Categories
Stability Wind Farms

Grid forming converter in CIGRE 928 benchmark system

For both grid-following (GFL) and grid-forming (GFM) control, the same hardware structure is used, specifically an IGBT-based two-level voltage-sourced converter (VSC) with a 12 MW rating. This converter is connected to the plant via a 0.69/66 kV transformer, with the grid-side reactance of the LCL filter acting as the step-up transformer.

The converter control is configured to operate in grid-forming mode, using vector control in the dq-reference frame. The control system includes the following components: power calculation, power regulation, voltage regulation, current regulation, and pulse-width modulation (PWM). Additionally, an active damping control utilizing feedback from the series reactor current is considered.

Subsystems of grid-forming converter

Anti-aliasing filtering and sampling: The anti-aliasing filter is a second-order Butterworth filter, with a cut-off frequency set to half the sampling frequency. Sampling delays are approximated using a third-order Padé formula.

Park transformation: Grid-forming control is implemented in a reference frame synchronized to an arbitrary input, provided in per unit (p.u.), where the reference frequency (f*) is set to 1. In practical applications, the phase can be provided by a low-bandwidth phase-locked loop (PLL) that slowly tracks the system frequency.

Power calculation: Instantaneous active and reactive power are calculated based on voltage and current measurements taken at the LCL filter's output.

Power regulation: In GFM mode, power synchronization control generates the reference for the voltage angle, while reactive power control sets the reference for the voltage amplitude. Active and reactive power droop characteristics are combined with a simple low-pass filter.

Voltage regulation: The voltage is controlled through a proportional-integral (PI) regulator, ensuring zero steady-state error. Active damping is also incorporated to enhance noise rejection and system stability.

Current regulation: A proportional (P) controller manages the current, as the integral (I) component is unnecessary; the voltage regulator already handles tracking the voltage reference. Adding complexity to the current regulator may adversely affect converter stability.

PWM: The modulation block calculates the switching functions and sends pulse patterns to the converter's gate drivers. The PWM delay is also accounted for in this block.

Note: The proposed GFM control does not include DC voltage regulation. It is assumed that the GFM converter maintains a stable DC link voltage, such as from a battery energy storage system (BESS), so DC voltage control is unnecessary.

Benchmark grid forming converter control block diagram.

Grid-forming converter parameters

The electrical circuit for GFM unit is assumed to be the same as for GFL converter. The control structure is visualized in figure below.

List of parameters in the grid forming converter control system.

References

[1] Ł. Kocewiak, Ch. Buchhagen, R. Blasco-Gimenez, J. B. Kwon, M. Larsson, Y. Sun, X. Wang et al., “Multi-frequency stability of converter-based modern power systems,” Technical Brochure 928, Page(s) 1-147, CIGRE, March 2024.

Categories
Measurements Stability Wind Farms

Grid following converter in CIGRE 928 benchmark system

The primary objective of the benchmark model is to serve as a reference for studying interactions between converters and the grid. It provides a foundation for evaluating small-signal stability analysis methods and instability mitigation techniques. The model includes aggregated grid following (GFL) converters, interconnected via a medium-voltage (MV) cable network.

The model is inspired by a real-life AC cable-connected offshore wind power plant (PP). The GFL converter model is a part of benchmark system, introduced by CIGRE WG C4.49 and published in the CIGRE 928 technical brochure.

The model is developed in the dq-reference frame to simplify modelling and avoid coupling in the control system. The converter is based on a standard insulated-gate bipolar transistor (IGBT)-based two-level voltage source converter (VSC), rated at 12 MW, as typically seen in modern offshore wind turbines (WTs). To simplify the analysis, the mechanical system and its controllers are not included.

Converter control is designed as GFL unit, employing vector control in the dq-reference frame and a synchronous reference frame (SRF) phase-locked loop (PLL) for grid synchronization. The dq currents regulate the DC link voltage and either the voltage or reactive power at the converter terminals. Active damping control, using capacitor current feedback, is also incorporated to enhance stability.

The converter system is linked to the grid through a 0.69/66 kV transformer, where the low-voltage reactance corresponds to the grid-side reactance of the LCL output filter.

Benchmark grid following converter control block diagram.

