As one of the most promising pathways in the transition period towards the low carbon economy, a large scale implementation of electric vehicles (EV) is expected in the near future. Concentration of EV charging in a small area or within a short time will dramatically affect the load demand profile, especially the peak load in the distribution network. As a result, distribution transformers are facing hazards of shortened lifetime due to extra loads, and direct failures caused by potential overloads. Considering the large number of distribution transformers and the massive investment involved, the adaptability of the population of distribution transformers under future EV scenarios should be assessed. In this thesis, an assessment strategy for the future adaptability of distribution transformer population under EV scenarios is introduced. Assessing the adaptability is to understand the impact of the hot-spot temperature, loss-of-life, expected lifetime and failure probability of each individual in the distribution transformer population. Determination of hot-spot temperature of distribution transformers is essential for the assessment. In order to achieve accurate prediction of hot-spot temperatures under EV scenarios, thermal parameters should be refined for individual distribution transformers so that their thermal characteristics can be reflected more accurately than using the generic values recommended for all distribution transformers in the IEC loading guide. Two methods for the refinement are proposed in this thesis. One method is to curve-fit hot-spot temperatures measured in the extended heat run test; and the other is to calculate each parameter with developed equations in the loading guide with standard heat run test results. The assessment strategy is introduced and demonstrated on a group of selected distribution transformers from the population under three EV scenarios, i.e. Business as usual (BAU), High-range and Extreme-range scenarios, which represent 0%, 32% and 58.9% EV penetration levels respectively. Results show that EV charging would be less concerned on the acceleration of thermal ageing than the direct failure due to breakdown caused by decrease of dielectric strength in an event of bubbling. Since the peak load and hot-spot temperature under EV scenarios only last for a short time and would be compensated by low values during the rest time of a day, which eventually leads to a moderate thermal ageing. Occasionally, severe over-ageing can be caused by extremely high hot-spot temperatures, and the lifetime will be reduced to an unacceptable level. However, on such occasions, hot-spot temperatures would be high enough to trigger bubbling and reduce the dielectric strength of transformer's insulation system to a level that is incapable of undertaking the voltage stress, which eventually causes breakdown of transformers. In terms of the failure probability, results show that no distribution transformers are facing failure risks due to bubbling under BAU scenario. Failure starts under High-range scenario. If transformers possessing a failure probability over 50% are identified as high risk, then 13% of investigated transformers are at high risk under High-range scenario, while it increases to 39% under Extreme-range scenario. It is found that the failure probability is dominantly controlled by the peak load, other factors such as transformer age and installation conditions are less influential. A threshold peak load of around 1.5 p.u. is observed that distinguishes transformers in high risk from others under Extreme-range scenario. This observation could be applied to assist the asset management under future EV scenarios that the peak load of distribution transformers should be restricted below 1.5 p.u. to prevent potential failure due to bubbling.

D.Ing. / In this thesis, contactless power transmission with sliding transformers to mobile loads is investigated. The sliding transformer comprises an extended primary winding - placed along the path of travel - which is fed through a core containing a secondary winding and which is attached to a mobile load. In the first chapter problems with existing power supply systems are discussed whereafter several possible applications of such a system are identified. In this chapter different methods of obtaining contactless power transmission are also discussed and a brief discussion of a practical, cost effective system is given. In chapter two the application of different resonant topologies to compensate the large structural inductance of the sliding transformer is investigated. A Series Resonant, Series Loaded topology is identified as the most practical and efficient topology for this application and a more detailed analysis of this topology - as applied to this case - is given. Formulas to calculate the sliding transformer parameters are given in the third chapter and two quantities relating the optimization of sliding transformers to the parameters thereof, are defined. Methods to optimize sliding transformers in terms of these two quantities, and also with respect to EMI, are also discussed in this chapter. The design of a five meter long experimental system which delivers 15 kVA to a mobile load is given in chapter four and experimental results of this system are presented. In chapter five a detailed investigation into different output power control methods is conducted. The effect of these control methods on EMI is investigated by means of numerical simulation of a Series Resonant, Series Loaded system under these different control methods. The most complex control method which generates the most EMI, namely frequency and pulse burst control, is implemented into the experimental system to investigate the limitations thereof. Experimental results of this controlled system are presented. It is found that this control method increases the control range obtainable with classical frequency control but introduces many limitations due to discontinuous primary current. It is concluded in the last chapter that it is possible to optimize sliding transformers within achievable cost constraints to such an extent that such systems have acceptable efficiencies and that reasonable system lengths are obtainable.

The aim of this thesis was to get acquainted with the electrical properties of distribution transformers, which are used in the area of high voltage technology. Further aim was to describe the existing diagnostic methods according to the standards and to introduce the measuring workstation. Finally, the diagnostic of transformer was performed, and the results were evaluated.

No description available.

MnZn ferrites are widely used in power electronics applications where the switching frequency is in the range of several tens of kilohertz to a megahertz. In this range of frequencies the combination of relatively high permeability and relatively low conductivity found in MnZn ferrite helps to minimize the size of magnetic devices while maintaining high efficiency. The continuing improvement in semiconductor switches and circuit topologies has led to use of high-frequency switching circuits at ever increasing power levels. The magnetic devices for these high-power, high-frequency circuits require magnetic CORES that are significantly larger than standard ferrite-core devices used at lower power levels. Often such large ferrite cores must be custom designed, and at present this custom design is based on available material information without regard for the physical size of the structure. This thesis examines the issues encountered in the use of larger MnZn ferrite cores for high-frequency, high-power applications. The two main issues of concern are the increased power dissipation due to induced currents in the structure and the change in inductance that results as the flux within the core is redistributed at higher frequencies. In order to model these problems using either numerical or analytical methods requires a reliable and complete set of material information. A significant portion of this work is devoted to methods for acquiring such material information since such information is not generally available from the manufacturers. Once the material constants required for the analysis are determined, they are used in both closed-form and numerical model to illustrate that large ferrite cores suffer significant increases in loss and significant decreases in inductance for frequencies as low as several hundred kilohertz. The separate impacts of the electrical and magnetic losses in the core are illustrated through the use of linear finite element analyses of several example core structures. The device impedances calculated using the FEA tools show fair agreement with measurement. An analysis of gapped structures and segmented cross-sections shows that these design options can reduce the severity of the dimensional problems for some designs. / Ph. D.

