An improved method for designing controllers for nonlinear stochastic systems is developed and analyzed. The resulting controller consists of a nonlinear control law coupled with an adaptive state estimator. The nonlinear control law is developed first. Using Taylor series expansion, linear approximations to the nonlinear systems are generated at selected points in the operating region. Then a control law which will produce the desired response is developed for each linearized configuration using conventional techniques for linear systems. The resulting control law parameters are treated as tabulated values from a set of unknown continuous functions of the nonlinear system parameters. These unknown functions are approximated at all points in the operating region by fitting curves to the tabulated data. The stability and convergence aspects of this nonlinear control law are analyzed in detail, with several derivations given and theorems proved. Two examples are given to illustrate the design procedure and evaluate its performance. The design procedure is extended to stochastic systems by incorporating a suitable state estimator. Two members of the class known as partitioned adaptive estimators (PAE's) are evaluated and their performance compared. The formulation known as the modified semi-Markov PAE is shown to be superior. The design, execution, and analysis of the experiments comprising the evaluation are discussed in detail, with particular attention given to correlating the performance of the estimators with the behavior of the weighting coefficients. Numerous figures and tables which amplify the discussions, along with some suggestions for further research, are also included. / Ph. D.

A standard form for linear and nonlinear perfect model following control problems is introduced, and the associated control laws developed. The error dynamics of such systems are analyzed with respect to stability of the error. The effects on the error dynamics of measurement errors and parameter variations are also analyzed, and it is seen that the perfect model following control problem is reduced to that of an error regulator. The linear problem is analyzed to show that virtually all common problems are equivalent to standard form problems through similarity transformations. In the standard form, simple expressions for the control law and error dynamics are used to solve the problem. The linear problem is also analyzed with respect to problems of different order model and plant systems, resulting in augmented system equations. These augmented systems are chosen so that the original dynamics are retained, and so that the higher order problem is in the standard form. The standard form problem is then solved as before. Imperfect model following control problems are analyzed, with three associated results. First, a new t€St for perfect model following is developed. Pairs of models and plants that fail this or other tests are imperfect model following control problems. Second, the effect of using perfect model following control laws on such problems is determined to be equivalent to the addition of a forcing function on the error regulator problem. Third, a new approach to the solution of imperfect model following control problems is shown. This approach seeks to find models that simultaneously satisfy the criteria for perfect model following while retaining the desired characteristics of the intended model. The methods developed in this analysis are applied to problems that illustrate all the principles addressed. The final example is a detailed application to a nonlinear simulation of the F-18 airplane involving control of all degrees of freedom over a large range of angles of attack. / Ph. D.

Submitted in fulfillment of the requirements of Doctor of Engineering, Durban University of Technology, Durban, South Africa, 2017. / The need to quantify and model transformations that have taken place in the cities of South Africa is one of the grand challenges linked to country’s transition to Democracy. Given the complexities associated with different stages of city transformation, it is imperative that models used to unpack processes of city transformation are novel. In this study it emerged that statistical methods alone are not adequate to fully present, in a comprehensible way, all facets of drivers of city transformation. As a result, statistical methods have been combined with mathematical and system dynamics models. Results revealed that city transformations derive from a number of triggers. Underlining these are income, migration and houses. The empirical data collected through questionnaire survey that was later incorporated into mathematical models demonstrated that income is a primary driver that fuels city migration. System Dynamic Models demonstrated that the availability of houses or accommodation serve as constraints that keep the city population within the limits of the carrying capacity of a city. In addition it was further confirmed, through mathematical models that income has varying effects on the attractiveness of cities. This was found to be linked to the shape of the distribution of income in the city. A normally distributed income with a peak in the middle results in a city being more attractive than an evenly distributed income that peaks either at very low or high income levels. This observation brought forth a need to test heterogeneity when analyzing city transformation using income as an index. Mathematical Models that incorporated heterogeneity demonstrated the usefulness of systems analysis in unpacking the mechanism of city transformation, a component of city management that requires serious consideration for planning, budgeting and provision of limited resources like houses in the cities. Success of methods used in this study led to a conclusion that these can be enhanced through other techniques like agent based models. With this call, improvements on this study that can be attained through these techniques are recommended. This will enrich the understanding of the transformation and dynamics of cities under different conditions than those that exist in South Africa. / D

Thesis: M.S., Massachusetts Institute of Technology, Department of Civil Engineering, 1978 / Includes bibliographical references. / M.S. / M.S. Massachusetts Institute of Technology, Department of Civil Engineering

Computing / (M.Sc. (Information Systems))

