Energy losses are an essential part of a power grid company’s costs. From anenvironmental and economic perspective, it is important to reduce these costs. Toknow where efforts should be deployed, it is important to have good knowledge ofwhere in the network losses are situated. When creating network tariffs, there is alsonecessary to have a correct distribution such that the network customers that use acertain part of the network, only pays for the network losses in this area. A customerthat is connected to the mid voltage network should therefore not have to pay forlosses in the low voltage grid.Mälarenergi elnät AB demand a deeper investigation to see if their reported net lossesfor Energimarknadsinspektionen, EI is reasonably accurate or completely misleading.Therefore, this thesis was performed.Today advanced software is used which retrieve measurements from the power gridaround cities. This type of software is monitoring networks in order to detectpossible failures and then fix those errors when necessary. These measurements areused in the thesis to identify the size of the losses in the network. The network isdivided into smaller areas with each area analysed separately. When all areas haveundergone deeper analysis, the results are compared and a final result will beobtained at the end.According to Energimarknadsinspektionen, Mälarenergi elnät energy losses are 2,96%. Losses according to the analysis in this thesis amounts to 3,11 %.

An important process for grid companies in order to reduce power losses in electricity distribution grids is to figure out where and how they appear. This report describes the factors that cause power losses in low voltage distribution grids and methods for how they could be measured and analyzed. Measurement data for consumed and injected energy were collected for twelve low voltage distribution grids owned by Gävle Energi AB in order to analyze the size of the power losses and how the measurement uncertainty affects the collected data. Then models of three of the grids were created in MATLAB in order to analyze the data and the power losses were calculated to be compared to the collected data. Curve fitting were also used as a method in order to analyze the power losses from collected data. The results show that the losses in the studied grids are low. The measurement uncertainty is also low when losses are studied for long time periods, but high when losses are studied for short periods such as single hours. The calculated power losses from the MATLAB calculations seem to be lower than the measured power losses, which could indicate that a more complex model is needed for these kinds of studies. With a good interaction between the grid information system and the system that is storing measurement data it should be possible for grid companies to chart their power losses and with an accurate model of the grids the power losses can be calculated theoretically which should make it possible to analyze the causes of power losses in low voltage distribution grids.

This study was carried out on behalf of Pite Energi with the guidance of Rejlers. The assignment was to develop a simulation model over part of Pite Energi's power grid. This will allow calculations of losses and capacity limits in the network, which then can form the basis for future efficiency improvements. The work has also investigated the possibility of a backup line for the voltage level of 40 kV and how the energy losses are affected under the periods of high and low load in the 10 kV grid, respectively. The grid is constructed with two 40 kV lines, ordinary and reserve, which are parallel to each other. The voltage in these lines is transformed down to two different voltage levels of 10 and 20 kV, respectively. Data has been collected on the exracted load at 10 kV as well as electricity production from 35 MW wind power at 20 kV. This data was collected by Pite Energi for all hours of 2019 and used as input in the thesis work for simulations and calculations of the grid. The investigated 40 kV grid shows a reduced loss as load increases in the 10 kV grid as long as the increase follows the load variations for 2019. Compared to the losses for the base case, reduced losses are obtained in 40 kV networks with loads up to 16-18 MW. The report also shows that there are opportunities for further energy savings through regulating production, or energy storage in the wind farm. The difference in network losses between high and low load times is very small. This is largely due to the large difference between production and consumption in the grid. With the load variations measured for 2019, it shows that the limitations in the 40 kV network are primarily due to the thermal limits for both ordinary and reserve. The result also shows that today there is a good margin to the network's capacity limits and will most likely be able to handle any load increases that may come in the near future.