A new method for analyzing axisymmetric, high-frequency transformers is presented. The method is based on the simultaneous solution of the coupled, nonlinear thermal and electromagnetic equations using the finite element method. A novel technique for modeling the reluctivity of the soft-ferrite core material permits a time-harmonic transformation of the electromagnetic equations. This eliminates the need to step through time while maintaining the effects of hysteresis losses. Also, a quasi-steady formulation of the heat-conduction equation eliminates the time dependency on the thermal problem. A direct substitution iterative scheme is used in conjunction with the finite element method to compensate for the coupled and nonlinear nature of the equations. To verify the magnetics portion of the finite element code numerically, a linear, uncoupled test case is given which compares the magnetic results from the present method to those from a commercial software package. To investigate the accuracy of the fully coupled and nonlinear model, an example is presented which compares the results from the numerical analysis of an inductor to those obtained by experimental measurement. / Master of Science

M.S.

High voltage transformers inherently contain undesirable parasitic reactances. A resonant circuit formed by the leakage inductance and parasitic reflected capacitance, makes various resonant technologies practical for high voltage dc applications. The components of a typical high voltage power stage will be investigated in order to identify these parasitics and determine their influence on converter operation. Proto-type high voltage converters will be designed and built using PWM and quasi-resonant techniques and experimental results will be given. Dc analysis will be performed for the PWM and the quasi-resonant designs in order to obtain insight into the actual operation of the two converters. The design of the two proto-type converters will be reviewed and design guidelines will be established. / Master of Science

The quality of an input sketch can affect performance of the computational algorithms. However, the quality of a sketch is not often considered when working with sketch tasks and automated sketch quality prediction has not been previously studied. This thesis presents quality prediction on the "Sketchy" dataset. The method presented here predicts a quality label rather than a zero to one quality metric. This thesis predicts an understandable label rather than a computer-generated quality metric with no human input. Previous tasks like sketch classification have used a transformer architecture to leverage the vector format of sketches. The architecture used in sketch classification was called Sketchformer. The Sketchformer was adopted and trained to predict quality labels of hand-drawn sketches. This Sketchformer architecture achieves 66% accuracy when predicting the 5-labels. The same transformer achieves up to 97% accuracy in a different experiment when combining the different labels into good versus bad (2-label) experiments. The sketchformer significantly outperforms the SVM baseline. The results of the experiments show that the transformer embedding space facilitates separation of 'good' sketch quality from 'bad' sketch quality with high accuracy. / Master of Science / If pictures are worth 1000 words, then sketches are worth a few hundred words. Sketches are easy to create using a pen and tablet. Objects in the sketches can be drawn many ways, depending on the talent of the creator and pose of the object. The quality of the sketches vary pretty drastically. When using sketches in computer vision tasks, the quality of a sketch can affect the performance of the computational algorithm. However, the quality of a sketch is not often considered when working with other sketch tasks. One common sketch task is called Sketch-Based Image Retrieval (SBIR). The input of this task is the sketch of an object/subject, and the model returns a matching image of the same object/subject. If the quality of the input sketch is bad, the output of this model will be poor. This thesis predicts the quality of sketches. The dataset used is called the "Sketchy" dataset, this dataset was originally used to study SBIR. However, the creators of the dataset provided quality labels for the sketches. This allows for quality prediction on this dataset, which has not previously been completed. There are 5 different labels assigned to sketches. One of the experiments completed for this thesis was predicting 1 of the 5 labels for each sketch. The other experiments for this thesis create good and bad labels by combining the 5 labels. The Sketchformer architecture created by Ribeiro et al. is used to run the experiments. The Sketchformer achieves 66% on the 5-label experiment and up to 97% on the good and bad (2-label) experiment. This transformer outperforms a Support Vector Machine baseline on this quality labels. The results of the experiments show that the transformer applied to this dataset is a valuable contribution by surpassing the baseline on multiple tasks. Additionally, accuracy values from these experiments are similar to values found in the corresponding image quality prediction task.

Present demagnetization methods for large power system transformers are time consuming and can be dangerous to persons performing demagnetization. The work of this thesis was to develop improved demagnetization methods and to construct an automated instrument that would implement the methods developed. One previously developed method was analyzed for effectiveness. Then, two new methods for demagnetization were developed and also analyzed for effectiveness. An automated test instrument prototype was redesigned to be able to accommodate these methods and to improve the safety of the user. The previously developed method attempts demagnetization based on current flow behavior characteristics. The first new method is a magnetic flux estimation based on saturation time. The second new method is also based on measuring saturation time, modified to account for the variable voltage loss due to wire resistance. The second of the two new methods developed proved to be the most effective for demagnetization and was able to demagnetize a transformer within an error margin of 2%. The instrument designed to perform the demagnetization with this new routine is now in early production stages for an expanded field trial with transformer maintenance teams.