<p dir="ltr">Deciding setpoints for distributed energy resources (DERs) via local control rules rather than centralized optimization offers significant autonomy. The IEEE Standard 1547 recommends deciding DER setpoints using Volt/VAR rules. Although such rules are specified as non-increasing piecewise-affine, their exact shape is left for the utility operators to decide and possibly customize per bus and grid conditions. To address this need, this thesis optimally designs Volt/VAR rules to minimize ohmic losses on lines while maintaining voltages within allowable limits. This is practically relevant as excessive reactive injections could reduce equipment's lifetime due to overloading. First, we consider a linearized single-phase grid model. Even under this setting, optimal rule design (ORD) is technically challenging as Volt/VAR rules entail mixed-integer models, stability implications, and uncertainties in grid loading. Uncertainty is handled by minimizing the average losses under voltage chance constraints. To cope with the piecewise-affine shape of the rules, we build upon our previous reformulation of ORD as a deep learning task. A recursive neural network (RNN) surrogates Volt/VAR dynamics and thanks to back-propagation, we expedite this chance-constrained ORD. RNN weights coincide with rule parameters, and are trained using primal-dual decomposition. Second, we consider a multi-phase grid model with alternating current power flow (AC-PF). Different from the single-phase model, we additionally add imbalance chance constraints and use chain rule differentiation to obtain derivatives during the primal-dual updates. Numerical tests corroborate the efficacy of this novel ORD formulation and solution methodology.</p>

This thesis explores the losses occurring in a specific region located in the southern part of Eskilstuna, where ESEM serves as the network owner. The study places particular emphasis on power factor and load factor as key factors. Its objective is to analyze and comprehend the extent of losses in the network and their implications. The results demonstrate that losses in the chosen area of the Eskilstuna network align with prior research and theoretical expectations. Based on calculations, the losses in this network segment amount to approximately 483 MWh per year, corresponding to a cost of approximately 272,000 Swedish Kronor for ESEM in 2022. The method of studies is based on calculations, analyzer, and literature studies to achieve a reliable result. The results show that losses within the selected area of the Eskilstuna network are consistent with previous research and theoretical expectations. Of the total loss, there are some customers who show larger losses compared to the average case. These customers were identified, and a summary analysis was made of their power factor and load factor standard deviation. These losses depend on various factors, such as the existence of reactive power, technical and non-technical losses. Measures to deal with these problems are suggested based on previous research and literature studies. Optimization of transformers, phase compensation and monitoring are some of these measures. This study provides significant insights into power grid losses and proposes potential strategies to mitigate these losses in the examined region. By minimizing such losses, ESEM has the potential to enhance the efficiency of the electrical grid, lower carbon dioxide emissions, and enhance economic outcomes for energy producers, the electricity grid operator, and consumers. For continued progress in the field, further research, and implementation of the proposed measures to streamline the power grid and reduce losses in the future is recommended. / Elnätförluster har konsekvenser för miljön, energisystemet och slutanvändarna. När energi går förlorad i elnätet minskar den totala effektiviteten i systemet och ökar belastningen på kraftgenereringen. Detta kan leda till ökade kostnader för energiproducenter och potentiellt högre priser för konsumenterna. Dessutom kan ökade elnätförluster bidra till en ökning av koldioxidutsläppen från kraftverken och därmed ha en negativ miljöpåverkan. Syftet med studien är att undersöka det befintliga elnätet i det utvalda området genom analys av data från mät-elnätavdelningen i ESEM. Metoden för studien är baserad på beräkningar, analyser och litteraturstudier för att uppnå ett pålitligt resultat. Resultaten visar att förluster inom det valda området av Eskilstunanätverket överensstämmer med tidigare forskning och teoretiska förväntningar. Baserat på beräkningar uppgår förlusterna i detta nätsegment till cirka 483 MWh, vilket motsvarar en kostnad på cirka 272 000 svenska kronor för ESEM 2022. Av den totala förlusten finns det vissa kunder som uppvisar större förluster jämfört med genomsnittsfallet. Dessa kunder identifierades, och en översiktsanalys gjordes av deras effektfaktor och standardavvikelse för belastningsfaktorn. Dessa förluster beror på olika faktorer, såsom existensen av reaktiv effekt, tekniska och icke-tekniska förluster. Åtgärder för att hantera dessa problem föreslås baserat på tidigare forskning och litteraturstudier. Optimering av transformatorer, faskompensering och övervakning är några av dessa åtgärder. Denna studie bidrar med viktig kunskap om elnätförluster och identifierar potentiella åtgärder för att minska förlusterna i det studerade området. Genom att minska förlusterna kan man öka elnätets effektivitet, minska koldioxidutsläppen och förbättra ekonomiska aspekter för energiproducenter, elnätägaren och konsumenter. För fortsatta framsteg inom området rekommenderas ytterligare forskning och implementering av de föreslagna åtgärderna för att effektivisera elnätet och minska förlusterna i framtiden.

The master’s thesis is focused on the creation of models of DC/DC converters and control program for input parameters and their configuration. The first part is theoretical and she is dedicated to the description of the program Matlab. The second chapter is dedicated to DC/DC converters. Their function is described there and the necessary equations for the design are derived. All of this is complemented by the calculation of line losses in semiconductor elements and the design of voltage and current regulator. The last two parts deal with the description of the created models in Simulink and control program in a graphical environment. The created program is able to edit and recalculate parameters of the converters, calculate line losses on semiconductor components, perform a model simulation or open it.