with teachers. A mixed methods approach was deployed via a survey embedded in an email, and data from 61 Florida reading coaches from elementary, middle, and high schools in a large urban school district were examined using an adaptation of the Technology Acceptance Model (TAM). The results showed that collectively all reading coaches have a high level of behavioral intentions towards using a data technology tool with teachers. The study also showed that elementary, middle, and high school reading coaches vary in their degree of behavioral intentions in using a data technology tool based on different variables. Trends in data showed that reading coaches think data technology tools are helpful, but that trainings are needed and that technology tools should be user-friendly. Discussion is provided regarding the implications of the study results for all stakeholders.

Low-aspect-ratio reinforced concrete (RC) shear walls, characterized by height-to-length ratios of less than two, have been widely used as a seismic force-resisting system (SFRS) in a wide array of structures, ranging from conventional buildings to critical infrastructure systems such as nuclear facilities. Despite their extensive applications, recent research has brought to light the inadequate understanding of their seismic performance, primarily attributed to the intricate nonlinear flexure-shear interaction behaviour unique to these walls. In this respect, the current research dissertation aims to bridge this knowledge gap by conducting a comprehensive evaluation to quantify the seismic performance of low-aspect-ratio RC shear walls when used in different applications. Chapter 2 focuses on low-aspect-ratio RC shear walls that are employed in residential and industrial structures. Considering their significance, the seismic response modification factors of such walls, as defined in various standards, are thoroughly examined and evaluated utilizing the FEMA P695 methodology. The analysis revealed potential deficiencies in the current code-based recommendations for response modification factors. Consequently, a novel set of response modification factors, capable of mitigating the seismic risk of collapse under the maximum considered earthquake, is proposed. Such proposed values can be integrated into the forthcoming revisions of relevant building codes and design standards. While the FEMA P695 methodology offers a comprehensive approach to assessing building seismic performance factors, its practical implementation is associated with many challenges for practicing engineers. Specifically, the methodology heavily relies on resource-intensive and time-consuming incremental dynamic analyses, making it less feasible for routine engineering practices. To enhance its practicality, a data-driven framework is developed in Chapter 3, circumventing the need for such demanding analyses. This framework provides genetic programming-based expressions capable of producing accurate predictions of the median collapse intensities—a key metric in the acceptance criteria of the FEMA P695 methodology, for different structural systems. To demonstrate its use, the developed framework is operationalized to low-aspect-ratio RC shear walls and the predictive expression is evaluated considering several statistical and structural parameters, which showed its adequacy in predicting the median collapse intensities of such walls. Furthermore, the adaptability of this framework is showcased, highlighting its applicability across various SFRSs. Chapters 4 and 5 tackle the scarcity of experimental assessments pertaining to the seismic performance of low-aspect-ratio RC walls in nuclear facilities. The seismic hybrid simulation testing technique is employed herein to merge the simplicity of numerical simulations with the efficiency of experimental tests. Hybrid simulation can overcome obstacles related to physical specimen sizes, limited actuator capacities, and space constraints in most laboratories. In these two chapters, the experimental program delves into evaluating the seismic performance of three two-storey low-aspect-ratio nuclear RC walls under different earthquake levels, including operational, design, and beyond-design-level scenarios. Diverse design configurations, including the use of increased thickness boundary elements and different materials (i.e., normal- and high-strength reinforcement), are considered in such walls to provide a comprehensive understanding of several structural parameters and economic metrics. Key structural parameters, such as the force-displacement responses, multi-storey effects, lateral and rotational stiffnesses, ductility capacities, displacement components, rebar strains, crack patterns and damage sequences, are all investigated to provide direct comparisons between the walls in terms of their seismic performances. Additionally, economic metrics, including the total rebar weights, overall construction costs and the expected seismic repair costs, are considered in order to evaluate the seismic performance of the walls considering an economic perspective. The findings of this experimental investigation are expected to inform future nuclear design standards by enhancing the resilience and safety of their structures incorporating low-aspect-ratio RC shear walls. / Thesis / Doctor of Philosophy (PhD)