Increased energy performance of the building stock of European Union is seen as an important measure towards mitigating climate change, increasing resource utilisation efficiency and energy supply security. Whether to improve the supply-side, the demand-side or both is an open issue. This conflict is even more apparent in countries such as Sweden with a high penetration of district heating (DH). Many Swedish DH systems have high share of secondary energy resources such as forest industry residuals, waste material incineration and waste heat; and resource efficient cogeneration of electricity in combined heat and power (CHP) plants. When implementing an energy conservation measure (ECM) in a DH connected building stock, it will affect the operation of the whole DH system. If there are CHP plants and the cogeneration of electricity decreases due to an ECM, and this electricity is valued higher than the fuel savings, the consequences of the ECM would be negative.  These complex relationships are investigated by conducting a case study on the Eskilstuna DH system, a renewable energy supply system with relatively high share of cogenerated electricity. Heat demand profiles of ECMs are determined by building energy simulation, using recently deep energy retrofitted multifamily buildings of the “Million Programme”-era in Eskilstuna as model basis. How implementing ECMs impact on the DH system’s heat and electricity production under different electricity revenue scenarios has been computed and evaluated in terms of resource efficiency and CO2 emissions.  The results show that different ECMs in the buildings impact differently on the DH system. Measures such as improved insulation level of the building’s envelope, that decrease the heat demand’s dependence to outdoor temperature, increase the amount of cogenerated electricity. While measures such as thermal solar panels, which save heat during summer, affects the absolute amount of cogenerated electricity negatively. Revenues from cogenerated electricity influence the amount of cost-effectively produced electricity much more than the impact from ECMs. Environmental benefits of the ECMs, measured in CO2 emissions and primary energy consumption, are quite small in DH systems that have high share of forest residual fuels and electricity cogeneration. The consequences can even be negative if ECMs lead to increased need of imported electricity that is produced resource inefficiently or/and by fossil fuels. However, all studied ECMs increase the relative amount of cogenerated electricity, the ratio between amount of cogenerated electricity and the heat load. This implied that all ECMs increase the overall efficiency of the Eskilstuna DH system.

Wave Energy Converters (WEC) are used to transform energy stored in ocean waves into electrical energy. One type of WECs consists of buoyant bodies. To extract energy from their motion, hydraulic cylinders can be used to generate hydraulic power. For conversion into electric power various systems have been analysed in literature. However, the focus was put on efficiency and rigorous analyses of the system behaviour are still missing. In this paper an exemplary system consisting of two hydraulic cylinders, switchable check valves, accumulators and three motor-generator sets is analysed with help of simulation and measurement. This exemplary system is called WavePOD and was installed at the Institute for Fluid Power Drives and Controls (IFAS) of RWTH Aachen University together with Aquamarine Power and Bosch Rexroth for testing. In this paper the data collected during various test phases is used for system analysis and for validating the simulation. The simulation model is presented. The system’s response to various switching operations is investigated. Comparing the simulation with measurements validates the system`s dynamic model.

<p>The most common way to treat wastewater in Sweden today is by a combination of an activated sludge process and a chemical treatment. These processes are not very energy efficient and may not be the most environmentally friendly. That is why it is interesting to evaluate new alternative methods to see if they can be less harmful to the environment and help to a lower resource consumption. New techniques are tested in a pilot installation at Hammarby Sjöstad, Sjöstadsverket. To be able to evaluate these techniques in a wide perspective system analyses are conducted where different impacts can be considered, advantages and disadvantages in the systems.</p><p>Five system structures for water treatment have been constructed in Matlab/Simulink in this study. The URWARE (URbanWAter REsearch) model was used to conduct a system analysis, which made a comparison of the different systems possible. Four of the systems contained a new technique for water treatment, consisting of an anaerobic membrane bioreactor (MBR) combined with a VSEP (Vibratory Shear Enhanced Process) unit. This is one of the techniques that now are tested in Sjöstadsverket. The fifth system was a reference system designed as a conventional treatment plant. The wastewater treated in the different systems was in some cases mixed wastewater, in other source sorted, with or without food waste from kitchen waste disposers.</p><p>Results from the system analysis shows that the MBR technique gives some advantages compared to the conventional water treatment. A large part of the nutrients in incoming materials are recycled. The highest recycling percent is reached when the MBR technique is combined with compost that handles the food waste. The total impact on the environment by greenhouse gases and eutrophic substances is lower in all the alternative systems compared to the reference system. The system with the source sorted urine has the most favourable exergy turnover. When reversed osmosis is used in the treatment a lot of energy is needed which leads to a poor exergy turnover and the conventional technique is more advantageous in these cases. All the alternative systems give a larger acidification which is a disadvantage with the MBR technique.</p> / <p>Vid behandling av avloppsvatten i Sverige används idag i de flesta fall en aktivslamprocess kompletterad med kemiskfällning. Denna process är inte så energieffektiv och kanske inte heller den mest optimala ur miljösynpunkt. Därför kan det vara intressant att utvärdera nya alternativa metoder för att se om dessa kan vara skonsammare mot miljön och bidra till en minskad resursförbrukning. I Sjöstadsverket, som är en försöksanläggning vid Hammarby Sjöstad, undersöks flera nya tekniker i mindre skala. För att kunna utvärdera dessa tekniker utifrån ett helhetsperspektiv görs systemanalyser, där för- och nackdelar och andra effekter kan vägas in.</p><p>I föreliggande studie har fem systemstrukturer för vattenrening byggts upp i Matlab/Simulink. Substansflödesmodellen URWARE (URban WAter REsearch) användes för att utföra en systemanalys, som gjorde en jämförelse av de olika systemen möjlig. Fyra av systemen innehöll ny reningsteknik bestående av en anaerob membranbioreaktor (MBR) kopplad till en VSEP- enhet (Vibratory Shear Enhanced Process). Det är en teknik som nu testas i Sjöstadsverket. Det femte systemet var ett referenssystem utformat som ett konventionellt reningsverk. De vatten som behandlades i reningsverken var i vissa fall blandat avloppsvatten, i andra källsorterat, med och utan matavfall från köksavfallskvarnar.</p><p>Utifrån systemanalysen kan det konstateras att MBR-tekniken ger en del fördelar gentemot den konventionella reningen. Bland annat eftersom en stor andel av näringsinnehållet i avloppsvattnet kan fångas upp och återföras till åkermark. Bäst kretsloppspotential fås när membranbioreaktorn kombineras med en kompost som tar hand om allt matavfall. Totalt sett bidrar också alla de nya reningssystemen till lägre växthusgasutsläpp och lägre utsläpp av eutrofierande ämnen än referenssystemet med konventionell rening. Ur exergisynpunkt är den nya tekniken bättre än den konventionella i det fall då urin källsorteras. I de fall då omvänd osmos används vid behandling är den konventionella processen mer fördelaktig. En nackdel med MBR-tekniken är att utsläpp av försurande ämnen är markant större än för den konventionella tekniken.</p>

