The goal of this project is to construct a machine learning algorithmthat improves over time. This was done by first constructing a datasetthat reflects real world messages, that would simulate receiving emailsfrom two different sources. The data set was constructed by combiningdata from two different online forums. Two application programminginterrfaces were used to collect and send data to the program. Thedataset was tested on 4 different methods where the best one would beused for the final product. The 4 different methods were: k-nearestneighbors, adaptive boosting, random forest and artificial neuralnetwork. All the above methods were tested and tuned to achieve the bestaccuracy. From the result it became clear that the artificial neuralnetwork outperformed the other methods by a large margin and would bemost suited for the final product. The final product was an algorithmthat would improve over time. This was achieved by using a feedback loopon the new data that was collected over time from the online forums. Ifthe algorithm was sure that a new datapoint was the right class it wouldincorporate it into the dataset and over time the dataset would growlarger and the algorithm would adapt to new data and trends. The finalresult became a growing dataset that started on a 1000 data points andended up at 8464 data points, where the total amount ofmisclassification ended up at 74.

Electrical bushings are a type of hollow or solid conductor withinsulation designed to allow a conductor to pass through a conductingbarrier without making electrical contact and are very important forsafe transportation of electricity. The bushings vary in size, but allbushings have a solid or hollow conductor.This project aims to investigate which resistivities of the siliconerubber, in combination with different air conditions for the airsurrounding the bushing (dry, average and humid air) and theirrespective resistivities, to see which combinations allow for anelectrical field that allows the bushing to safely work without beingdamaged. The different air conditions are an important factor since theyall correspond to different absolute humidities present in the air,which have a direct impact on the strength of the electrical fieldsurrounding the bushing. Since this cannot be done by hand, a computersoftware called COMSOL Multiphysics will be used. COMSOL is aMultiphysics software, meaning one can simulate many types of physics atonce. With the help of this software, and a model provided by HitachiEnergy, results could be found stating that dry air overall is the bestcondition of air for the bushing, followed by average (not either dry orhumid) air and lastly humid air.

To face climate change and transform the electricity supply to an environmentally friendly generation, wind plays an important role. Due to a yearly increase in installed wind power turbines, in the European Union, the need for maintenance increases as well. For reducing the maintenance times and, with that, the standstill time and resulting economical losses, the time for troubleshooting must be reduced. This work aims to show that the troubleshooting process of wind turbines can be reduced to a minimum with the automation. This can be reached by creating a scatter plot of the active power over the wind speed curve and investigating the data points where the turbine is not performing as it should. The data is extracted from a wind farm located in Finland for the wind year 2021. The methodological approach taken in this study is to build a normalised threshold power curve and compare it to monthly binned power curves of two selected turbines. The deviation between the threshold and the monthly power curve is investigated, and the months with a high deviation are chosen for further analysis, which includes the separation of the outlier data into four different categories. The outlier in bins with a higher deviation than 5 % are selected. The four categories are further inspected, and the reasons for the curtailments are extracted and analysed. In summary, these results show that the analysis of curtailment reasons based on a scatter plot of the active power of a wind turbine is possible. Moreover, the troubleshooting process can be reduced in time. Due to practical constraints, this work cannot provide an analysis with a threshold power curve built with data from more than one year. This makes the results less objective since fluctuations, which can occur during only one year, cannot be minimised.

Εlectric vehicles are replacing the internal combustion engine vehicles rapidly and they will dominate the market completely in the next years. The amount of energy and power needed to support this new technology is huge. This will increase the already high electricity demand of our societies. The electric vehicles can provide a solution by using them to transfer energy to any other vehicles or infrastructure in combination with electricity management. This can be achieved by controlling the electric vehicle chargers and by knowing the exact consumption of the other vehicles or infrastructures. In Dansmästaren, Uppsala, there is a parking garage with 30 Charge Amps Aura charging stations. The same type of charger has been used in order to examine if it is possible to extract and update data through programming, as well as its functions regarding Vehicle-to-everything (V2X). This thesis presents two Python scriptswhere the first is used to update different functions of the charger and the secondto get high resolution electricity data and the energy consumption of the charger.The collected data is stored in two MySQL database every 30 seconds for future use. The data that can be updated by the user immediately, from anywhere and at any time. Similarly, the data collection has shown that different charging patterns exist and they can be observed by using the data that are generated and saved in the databases.

