Research

Modal Analysis

Modal analysis is a mathematical tool used to study the dynamic properties of various physical systems in the frequency domain. It is a method for analysing, measuring, and calculating the dynamic responses of different (excited) systems.

In power systems, modal analysis is primarily applied to small-signal angle stability studies. A key research question is whether its application can be extended to assess the impact of resonant conditions on harmonic propagation within the grid.

The objective of our research is to explore the potential of modal analysis for investigating and mitigating resonance phenomena in power systems.

Reference Models of LV and MV Networks

The operation of networks with distributed generation and active users can be simulated using models implemented in various simulation environments. For accurate analysis of the impact of different consumers on a specific network segment, that segment must be precisely modelled and its actual operating conditions known. In practice, however, detailed network data are often unavailable.

Moreover, assessing the general impact of emerging technologies on the entire network is challenging, as the analysed segment may not be representative. For this reason, new technologies are frequently evaluated using reference (representative) network models, allowing results to be generalised to the wider distribution system.

Machine learning methods, such as clustering and classification, are often used to identify representative networks. This requires a dataset of networks described by selected attributes (features), such as total feeder length, overhead line length, transformer rated power, and the share of residential consumers. After preprocessing and selecting an appropriate clustering method, networks are grouped based on similar characteristics. Representative networks are then defined for each group, enabling simulation results to be extrapolated more reliably to the overall distribution network.

Impact of Electric Vehicles on the Grid

The transport sector accounts for a significant share of global CO₂ emissions, making its decarbonisation a key priority worldwide. In recent years, we have witnessed an accelerated transition from internal combustion engine vehicles to plug-in electric and hybrid vehicles, which rely on electricity from the grid.

A growing number of electric vehicles (EVs), particularly due to high charging power demands, will significantly affect the operation of transmission and distribution networks. Uncontrolled charging of large EV fleets may cause overloading of transformers and lines, unacceptable voltage drops, and ultimately the need for costly grid reinforcements.

These challenges can be mitigated through the implementation of smart charging concepts, which adapt charging processes to real-time grid conditions, reducing violations of operational limits.

When simulating EV impacts, uncertainties arise from charging power levels, battery capacity, user behaviour, and driving patterns. The availability and appropriate spatial placement of charging infrastructure will also play a crucial role in enabling the large-scale adoption of electric vehicles.

Distribution Network State Estimation

Distribution networks are facing increasing challenges due to higher integration of distributed generation, heat pumps, electric vehicles, and the need for greater system flexibility. This requires enhanced system observability as a prerequisite for improved controllability.

System observability depends directly on the availability and accuracy of measurement data. Due to technical and economic constraints, existing metering infrastructure often does not provide sufficient observability. In such cases, the required level can be achieved through state estimation.

A state estimation algorithm calculates the most probable system state based on available measurements and the mathematical network model. Although this method has been used in transmission networks for decades, it shows promising results when applied to distribution systems of various types.

While the fundamental principles of state estimation are similar in transmission and distribution networks, its design and implementation in distribution systems differ significantly due to structural and operational distinctions. The primary challenge lies in ensuring robustness — enabling fast and reliable state estimation across diverse distribution network configurations.