The large dispersion of flexibility providing units among all voltage levels of the power system requires large, comprehensive grid models to assess the impact of flexibility usage during system operation. For this reason, aggregation methods are used to enable distributed concepts. This paper proposes a decentralized and hierarchical approach for the practical implementation of coordinated, vertical flexibility provision across multiple voltage levels, with the feasible operation region as an interface between adjacent grids. A top-down disaggregation process, based on a linear OPF model, is introduced in this paper. It is used during a flexibility provision request to distribute the requested operation point change to the underlying flexibility providing units. The objective function for this disaggregation process is chosen according to operational aspects. Both implemented methods for aggregation and disaggregation of flexibility consider grid constraints. The proposed concept is field tested during regular grid operation in a real distribution system. The results of the conducted studies show that accurate system operation in the more volatile distribution systems requires shorter time intervals compared to the transmission system. Good results with small control deviations were achieved by using time intervals of one minute for the MV and ten seconds for the LV level.

So far, several simulation methods for the aggregation of distributed flexibility providing units have been developed, which allows an estimation of the flexibility potential in distribution systems. However, the accuracy of the result in a real grid as well as the particularities of flexibility deployment as an ancillary service have yet to be demonstrated. Therefore, a holistic approach including the prediction, online aggregation and activation of flexibility was integrated into an existing grid monitoring system. Based on real grid measurements, the uncertainty of flexibility usage can be evaluated in a field test. The results prove, that a linearized OPF based aggregation method is suitable for use in grid operation. However, due to the stochastic behavior of LV-grids, a valid prediction of the flexibility potential cannot be guaranteed. Moreover, fluctuations in the flexibility provision may prevent a constant setpoint for a grid in a typical time interval of e.g. 15 minutes.

DOI: 10.1049/icp.2021.2085 ISBN: 978-1-83953-591-8 Location: Online Conference Conference date: 20-23 September 2021 Format: PDF XXXXX Inspec keywords: power grids; power system measurement; optimisation; power system state estimation; least squares approximations; load (electric); distribution networks Subjects: Optimisation techniques; Power system measurement and metering; Other topics in statistics; Distribution networks

The ongoing trend towards active distribution systems formulates a need to know the current state of the system. This information can be generated through a distribution system state estimation (DSSE). The accuracy of DSSE usually suffers from the lack of available measurements. In this paper a three-phase DSSE is performed for a real low-voltage system in Germany. The accuracy of the DSSE is evaluated based on the differences between the measurements and estimation results. Different measurement configurations are analysed, in order to determine a minimal measurement effort required for a given accuracy. Additionally, the optimal placement of new measurement devices is investigated. The results show that the performed DSSE is accurate enough to judge whether the operation of an unmeshed low-voltage system is within its thresholds or not.

In smart grids different assets and devices need to be able to communicate with each other and preferably with the grid operator as well. There are different technologies available to establish such a communication, with certain advantages and disadvantages over one another. In the research project “Smart Grid Demonstrator Sonderbuch” two different communication technologies are implemented and operated in parallel: wireless broadband LTE and wire-based broadband power line communication (PLC). In this paper, an overview of the scope and objectives of the project are presented; from which requirements for the communication infrastructure were derived. Comparisons of the operation of both communication technologies were drawn, regarding their usability and reliability for real-time applications. Overall, broadband PLC shows better performance in terms of networklatency, while LTE proofs to be less susceptible to short-term interruptions resulting in a higher overall reliability.

The aim of the presented work is to develop an approach for the proper assessment of pseudo measurements for a three-phase distribution system state estimation under the conditions of a real medium voltage grid. The evaluation is based on the validation of the voltage estimation error, which describes the difference between the calculated node voltages determined by the distribution system state estimation and the actually measured node voltages, as a measure of the quality of the calculated results. By replacing real measurement values with pseudo measurements, the influence on the quality of the three-phase distribution system state estimation for incomplete measurement acquisition will be investigated. For this purpose, customizable synthetic profiles are used to replace real power values in this work.

Emerging smart-grid applications in low-voltage systems generate a need to forecast not only aggregated load profiles, b ut i ndividual, c onsumer s pecific pr ofiles in a high temporal resolution. These load profiles a re n oisy a nd volatile, which makes forecasting them challenging. A short term load forecasting (STLF) model, based on Markov chains, is presented in this paper. It is designed for continuous training during operation and can be pre-trained, making it universally applicable. Its performance is evaluated at a synthetic dataset of individual households. The predictive accuracy of the model is investigated for different forecasting resolutions and levels of load aggregation. The results show that the general characteristics of individual, high resolution load profiles i s m odelled reasonably well, outperforming a naive prediction, utilizing a scaled standard load profile.