Buchta, Eva (2024)
Uncertainty-Aware Distribution System State Estimation - A State Estimation Method for Medium Voltage Grids Using Heterogenous Measurement Input.
Technische Universität Darmstadt
doi: 10.26083/tuprints-00027467
Dissertation, Erstveröffentlichung, Verlagsversion
Kurzbeschreibung (Abstract)
In Germany, as in many other countries, the “Energiewende” is a central political and social goal to counteract climate change. To achieve this goal, dependence on fossil fuels must be reduced and a transition to a sustainable energy economy promoted. However, this paradigm shift towards renewable energies, decentralized energy sources, and electrification of different sectors significantly impacts the distribution grids’ structure and strains. Integrating decentralized, primarily volatile energy sources, such as solar and wind power plants, leads to a locally and temporally fluctuating electricity feed in. In addition, new consumers, such as electric vehicles and heat pumps, place heavy strains on the lower voltage levels. The distribution grid operators are now faced with the task of keeping the power supply stable despite highly variable feed-in and load. In order to continue to ensure the reliability and efficiency of the power supply, the integration of technologies for real-time monitoring and analysis of the status is crucial. By using innovative monitoring that provides early warning of critical system conditions and automated control applications, existing grids can usually continue to operate safely for a longer period without needing immediate grid expansion. In this thesis, a state estimation method for medium-voltage grids is developed. The aim is to estimate the current state of the power system as accurately as possible. The focus here is on detecting limit violations, such as voltage bands or thermal limit currents, and on the modeling of uncertainties. Since medium-voltage grids are usually only equipped with a scarce measurement infrastructure, additional input, usually only available historically, is used as background information for the state estimation. This input has significantly higher uncertainties than the precise real-time measurements. Therefore, it is particularly important in medium-voltage grids to apply state estimation methods that consider these uncertainties in the output assessment. Based on this requirement, a probabilistic state estimation method is developed that is suitable for practice-relevant measurement availability scenarios. The basic algorithm of the developed state estimation method is based on the Bayes’ rule. This algorithm was extended in the present work by corresponding modules in order to fulfill the analyzed requirements for state estimation methods for medium-voltage grids. These extensions include evaluating the probability distribution regarding critical system states and parameterizing the statistical properties of loads using the available measurement information. Annual simulations of a representative German medium-voltage grid are used for the evaluation. With the developed method, it is possible to estimate the probability of critical system states. The estimation takes less than one second, and the critical system states can be reliably identified with a detection rate of over 90 %. The method also includes classification and output of warning and alarm stages, which provides an early warning of bottlenecks in the grid. In order to be practicable for field usage, the method places particular emphasis on a realistic assumption of the temporally available input data in medium-voltage grids.
Typ des Eintrags: | Dissertation | ||||
---|---|---|---|---|---|
Erschienen: | 2024 | ||||
Autor(en): | Buchta, Eva | ||||
Art des Eintrags: | Erstveröffentlichung | ||||
Titel: | Uncertainty-Aware Distribution System State Estimation - A State Estimation Method for Medium Voltage Grids Using Heterogenous Measurement Input | ||||
Sprache: | Englisch | ||||
Referenten: | Niessen, Prof. Dr. Stefan ; Hanson, Prof. Dr. Jutta | ||||
Publikationsjahr: | 10 Juni 2024 | ||||
Ort: | Darmstadt | ||||
Kollation: | xvi, 119 Seiten | ||||
Datum der mündlichen Prüfung: | 2 Mai 2024 | ||||
DOI: | 10.26083/tuprints-00027467 | ||||
URL / URN: | https://tuprints.ulb.tu-darmstadt.de/27467 | ||||
Kurzbeschreibung (Abstract): | In Germany, as in many other countries, the “Energiewende” is a central political and social goal to counteract climate change. To achieve this goal, dependence on fossil fuels must be reduced and a transition to a sustainable energy economy promoted. However, this paradigm shift towards renewable energies, decentralized energy sources, and electrification of different sectors significantly impacts the distribution grids’ structure and strains. Integrating decentralized, primarily volatile energy sources, such as solar and wind power plants, leads to a locally and temporally fluctuating electricity feed in. In addition, new consumers, such as electric vehicles and heat pumps, place heavy strains on the lower voltage levels. The distribution grid operators are now faced with the task of keeping the power supply stable despite highly variable feed-in and load. In order to continue to ensure the reliability and efficiency of the power supply, the integration of technologies for real-time monitoring and analysis of the status is crucial. By using innovative monitoring that provides early warning of critical system conditions and automated control applications, existing grids can usually continue to operate safely for a longer period without needing immediate grid expansion. In this thesis, a state estimation method for medium-voltage grids is developed. The aim is to estimate the current state of the power system as accurately as possible. The focus here is on detecting limit violations, such as voltage bands or thermal limit currents, and on the modeling of uncertainties. Since medium-voltage grids are usually only equipped with a scarce measurement infrastructure, additional input, usually only available historically, is used as background information for the state estimation. This input has significantly higher uncertainties than the precise real-time measurements. Therefore, it is particularly important in medium-voltage grids to apply state estimation methods that consider these uncertainties in the output assessment. Based on this requirement, a probabilistic state estimation method is developed that is suitable for practice-relevant measurement availability scenarios. The basic algorithm of the developed state estimation method is based on the Bayes’ rule. This algorithm was extended in the present work by corresponding modules in order to fulfill the analyzed requirements for state estimation methods for medium-voltage grids. These extensions include evaluating the probability distribution regarding critical system states and parameterizing the statistical properties of loads using the available measurement information. Annual simulations of a representative German medium-voltage grid are used for the evaluation. With the developed method, it is possible to estimate the probability of critical system states. The estimation takes less than one second, and the critical system states can be reliably identified with a detection rate of over 90 %. The method also includes classification and output of warning and alarm stages, which provides an early warning of bottlenecks in the grid. In order to be practicable for field usage, the method places particular emphasis on a realistic assumption of the temporally available input data in medium-voltage grids. |
||||
Alternatives oder übersetztes Abstract: |
|
||||
Status: | Verlagsversion | ||||
URN: | urn:nbn:de:tuda-tuprints-274675 | ||||
Sachgruppe der Dewey Dezimalklassifikatin (DDC): | 600 Technik, Medizin, angewandte Wissenschaften > 621.3 Elektrotechnik, Elektronik | ||||
Fachbereich(e)/-gebiet(e): | 18 Fachbereich Elektrotechnik und Informationstechnik 18 Fachbereich Elektrotechnik und Informationstechnik > Technik und Ökonomie Multimodaler Energiesysteme (MMES) |
||||
Hinterlegungsdatum: | 10 Jun 2024 11:36 | ||||
Letzte Änderung: | 11 Jun 2024 13:15 | ||||
PPN: | |||||
Referenten: | Niessen, Prof. Dr. Stefan ; Hanson, Prof. Dr. Jutta | ||||
Datum der mündlichen Prüfung / Verteidigung / mdl. Prüfung: | 2 Mai 2024 | ||||
Export: | |||||
Suche nach Titel in: | TUfind oder in Google |
Frage zum Eintrag |
Optionen (nur für Redakteure)
Redaktionelle Details anzeigen |