Today’s electrical grid companies understand that better knowledge of and information about their grid can help realize a smart grid network. One way to gain better knowledge and information about the low voltage network is to install measuring equipment on the substation transformer´s secondary low voltage side that measures the low voltage network losses for that specific substation. By doing so, grid companies can, due to the calculation of the low voltage network losses, identify and address issues such as illegal power dissipation, unmeasured power outtakes, and unsatisfied technical documentation for the substations low voltage network. During 2012, Mälarenergi Elnät AB had a total electrical network loss (both medium and low voltage network losses included) of three to four percent of their total distributed electrical volume. Converting this energy volume into a monetary figure this accounts for a loss of around 30 million Swedish crowns. To address Mälarenergi Elnät AB’s issue about their low voltage network losses, Mälarenergi Elnät AB asked me to perform my civil engineering master thesis around this subject. Mälarenergi Elnät AB wanted me to perform measurements in five pre-existingsubstations. I had seven EDMI MK10A CT electrical meters for substation measurements installed in five pre-existing substations in the surrounding area of Västerås in order to investigate the potential costs and benefits of increased measurement in the low voltage network system. The total cost of my substation measurement installation reached 35 000 SEK or around 5 000 SEK per installed electrical meter. The calculated low voltage network losses from my measurements in the selected pre-existing substations showed that 75 % of the studied low voltage networks had losses in the range of zero to five percent. From the calculated results I also identified several benefits of measurements in pre-existing substations, including improved calculation of low voltage network losses, increased awareness of load levels so that maintenance could be more efficiently scheduled, and more accurate technical documentation, all of which can lead to increased customer satisfaction. The disadvantages of measurement in pre-existing substations that I identified during my master thesis study were the high installation cost of the meters and the increased demand on human resources during the implementation phase. Overall, my economic evaluation of the project showed that the investment was not cost effective, despite the acknowledged benefits of these substation measurements. The overall results suggest that Mälarenergi Elnät AB could benefit from installing meters in new substations, and in some existing substations where specific information about the low voltage network loss is needed for issues such as identifying suspicious illegal power dissipation or unmeasured power outtakes. The high installation cost for substation measurements are, in those cases, motivated by the fact that the substation measuring equipment costs are low compared to a whole new substation or compared to the fact of finding unmeasured outtakes, illegal powerdissipations etc.

The Swedish Energy Market Inspectorate, Ei, has an assignment toregulate the revenue cap for the network operators in Sweden. As apart of this regulation there is an incentive for the networkoperators to lower their network losses. This incentive wereintroduced 2016 as a norm level based upon the network operatorsown historical values of network losses. If the operators manageto lower their network losses compared to their own norm levelthey are given a higher return on their asset base. The aim of this thesis was to evaluate the possibilities to createa function form calculations of norm levels based upon theconditions under which each network operator perform and ifpossible direct this function. The conditions for creating suchfunction were investigated by analysing data of differentparameters in the Swedish networks to find which of them thatcould describe the losses through a function. The number of customers per kilometre line was found to be thesingle most important parameter to describe the network losses. Bycombining the number of customers per kilometre line with aparameter for the part of the energy that is feed out at highvoltage the function could describe the losses even better. Usingthis function will give the network operators more long timeincentives to lower their network losses than the norm levelsbased upon their own historical values of network losses.

One step towards a more sustainable energy system is to create a more flexible electrical grid, where increased demand response among electricity consumers can play an important role. A distribution grid owner can encourage their customers to use electricity more evenly distributed during the day by introducing different types of grid fees such as time-differentiated power tariffs. In this master thesis, the theoretical economic impact of a flattened load profile for a distribution grid owner is investigated. Different factors that impact the distribution grid owner’s economy are identified and two are chosen to be quantified; losses in the grid and the fee to the feeding grid. The possibility to save money by avoiding future investments is discussed but not quantified. Models are developed for modifying an existing load profile, calculating the losses associated with a certain grid and load profile, and calculating the value of a lowered subscribed power to the feeding grid. The models are applied on the distribution grid owner Sala-Heby Energi Elnät AB. The results show that with a load profile that is flattened out 100% every day, which is the theoretical best possible scenario, the losses can be reduced with 2.6% which corresponds to a value of 81 000 SEK. By lowering the subscribed power to the feeding grid as much as the highest peak of the year is reduced in the modified load curve, the cost to the feeding grid would be reduced 1.2 million SEK, or 10% of the fee to the feeding grid. In the 20% curve modification case, which would be more realistic to achieve in reality, the losses decreased by 0.9% which corresponds to a value of 29 000 SEK. Furthermore, the fee to feeding grid would be reduced with 0.34 million SEK which corresponds to 2.8% of the total fee. In summary, the theoretical economic saving potential is around 10 times higher for the fee to feeding grid than for grid losses. / Marknadsbaserade styrmedel i bostadssektorn

No description available.

In 2018 the Swedish government introduced a new legislation for the electricity meters installed in the low voltage networks. The legislation includes, among other things, that the electricity meters should be able to measure the current and voltage per phase with a time resolution of one hour or 15 minutes. To meet the new requirements Falu Energi & Vatten have started a rollout of new electricity meters in their low voltage networks. In addition to this, electricity meters have also been installed in a few secondary substations in Falun. The overall purpose of this project was to investigate what the data from the new electricity meters can be used for and what the benefits are with having metering equipment in secondary substations. Data from secondary substations can be useful when dimensioning the grid or when analysing power losses. During the project, data was collected from one secondary substation and the underlying low voltage network. The parameters current and voltage were measured per phase with a time interval of 15 minutes. The data was used in three different investigations. The results from two of them indicated that the network information system that Falu Energi & Vatten uses for dimensioning tends to overestimate the power consumption and the voltage drop between the secondary substation and the low voltage customers. In the last investigation a method for localising non-technical power losses was evaluated. The result indicated that the method was not suitable for the purpose.