The increasing volume of smart electricity meter readings presents a challenge forelectricity providing companies in accurately validating and correcting the associated data. This thesis attempts to find a possible solution through the application ofunsupervised machine learning for detection of anomalous readings. Through thisapplication there is a possibility of reducing the amount of manual labor that is required each month to find which meters are necessary to investigate. A solution tothis problem could prove beneficial for both the companies and their customers. Itcould increase abnormalities detected and resolve any issues before having a significant impact. Two possible algorithms to detect anomalies within these meters areinvestigated. These algorithms are the Isolation Forest and a Autoencoder, wherethe autoencoder showed results within the expectations. The results shows a greatreduction of the manual labor that is required up to 96%.

I detta examensarbete utreds möjligheten att med hjälp av nya smarta elmätare få en ökad inblick i elkvalitén i elnäten samt att ge nätägare möjligheten att automatiskt eller via fjärrsignal koppla bort solcellsanläggningar i de fall de orsakar dålig elkvalité. Motiveringen till studien har varit att en ökad utbyggnad av solcellsanläggningar med tilhörande växelriktare och en trolig framtida ökning av snabbladdare och andra olinjära laster i näten ökar utmaningen för elnätsföretag att klara gällande krav på elkvalité. Samtidigt har utvecklingen av elmätare gjort att de idag har många funktioner som elnätsföretag kan dra nytta av på sätt som tidigare inte varit möjligt. Studien inleddes med en litteratur-studie där information om vad elmätare idag är kapabla till togs fram. Ett praktiskt test genomfördes på en solcellsanläggning där en elmätare installerades och konfigurerades att koppla bort solcellerna vid över-spänning. Studien har visat att smarta elmätare bör kunna användas som ett långsammare, sekundärt skydd som kopplar bort solceller i de fall som växelriktarna brister i sina säkerhetsfunktioner. Elmätarna kommer också kunna ge nätägare en mängd intressant information om elkvalitén. / In this thesis the possibility for smart meters to give utility companies an increased knowledge about the power quality in the grid has been investigated. During the investigation the possibility to use smart meters to automatically or by remote control disconnect solar power units in case of poor power quality caused by the production units has also been researched. The motivation to perform the study was that a future increase in distributed generation, rapid chargers for electric vehicles and other non-linear loads will make it harder for utility companies to meet the demands concerning power quality. At the same time electricity meters has developed over time and are today capable of providing grid operators with functions that has not been possible in the past. The first part of the study focused on obtaining information about what smart meters are capable of today through a literary study. A practical test was conducted on a solar power unit where a smart meter was installed and configured to automatically disconnect the solar power unit in case of over voltage. The study has shown that smart meters could be used as a slower, secondary protection that disconnects the solar power units in cases where the solar inverters safety functions fails. The smart meters will also be able to give grid operators a large amount of information about the power quality in the grid.

This work deals with setting of dependability of static electricity meters. The first two chapters deal about electricity meters and dependability in general than there are introduced three possible ways for getting dependability parameters. The first methode is data collection from the field. The second methode is reliability prediction from component's reliability. The third methode is aging life tests. Conclusion contains evaluation of results and highligting of importance for practical usage.