This master’s thesis describes the design, implementation and testing of an active antenna intended for use with global navigation satellite systems. The active antenna is composed of two major parts, a dual-band circular patch antenna and a dual-band low-noise amplifier. To streamline the design process, a modular solution was adopted. This enabled the functionality of every part in the signal path to be verified before the final active antenna was designed. A practical method to develop dual-band stacked circular patch antennas, along with a systematic way to tune the resonant frequencies and impedance of the antenna, is also presented. Testing of the antenna in realistic scenarios shows that the active antenna performs as expected and predicted by simulations. / <p>Examensarbetet är utfört vid Institutionen för teknik och naturvetenskap (ITN) vid Tekniska fakulteten, Linköpings universitet</p>

When the wind blows and wind turbine generators harvests the kinetic energy and trans- forms it to electrical power, there is a need for predicting how much power that will be dispatched from the turbines. Even the most perfect computer model with high computa- tional power could not model the beauty of the forces of nature and we must accept some degree of forecasting error in the predicted power output due to the inherently stochastic patterns in the atmosphere.  This project set out to investigate the main reasons and factors that impacts the forecasting error related to wind power assets on Gotland. From theory and the performed case study, wind speed is the strongest predictor of wind power production, to claim anything else would be severely inaccurate. However, the main predictors of wind power prediction are summarized from a literature study, extracted from a weather model and tried in a case study for the wind farm Stugylparken on Näsudden, Gotland. Three different prediction methods were tried and the ensemble trees model was the best model by the evaluation metrics that was chosen. The second-best performing model was the artificial neural network, and prediction by theoretical power curve performed worse than the standard machine learning methods what was tested in the study. It can be noted that when assessing what model to choose, it depends on how the evaluation is done and which metric is deemed most important. Besides that wind speed will have the most significant impact in all models, forecasting error seem to have correlation to the diurnal cycle. One reason could be land-sea interaction during the day, especially at the period April-September. Higher forecasting errors correlates strongly to periods of a higher mean wind speed and times of varying weather will impact the forecastability and larger errors should be expected. In this project, numerical weather prediction data is used to investigate the forecasting error. A lower error can be seen at the first hours from the model run. This should be expected because it is when we are closest to the initial conditions, in other words, the real world. However, it seems like wind speed and diurnal cycle are more significant than the performance of the numerical weather prediction model in the first 24 hours.  Predicting the future power output of wind assets is expected to be even more impor- tant in the future years due to larger installed capacity. Even with an increase in installed capacity, an over capacity is not wanted and flexibility will be more important. There are challenges, but also opportunity to have a more efficient use of resources in our society and lowering the climate impact that our society has on the planet through a more flexible use of resources.

To evolve and develop your skills as a driver, it is worthwhile to review and reflect on your last drive at a professional as well as beginners level. The purpose of the project is to create such a system that can be used to record a drive by filming from the dashboard of the car and logging data from the vehicle. The summarized log is then used to evaluate a run.  The system consists of two major parts: 1) an OBD-II reading-unit (hardware) that can be plugged in the OBD-II port in the vehicle to read data, and 2) an Android application run on an Android device. The application records the drive as well as a heads-up display (HUD) showing acceleration, GPS-location and some data gathered from the vehicle continuously. The OBD-II port is connected to engine control units in the vehicle. The reading unit follows the commands given by the application, reads the data and then sends them to the smartphone via Bluetooth.  The reading unit consists of a STN-1170 microcontroller which supports a large roster of diagnostics protocols, and a HC-06 Bluetooth module. The unit is plugged in the OBD-II port to get the measurement data and transmits the data to the smartphone.  To validate the system, a CAN simulation unit (CAN simulator) was created that acts as the OBD-II port (or socket) in a vehicle based on OBD standard diagnostics protocol. The simulator was designed as a shield for an Arduino UNO. The Arduino was programmed to act as an OBD-II port in a vehicle supporting the CAN protocol. Utilizing two components: MCP2515, which translates between SPI and CAN, and MCP2551, which prepares the CAN signal for transmission on the physical bus, allows the Arduino to communicate the simulated data through SPI which, in turn, was translated into and transmitted on the physical bus using CAN. The parameters: vehicle speed, vehicle RPM and throttle-position, could be controlled using a set of potentiometers, these parameters were used to validate the system. The system was tested using the testbench with the simulator. The test results have shown that both the system and the simulator work well. The Android application requested parameters such as RPM, speed and throttle positioning while also updating the GPS-location of the vehicle and reading the acceleration using the smartphone’s accelerometer. The OBD-II reading unit received the request from the application through Bluetooth, transferred it to the microcontroller via UART, translated the request into a command in the CAN protocol and sent it to the simulator. At the next command transmitted to the simulator, the reading unit would collect the previous command’s result, which was transferred back to the phone in the reverse order. The information collected from the vehicle was tagged using a timestamp and subsequently logged in a .txt-file.

