In this contribution, measurement data of phase, neutral, and ground currents from real low voltage (LV) feeders in Germany is presented and analyzed. The data obtained is used to review and evaluate common modeling approaches for LV systems. An alternative modeling approach for detailed cable and ground modeling, which allows for the consideration of typical German LV earthing conditions and asymmetrical cable design, is proposed. Further, analytical calculation methods for model parameters are described and compared to laboratory measurement results of real LV cables. The models are then evaluated in terms of parameter sensitivity and parameter relevance, focusing on the influence of conventionally performed simplifications, such as neglecting house junction cables, shunt admittances, or temperature dependencies. By comparing measurement data from a real LV feeder to simulation results, the proposed modeling approach is validated.

With decreasing rotational inertia from traditional synchronous generation, frequency is becoming more volatile. Hence, a closer assessment of frequency-dependent load behavior is required. This paper presents an experimental investigation for residential loads, including sixteen typical domestic appliances. The perturbation-based method (range 50±1 Hz) is applied to determine frequency sensitivity values as parameters for commonly used exponential and linear frequency-dependent load models. Subsequently, these models with their identified parameters are compared with the measured load behavior of four representative devices during dynamic and steady-state frequency variations. The comparison shows a significant increase in the reconstruction error beyond the 50±1 Hz range, which suggests that the model parameters identified for a small frequency range may not be the most effective for larger frequency variations. The experimental study presented in this paper also includes a PV inverter as a representative of the increasing number of distributed generator and storage systems. The frequency-dependent behavior of such systems is defined in grid codes in some countries (e.g. VDE-AR-N 4105 in Germany). The measured response from the PV inverter under test shows that there is a non-negligible response delay time, and thus modeling approaches must be upgraded in the future.

The power-to-voltage sensitivity of a load describes how power consumption depends on changes in voltage magnitude. It is a critical system characteristic used for power system control actions such as conservation voltage reduction and voltage-led load management. Voltage sensitivity depends on the operating state of loads, which can vary over time. However, the time-varying characteristic of voltage sensitivity is often ignored in power system analysis and has been considered as a constant parameter. This paper seeks to contribute to filling this gap by experimentally analyzing the time-varying voltage sensitivity profiles of 21 common residential appliances with a sub-minute time scale resolution, conducted in a 60 m ² apartment-like laboratory. The obtained sensitivity profiles, provided as an open-access dataset, can be used as building blocks that allow a flexible synthesis of voltage sensitivity profiles for complex aggregated residential loads. In addition to the individual load profiles, a typical daily power profile for a 2-person apartment was experimentally reproduced to study the voltage sensitivity profile of an aggregated residential load.

In recent years, the electricity supply has become more volatile, and advanced real-time controllers are needed to manage the grid safely. Demand-side management represents a promising solution, where regulating load consumption through controlled voltage variations offers a valuable approach, which can be applied using power electronics actuators. This approach relies on understanding how power consumption reacts to changes in voltage magnitude or frequency. One proposed method is perturbation-based load sensitivity identification, which introduces controlled perturbation into the grid, for instance through a Solid-State Transformer, and calculates load parameters via power measurements. However, existing methods often require synchronization with the perturbation actuator and lack resiliency to noise or uncorrelated power variations, limiting their practical applicability. This paper proposes a novel approach for perturbation-based load sensitivity identification, utilizing a pre- and post-filtering process. This method has been tested under realistic grid conditions, with autonomous computation of the load sensitivity, triggered by variation-based perturbation detection. It offers more global and flexible control possibilities, and eliminates the need for complex communication layers.

The increase in renewable power generation leads to the need for more controllability on the demand side. The integration of power electronic devices, such as Smart Transformers, brings more flexibility into the modern grid. Demand-side management has large potential, but the lack of grid information limits the realization of precise control. Thus, there is a renewed interest in investigating the power-to-voltage and power-to-frequency sensitivity of the load in real-time. The exponential load model is a commonly used model to describe the load dependency on voltage and frequency. As a non-linear equation, the load model can be determined by linearization or by using an iterative algorithm. This work compares two load sensitivity identification methods: the online load sensitivity identification method, using the linearization approach, and the Newton method, using an iterative algorithm. The variance in the distribution of the load sensitivity identification results is studied, which is an important feature of precision. The reliability and accuracy of the results are also analyzed by using the calculated parameters to reconstruct the power signals, which are compared with the power measurements.