<p>The section for volume, flow and temperature at SP Technical Research</p><p>Institute of Sweden performs measurements of volume, flow and temperature</p><p>in liquids.</p><p>Flow meters are best calibrated in its installation to take sources of error like</p><p>installation effects and the medium into account. If this can be done without</p><p>having to place measurement equipment inside the pipe it will mean several</p><p>practical benefits.</p><p>Since many years, clamp-on ultrasonic flow meters have been available on the</p><p>market. But even with today’s improvements they still have a measurement</p><p>uncertainty in the measurements that is five to ten times too big to make them</p><p>useful as references for calibration procedures.</p><p>This thesis focuses on analysis, using reversed engineering, of an existing</p><p>clamp-on ultrasonic flow meter.</p><p>The goal of the project is evaluation and further development of the ultrasonic</p><p>flow meter’s existing computer interface with the purpose of offering the</p><p>option of using Microsoft Excel and Visual Basic for data acquisition and</p><p>measurement of the flow rate of liquids.</p>

The main part of this thesis focuses on optimal experiment design for system identification within the prediction error framework. A rather flexible framework for translating optimal experiment design into tractable convex programs is presented. The design variables are the spectral properties of the external excitations. The framework allows for any linear and finite-dimensional parametrization of the design spectrum or a partial expansion thereof. This includes both continuous and discrete spectra. Constraints on these spectra can be included in the design formulation, either in terms of power bounds or as frequency wise constraints. As quality constraints, general linear functions of the asymptotic covariance matrix of the estimated parameters can be included. Here, different types of frequency-by-frequency constraints on the frequency function estimate are expected to be an important contribution to the area of identification and control. For a certain class of linearly parameterized frequency functions it is possible to derive variance expressions that are exact for finite sample sizes. Based on these variance expressions it is shown that the optimization over the square of the Discrete Fourier Transform (DFT) coefficients of the input leads to convex optimization problems. The optimal input design are compared to the use of standard identification input signals for two benchmark problems. The results show significant benefits of appropriate input designs. Knowledge of the location of non-minimum phase zeros is very useful when designing controllers. Both analytical and numerical results on input design for accurate identification of non-minimum phase zeros are presented. A method is presented for the computation of an upper bound on the maximum over the frequencies of a worst case quality measure, e.g. the worst case performance achieved by a controller in an ellipsoidal uncertainty region. This problem has until now been solved by using a frequency gridding and, here, this is avoided by using the Kalman-Yakubovich-Popov-lemma. The last chapter studies experiment design from the perspective of controller tuning based on experimental data. Iterative Feedback Tuning (IFT) is an algorithm that utilizes sensitivity information from closed-loop experiments for controller tuning. This method is experimentally costly when multivariable systems are considered. Several methods are proposed to reduce the experimental time by approximating the gradient of the cost function. One of these methods uses the same technique of shifting the order of operators as is used in IFT for scalar systems. This method is further analyzed and sufficient conditions for local convergence are derived.

<p>The main part of this thesis focuses on optimal experiment design for system identification within the prediction error framework.</p><p>A rather flexible framework for translating optimal experiment design into tractable convex programs is presented. The design variables are the spectral properties of the external excitations. The framework allows for any linear and finite-dimensional parametrization of the design spectrum or a partial expansion thereof. This includes both continuous and discrete spectra. Constraints on these spectra can be included in the design formulation, either in terms of power bounds or as frequency wise constraints. As quality constraints, general linear functions of the asymptotic covariance matrix of the estimated parameters can be included. Here, different types of frequency-by-frequency constraints on the frequency function estimate are expected to be an important contribution to the area of identification and control.</p><p>For a certain class of linearly parameterized frequency functions it is possible to derive variance expressions that are exact for finite sample sizes. Based on these variance expressions it is shown that the optimization over the square of the Discrete Fourier Transform (DFT) coefficients of the input leads to convex optimization problems.</p><p>The optimal input design are compared to the use of standard identification input signals for two benchmark problems. The results show significant benefits of appropriate input designs.</p><p>Knowledge of the location of non-minimum phase zeros is very useful when designing controllers. Both analytical and numerical results on input design for accurate identification of non-minimum phase zeros are presented.</p><p>A method is presented for the computation of an upper bound on the maximum over the frequencies of a worst case quality measure, e.g. the worst case performance achieved by a controller in an ellipsoidal uncertainty region. This problem has until now been solved by using a frequency gridding and, here, this is avoided by using the Kalman-Yakubovich-Popov-lemma.</p><p>The last chapter studies experiment design from the perspective of controller tuning based on experimental data. Iterative Feedback Tuning (IFT) is an algorithm that utilizes sensitivity information from closed-loop experiments for controller tuning. This method is experimentally costly when multivariable systems are considered. Several methods are proposed to reduce the experimental time by approximating the gradient of the cost function. One of these methods uses the same technique of shifting the order of operators as is used in IFT for scalar systems. This method is further analyzed and sufficient conditions for local convergence are derived.</p>