In power system asset management, component data is crucial for decision making. This thesis mainly focuses on two aspects of asset data: data quality and data availability. The quality level of data has a great impact on the optimality of asset management decisions. The goal is to quantify the impact of data errors from a maintenance optimization perspective using random population studies. In quantitative terms, the impact of data quality can be evaluated financially and technically. The financial impact is the total maintenance cost per year of a specific scenario in a population of components, whereas the technical impact is the loss of a component's useful technical lifetime due to sub-optimal replacement time. Using Monte-Carlo simulation techniques, those impacts are analyzed in a case study of a simplified random population of independent and non-repairable components. The results show that missing data has a larger impact on cost and replacement year estimation than that of under- or over-estimated data. Additionally, depending on problem parameters, after a certain threshold of missing data probability, the estimation of cost and replacement year becomes unreliable. Thus, effective decision making for a certain population of components requires ensuring a minimum level of data quality. Data availability is another challenge that faces power system asset managers. Data can be lacking due to several factors including censoring, restricted access, or absence of data acquisition. These factors are addressed in this thesis from a decision making point of view through case studies at the operation and maintenance levels. Data censoring is handled as a data quality problem using a Monte-Carlo simulation. While the problems of restricted access and absence of data acquisition are studied using event trees and multiphysics modelling.  While the quantitative data quality problem can be abstract, and thus applicable to different types of physical assets, the data availability problem requires a case-by-case analysis to reach an effective decision making strategy. / <p>QC 20210528</p> / CPC5

This paper studies the effects of the Firmware Over-The-Air update process done through Bluetooth Low Energy using a compression-based update solution compared to a non-compressed update solution.   The method that is used in this study is an experimental method using a Nordic semiconductor board as an IoT device. To this, a firmware was sent Over-The-Air and energy consumption along with time was measured using a measuring tool. The firmware was sent compressed and uncompressed. The compression algorithm that was used in the experiment was LZ4. Different sizes of firmware were tested and decompressed through the update process, and by using the measuring tool, the time taken, and average current were measured to get the results of the study.    The result of this study is that time taken increases as firmware gets larger on both compressed and non-compressed firmware, and the increase in firmware file sizes had no effect on the average current. At the same time, including decompression into the FOTA solution increases the average current and time taken, which led to the conclusion that a compressed FOTA update has a higher total time and average current on the hardware setup tested on. It was also concluded that an embedded system with a faster processor, in conjunction with a higher compression ratio, would be needed to reach a threshold value where a compression based FOTA process is preferable.

A dataset consisting of logs describing results of tests from a single Build and Test process, used in a Continous Integration setting, is utilized to automate categorization of the logs according to failure types. Two different features are evaluated, words and log keys, using unordered document matrices as document representations to determine the viability of log keys. The experiment uses Multinomial Naive Bayes, MNB, classifiers and multi-class Support Vector Machines, SVM, to establish the performance of the different features. The experiment indicates that log keys are equivalent to using words whilst achieving a great reduction in dictionary size. Three different multi-layer perceptrons are evaluated on the log key document matrices achieving slightly higher cross-validation accuracies than the SVM. A shallow-and-wide Convolutional Neural Network, CNN, is then designed using temporal sequences of log keys as document representations. The top performing model of each model architecture is evaluated on a test set except for the MNB classifiers as the MNB had subpar performance during cross-validation. The test set evaluation indicates that the CNN is superior to the other models. / Ett dataset som består av loggar som beskriver resultat av test från en bygg- och testprocess, använt i en miljö med kontinuerlig integration, används för att automatiskt kategorisera loggar enligt olika feltyper. Två olika sorters indata evalueras, ord och loggnycklar, där icke- ordnade dokumentmatriser används som dokumentrepresentationer för att avgöra loggnycklars användbarhet. Experimentet använder multinomial naiv bayes, MNB, som klassificerare och multiklass-supportvektormaskiner, SVM, för att avgöra prestandan för de olika sorternas indata. Experimentet indikerar att loggnycklar är ekvivalenta med ord medan loggnycklar har mycket mindre ordboksstorlek. Tre olika multi-lager-perceptroner evalueras på loggnyckel-dokumentmatriser och får något högre exakthet i krossvalideringen jämfört med SVM. Ett grunt-och-brett faltningsnätverk, CNN, designas med tidsmässiga sekvenser av loggnycklar som dokumentrepresentationer. De topppresterande modellerna av varje modellarkitektur evalueras på ett testset, utom för MNB-klassificerarna då MNB har dålig prestanda under krossvalidering. Evalueringen av testsetet indikerar att CNN:en är bättre än de andra modellerna.