Subsystems of grid-following converter

The converter control system has been tuned to mimic the behavior of a generic converter model and has not yet been customized for the specific grid under study. As a result, various instabilities may arise in both the base case and during disturbance scenarios.

Anti-aliasing Filter and Sampling: the anti-aliasing filter is implemented using a second-order Butterworth filter, with the cutoff frequency set at half of the sampling frequency and the sampling delay is approximated using a third-order Padé approximation.

Park Transformation: the GFL control is implemented in a synchronous reference frame (SRF), with the phase determined by a phase-locked loop (PLL) that tracks the system frequency.

Power Calculation: instantaneous active and reactive power are calculated from voltage and current measurements taken at the output of the LCL filter.

DC Voltage Control: the DC voltage regulation is managed through a proportional-integral (PI) controller.

AC Voltage Control: the AC voltage regulation is implemented using a simple droop control method.

Reactive Power Control: the reactive power regulation is handled by a proportional-integral (PI) controller.

Phase-Locked Loop (PLL): the grid synchronization system uses a PLL, where the voltage’s q-component is filtered by a first-order low-pass filter and regulated by a PI controller, which provides the system’s angular frequency, which is then integrated to determine the phase for the Park transformation.

Current Control: the converter reactor current regulation is achieved using PI controllers with decoupling in the SRF, and active damping is incorporated to attenuate the capacitor current in the LCL filter, and an output voltage feed-forward component is added to the voltage reference.

Pulse Width Modulation (PWM): the modulation block computes the switching functions and provides the pulse patterns to the converter gate drivers and the PWM delay is also included.

Grid following converter parameters

List of parameters in the grid following converter electrical circuit.
List of parameters in the grid following converter control system.

References

[1] Ł. Kocewiak, R. Blasco-Gimenez, C. Buchhagen, J. B. Kwon, M. Larsson, Y. Sun, X. Wang, “Practical Aspects of Small-signal Stability Analysis and Instability Mitigation,” in Proc. The 21st Wind & Solar Integration Workshop, 12-14 October 2022, The Hauge, The Netherlands.
[2] Ł. Kocewiak, R. Blasco‐Giménez, C. Buchhagen, J. B. Kwon, M. Larsson, A. Schwanka Trevisan, Y. Sun, X. Wang, “Instability Mitigation Methods in Modern Converter-based Power Systems,” in Proc. The 20th International Workshop on Large-Scale Integration of Wind Power into Power Systems as well as Transmission Networks for Offshore Wind Farms, Energynautics GmbH, 29-30 September 2021.
[3] Ł. Kocewiak, R. Blasco‐Giménez, C. Buchhagen, J. B. Kwon, Y. Sun, A. Schwanka Trevisan, M. Larsson, X. Wang, “Overview, Status and Outline of Stability Analysis in Converter‐based Power Systems,” in Proc. The 19th International Workshop on Large-Scale Integration of Wind Power into Power Systems as well as Transmission Networks for Offshore Wind Farms, Energynautics GmbH, 11-12 November 2020.

Categories
Stability

Converter-based benchmark system from CIGRE 928 technical brochure

Introduction

A small-scale model of an actual AC cable-connected offshore wind power plant (PP) is proposed to compare different stability analysis methods and strategies for mitigating instability in converter-based power systems. This post provides a more detailed description of the benchmark system proposed by CIGRE WG C4.49. The system features power generation units (PGUs) connected to an AC grid via an extensive offshore 66-kV array cable system, offshore step-up transformers, both offshore and onshore HVAC transmission cables, and an onshore step-up transformer.

The primary goal of this benchmark power system is to offer a reference model where interactions between converters, as well as between converters and the grid, can be studied. It is designed to support small-scale, easy-to-model system studies and serve as a foundation for evaluating various instability mitigation methods introduced in CIGRE 928 technical brochure.

The benchmark system includes either aggregated converters or a collection of individual converters linked by a complex MV cable network. The model is formulated in the dq-reference frame to facilitate stability analysis using both impedance and modal approaches, allowing for a direct comparison of their results.

Study cases

Case 1: Aggregated grid following converters connected to a Thevenin equivalent

The transmission system is modeled as a long HVAC cable connected to a simplified Thevenin grid equivalent. The 420 MW power plant is represented by two aggregated power generation units (PGUs) of 180 MW and 240 MW. In this scenario the converters operate in grid-following mode.

