To draw reliable conclusions about the thermal characteristic of or a preferential cooling strategy for a lithium–ion battery, the correct set of thermal input parameters and a detailed battery layout is crucial. In our previous work, an electrochemical model for a commercially-available, 40 $\mathrm{Ah}$ prismatic lithium–ion battery was validated under heuristic temperature dependence. In this work the validated electrochemical model is coupled to a spatially resolved, three dimensional (3D), thermal model of the same battery to evaluate the thermal characteristics, i.e., thermal barriers and preferential heat rejection patterns, within common environment layouts. We discuss to which extent the knowledge of the batteries’ interior layout can be constructively used for the design of an exterior battery thermal management. It is found from the study results that: (1) Increasing the current rate without considering an increased heat removal flux at natural convection at higher temperatures will lead to increased model deviations; (2) Centralized fan air-cooling within a climate chamber in a multi cell test arrangement can lead to significantly different thermal characteristics at each battery cell; (3) Increasing the interfacial surface area, at which preferential battery interior and exterior heat rejection match, can significantly lower the temperature rise and inhomogeneity within the electrode stack and increase the batteries’ lifespan.

The increasing penetration of the power grid with renewable distributed generation causes significant voltage fluctuations. Providing reactive power helps balancing the voltage in the grid. This paper proposes a novel adaptive volt-var control algorithm on the basis of deep reinforcement learning. The learning agent is an online-learning deep deterministic policy gradient that is applicable under real-time conditions in smart inverters for reactive power management. The algorithm only uses input data from the grid connection point of the inverter itself; thus, no additional communication devices are needed and it can be applied individually to any inverter in the grid. The proposed volt-var control is successfully simulated at various grid connection points in a 21-bus low-voltage distribution test feeder. The resulting voltage behavior is analyzed and a systematic voltage reduction is observed both in a static grid environment and a dynamic environment. The proposed algorithm enables flexible adaption to changing environments through continuous exploration during the learning process and, thus, contributes to a decentralized, automated voltage control in future power grids.

With increasing changes in the contemporary energy system, it becomes essential to test the autonomous control strategies for distributed energy resources in a controlled environment to investigate power grid stability. Power hardware-in-the-loop (PHIL) concept is an efficient approach for such evaluations in which a virtually simulated power grid is interfaced to a real hardware device. This strongly coupled software-hardware system introduces obstacles that need attention for smooth operation of the laboratory setup to validate robust control algorithms for decentralized grids. This paper presents a novel methodology and its implementation to develop a test-bench for a real-time PHIL simulation of a typical power distribution grid to study the dynamic behavior of the real power components in connection with the simulated grid. The application of hybrid simulation in a single software environment is realized to model the power grid which obviates the need to simulate the complete grid with a lower discretized sample-time. As an outcome, an environment is established interconnecting the virtual model to the real-world devices. The inaccuracies linked to the power components are examined at length and consequently a suitable compensation strategy is devised to improve the performance of the hardware under test (HUT). Finally, the compensation strategy is also validated through a simulation scenario.

Faults and unintended conditions in grid-connected photovoltaic systems often cause a change of the residual current. This article describes a novel machine learning based approach to detecting anomalies in the residual current of a photovoltaic system. It can be used to detect faults or critical states at an early stage and extends conventional threshold-based detection methods. For this study, a power-hardware-in-the-loop approach was carried out, in which typical faults have been injected under ideal and realistic operating conditions. The investigation shows that faults in a photovoltaic converter system cause a unique behaviour of the residual current and fault patterns can be detected and identified by using pattern recognition and variational autoencoder machine learning algorithms. In this context, it was found that the residual current is not only affected by malfunctions of the system, but also by volatile external influences. One of the main challenges here is to separate the regular residual currents caused by the interferences from those caused by faults. Compared to conventional methods, which respond to absolute changes in residual current, the two machine learning models detect faults that do not affect the absolute value of the residual current.

Due to climate targets of the German government, the share of renewable energy in the power grid will be increased and the number of grid participants connected to the low voltage level of the power grid will rise. This leads to new requirements in voltage control, especially in low voltage distribution grids. In order to achieve a stable power grid in future, further development of control strategies is necessary. In this paper, a load recognition concept, which was tested on simulative data in previous work, is further developed to reduce simulation effort. Additionally, the concept is adapted for real hardware influences and active grid participants complicating the recognition task. Thus, the main contribution of this study is the successful application of the methodology within a hardware-based test grid containing a charging electric vehicle. Using a convolutional neural network in a time series classification setting, the recognition rates in this use-case exceeded 99 % while benefiting from an asymmetric charging behavior. Due to these promising results, future voltage control strategies could be supported based on gained information through integration of the presented concept.

The decommissioning of conventional power plants and the installation of inverter-based renewable energy technologies decrease the overall power system inertia, increasing the rate of change of frequency of a system (RoCoF). These expected high values of RoCoF shorten the time response needed before load shedding or generation curtailment takes place. In a future scenario where renewables are predominant in power systems, the ability of synchronous machines to meet such conditions is uncertain in terms of capacity and time response. The implementation of fast power reserve and synthetic inertia from inverter-based sources was assessed through the simulation of two scenarios with different grid sizes and primary reserve responses. As main results it was obtained that the full activation time for a fast power reserve with penetration above 80% of inverter-based generation would need to be 100 ms or less for imbalances up to 40%, regardless of the synchronous response and grid size, meaning that the current frequency measurement techniques and the time for fast power reserve deployment would not ensure system stability under high unbalanced conditions. At less-unbalanced conditions, the grid in the European scale was found to become critical with imbalances starting at 3% and a non-synchronous share of 60%.