This research explores the combined use of soft systems methodology (SSM) and UML based business process modelling (BPM) techniques. These two techniques are integrated to provide a framework for the analysis and definition of suitable business process models. Such integration better supports developers following objectoriented (00) approaches than traditional business process modelling. The thesis describes the importance and difficulties in getting development proj ects aimed at the correct needs. We provide an overview of current business process modelling practices. From this it is argued that current practices show two major weaknesses. Firstly, the modelling language that is used is not a current standard amongst developers who now expect 00 and UML based approaches. Secondly, the techniques used do not emphasise analysis, often resulting in a lack of appreciation of the problem. In order to deal with these inadequacies, the thesis critically examines suitable techniques that can be used to analyse and model business processes to support the developer's requirements. The examination of SSM reveals that the technique does deal with the analysis limitations of current business process modelling techniques. SSM has been linked to information systems provision by previous researchers. Unfortunately the examination ofthese research attempts shows that the linking is conducted in an ad-hoc manner with no underlying theoretical basis or emphasis on business process modelling. We show how soft systems methodology techniques can be married with Eriksson and Penker (2000) UML business process modelling techniques following Mingers (2001) multi-methodology framework in a way that can over come these difficulties. This combined business analysis and modelling technique is applied to the production process in an aluminium rolling plant. Based on the experiences at one site, the integrated approach is able to deal with the complexities caused by multiple stakeholders, and is able to provide a UML representation of the required business process to guide developers. / Thesis (M.Sc.)-University of Natal, Pietermaritzburg, 2003.

The problem motivating this research is that ineffective communication may hamper systems development efforts. Specifically germane to this study are development efforts characterized as task-oriented, and which require information-sharing and problem-solving activities. This research problem motivated an analysis of the communication process and lead to the development of a temporal framework that delineates variables associated with task-oriented, end user/systems analyst communication interactions. Several variables within this framework are depicted in two theoretical models. The first model shows the theoretical relationship between an independent variable, communication mode (represented by asynchronous computer conferencing and face-to-face conferencing), and five dependent variables: (1) the amount of information shared, (2) the significance of the information shared, (3) the comprehensiveness of the information shared, (4) the perception of progress toward the goal, and (5) the perception of freedom to participate. The second model depicts the assumed interaction between communication mode, the five variables cited above (now acting as independent variables), and a dependent variable, communication quality. There are two theoretical components of communication quality: (1) deviation from an optimal set of user requirements, and (2) the degree of convergence (unity based on mutual understanding and mutual agreement) emanating from a communication interaction. Using the theoretical models as a guide, an experiment was designed and performed to test the relationships among the variables. The experimental results led to the rejection of all null hypotheses; the results strongly favored face-to-face conferencing for solving task-oriented, information-sharing problems analagous to the case used in the present study. The findings indicate that asynchronous computer conferencing may have a detrimental effect on the thoroughness of information exchange, on the relevance of the information shared in terms of making the correct decision, and on the completeness of the consideration given to all problem dimensions.