Static VAR Generators (SVGs) are power electronics devices used to compensate unbalanced currents and correct power factor. SVGs can be used in networks with neutral connection if they use a four-wire topology. The main topologies for two-level three-phase converters are the four-leg four-wire (4L4W) converter and the three-leg converter with split capacitor, a.k.a. three-leg four-wire (3L4W) converter. The selection of the topology and sizing of the dc capacitor are crucial aspects in the SVG design. However, in the literature the dc capacitor is usually sized by numeric simulations that limits the comprehension of its sizing. In this paper, a methodology for the sizing of the dc capacitor of four-wire SVGs is proposed. This methodology is based on the calculation of the ripple of the dc-link voltage using analytic expressions, when the SVG compensates negative-and zero-sequence currents. The proposal simplifies the sizing of the power stack, and helps to select the most adequate topology depending on the application. The contributions have been validated by using numeric simulations in MATLAB/Simulink to design a 15 kvar SVG.

The rapid growth in electricity generation from renewable energy sources represents a major challenge to maintaining the stability of power systems. In this scenario, virtual power plants (VPPs) represent a suitable solution for providing ancillary services to the grid. However, to achieve this goal, renewable energy sources and storage units must be adequately coordinated. In this article, a secondary controller for VPPs that allows the provision of frequency response services is proposed. This controller takes into account the resource availability of the VPP elements (wind, sun irradiance, energy stored, etc.) to calculate the power references for each unit. The effectiveness of the proposed control strategy is tested using a simplified scenario in which a VPP is connected to the transmission network. The VPP consists of two wind power plants, a solar photovoltaic power plant, an energy storage system, and local loads. Four case studies are analysed that include sudden changes of the VPP load, impact of external network, and variations of the power generated by the VPP using realistic generation profiles. Numerical simulation results obtained in Matlab/Simulink are used for validation purposes. In all the cases, the frequency is restored to its nominal value while the VPP power exchange converges to the reference set-point.

In recent years, power electronic converters have been increasingly used for integration of different generation technologies to electricity networks. However, due to their fast control loops and non-ideal grid conditions, these devices are prone to interact with each other. This can lead to an unstable grid operation with possible catastrophic results. In order to study such interactions between devices before commissioning of new power plants, power hardware in the loop (PHIL) techniques have been increasingly used. However, emulation of complex grid scenarios is not straightforward and requires in-depth knowledge of the emulated grids and dedicated hardware and software. In this paper, the emulation of complex grid scenarios in a PHIL platform is presented. The proposed methodology can be used to emulate not only the dynamics of power devices but also the primary, secondary and tertiary control levels. Four complex scenarios are described and their implementation discussed, with special emphasis on practical details that are commonly not explained in the literature. Results show that the low-frequency dynamics of diesel generators, microgrids, electrical systems and local-power networks can be effectively emulated in the PHIL platform. All the scenarios are validated experimentally.

The following topics are dealt with: power grids; invertors; distributed power generation; DC-DC power convertors; voltage control; control system synthesis; electric current control; power convertors; machine control; switching convertors.

Most electronic devices are connected to the grid by using power electronics converters and, in order to reduce the filter size and cost, novel filter topologies are being proposed in the literature. For high-power applications, LTCL filters are an attractive solution since the size of inductors can be reduced while maintaining a low switching frequency. However, this type of filter requires additional passive damping elements unless an active damping strategy is included in its control system. In this paper, the hardware and control system design of a VSC connected to the grid via a LTCL filter is addressed, with emphasis on the damping problem. It is shown that adequate damping can be provided by applying two different strategies. The first one is by changing the hardware of the converter (without additional passive elements), while the second one is by adding a compensator in series with the current controller. The proposed design methodology and control system improvements were tested in a simulation of a 15 kW VSC connected to the grid via a LTCL filter.

Two different methods for implementing inertial damping on a Virtual Synchronous Machine (VSM) and their potential for attenuating power system oscillations when utilized in a VSC HVDC terminal are investigated in this paper. As a reference case, the VSM is considered with only a frequency droop providing damping in the virtual swing equation. Then, the effect of damping based on high-pass filtering of the virtual speed is compared to damping based on high-pass filtering of the measured grid frequency. A simplified model of a power system with two equivalent generators and a VSC HVDC terminal is introduced as a case study. Analysis of the small-signal dynamics indicates that damping based on the VSM speed has limited influence on the power system oscillations, while improved attenuation can be obtained by introducing damping based on the locally measured grid frequency. The presented analysis and the operation of the proposed VSM-based damping strategy is validated by numerical simulation of a 150 MVA VSM controlled VSC HVDC terminal connected to a dynamic model of the assumed grid configuration.

The emulation of synchronous machines is an attractive solution to improve the integration of distributed energy resources in weak grids. In this sense, synchronverters have been widely analysed in the literature, but their digital implementation has been seldom studied. This paper presents the discrete-time implementation of the control algorithms for a battery-supported synchronverter in a weak grid. First of all, the formulation is described in a synchronous reference frame and a simple method to adjust the parameters is described. Some modifications are proposed to the classical synchronverter that will be implemented in its incremental form. By using this implementation method the control system a) becomes easier to implement, b) wind-up problems are avoided, and c) command signals are easier to saturate. Since the control system is non-linear, these features are of a particular relevance. All the control developments are tested experimentally in a 15 kVA converter with a 47.5 kWh Li-ion battery connected to a weak grid.