The section for volume, flow and temperature at SP Technical Research Institute of Sweden performs measurements of volume, flow and temperature in liquids. Flow meters are best calibrated in its installation to take sources of error like installation effects and the medium into account. If this can be done without having to place measurement equipment inside the pipe it will mean several practical benefits. Since many years, clamp-on ultrasonic flow meters have been available on the market. But even with today’s improvements they still have a measurement uncertainty in the measurements that is five to ten times too big to make them useful as references for calibration procedures. This thesis focuses on analysis, using reversed engineering, of an existing clamp-on ultrasonic flow meter. The goal of the project is evaluation and further development of the ultrasonic flow meter’s existing computer interface with the purpose of offering the option of using Microsoft Excel and Visual Basic for data acquisition and measurement of the flow rate of liquids.

Intensive development in the field of high-temperature superconductors (HTS) has resulted in long length HTS with high current carrying capacity. The performance of HTS in the form of multifilamentary silver sheathed Bi-2223 tapes is close to the requirements for power application. Several power apparatus prototypes based on HTS have been and are being demonstrated. Recently the first commercial order for a HTS power apparatus was placed. Tennessee Valley Authority ordered two 12 MVA synchronous condensers with HTS rotor windings for reactive power compensation. In most power devices the conductor is carrying an AC transport current while it is exposed to an AC magnetic field transverse to the current path. In certain applications such as multi-layer power transmission cables or the control winding in a controllable reactor, the conductors are exposed to a magnetic field component longitudinal to the tape axis that is parallel with the current path. In this thesis, the losses in HTS tapes in longitudinally applied magnetic field are investigated and a single phase small scale controllable reactor with a loss optimized HTS control winding is designed, constructed and tested. An experimental setup to measure the losses in tapes exposed to longitudinal magnetic fields, transport currents, frequencies, and temperatures is presented. The system is based on a calorimetric loss measuring method where the losses are determined by the temperature response of the conductor when exposed to magnetic field and/or carrying transport current. Semi-empirical loss models are developed from the results of the loss measurements of non-twisted multifilamentary Bi-2223 tapes. The losses when the sample is carrying transport currents and is exposed to longitudinal magnetic fields are found to consist of hysteresis, flux flow, and eddy current losses. The hysteresis losses can be modeled with the critical state model. The flux flow losses can be described with a power law dependent current voltage characteristic. The loss models are well suited for the design of power devices in which HTS is exposed to longitudinal magnetic fields. Based on the knowledge of the losses a single phase small scale controllable reactor with a loss optimized HTS control winding is designed, constructed and tested. A simple model of the magnetic circuit of the controllable reactor is presented. Calculations from the model are compared with experimental measurements. The feasibility to design controllable shunt reactors with large linear dynamic range, low losses and limited harmonic distortion is demonstrated. Compared to a copper control winding the HTS control winding increases the dynamic range and reduces the size due to the high current density and reduces the losses of the control winding. / <p>QC 20100927</p>

University of Macau / Faculty of Science and Technology / Department of Electrical and Electronics Engineering

University of Macau / Faculty of Science and Technology / Department of Electrical and Electronics Engineering

University of Macau / Faculty of Science and Technology / Department of Electrical and Electronics Engineering

University of Macau / Faculty of Science and Technology / Department of Electrical and Electronics Engineering

University of Macau / Faculty of Science and Technology / Department of Electrical and Electronics Engineering