The parameters for this aggregated system were derived from a detailed system representation, ensuring that the dynamics of the converters and their interaction with the grid remain consistent between the two. While the original benchmark configuration falls well within the capabilities of the modal and impedance analysis techniques discussed in CIGRE technical brochure 928, a simplified reduced-order model was developed to make the benchmark more accessible for researchers exploring new stability analysis methods, as well as to present results more concisely and clearly.

In this reduced version, each group of 5 PGUs, totaling 20 or 15 PGUs per group, is simplified into an equivalent cable segment and a single PGU. While the reduced model aligns well with the detailed model in the low-frequency range, significant deviations appear at higher frequencies, particularly beyond 500 Hz, as the Nyquist frequency of 1475 Hz, associated with the PGU control system, is approached. Consequently, the reduced-order model is not recommended for studying high-frequency resonance phenomena, but it is suitable for analyzing control interactions in the lower frequency range.

Case 2: Aggregated grid forming converters connected to a Thevenin equivalent

The transmission system is modeled as a long HVAC cable connected to a simplified Thévenin grid equivalent. The 420 MW power plant is represented by two aggregated power generation units of 180 MW and 240 MW. In this scenario, the converters operate in grid-forming control mode.

Download benchmark system

In Matlab

In PSCAD

In PowerFactory

To be ready soon...

References

[1] Ł. Kocewiak, Ch. Buchhagen, R. Blasco-gimenez, J. B. Kwon, M. Larsson, Y. Sun, X. Wang et al., “Multi-frequency stability of converter-based modern power systems,” Technical Brochure 928, Page(s) 1-147, CIGRE, March 2024.
[2] Ł. Kocewiak, R. Blasco-Gimenez, C. Buchhagen, J. B. Kwon, M. Larsson, Y. Sun, X. Wang, “Practical Aspects of Small-signal Stability Analysis and Instability Mitigation,” in Proc. The 21st Wind & Solar Integration Workshop, 12-14 October 2022, The Hauge, The Netherlands.

Categories
Stability

Instability Mitigation Measures in Modern Converter-Based Power Systems

Converter setpoint adjustment

The power generation unit frequency response changes depending on an operating point, e.g. active power setpoint, voltage setpoint. That is important for stability analysis within lower frequency range where the influence of phase-locked loop and controllers with lower dynamics, e.g. voltage control, is visible. Combination of specific setpoints and grid impedances, especially in case of low short-circuit ratio, can trigger unwanted instability. One can perform mapping of different operational points and schedule only the one not leading to any oscillatory behaviour. This mitigation measure can be applied within the design phase and during power plant operation. However, it is rather considered as corrective mitigation measure when any hardware changes are not available.

Converter control adjustment

Controller parameter tuning, digital delay compensation and pulse-width modulation filter resonance damping (or shifting) methods can be applied to improve the inherent stable operating region, all of which are digital modification that can be rapidly applied mostly without modification of hardware. Controller can be adapted (e.g. loop gain reduction, impedance shaping) to local grid conditions during all phases of the asset lifecycle, i.e. design, commissioning, operation, thus can be considered as preventive as well as corrective instability mitigation measure. However, much more flexibility in controller parameters adjustment as well as control structure modification is possible during the design phase. Adaptive control where the parameters are changed on the fly is also an attractive option, however complex to implement.

Internal and external operational scenario adjustment

Control interactions can often be avoided by changing the electrical topology. A change of the power plant internal electrical infrastructure will affect the short-circuit power and / or shift the resonance points in the system. Avoiding specific contingency operation cases can maintain the short-circuit ratio at a satisfactory high level. Moreover, if the outcome of the analysis demonstrates that a hazardous resonance point can be avoided in a specific power plant electrical infrastructure configuration, then this grid topology can be considered as an intermediate or final mitigation measure. The operational measures can only be chosen according to the individual situation and will always be very specific. Adjustment of operational scenarios within the power plant is rather considered as corrective mitigation measure, however, can be applied also during the design phase as a preventive measure.

All electrical components and subsystems are interconnected within the entire power system and contribute to a certain degree to assure stable operation. Thus, operational scenarios may not be changed only within the power plant internal electrical infrastructure but also within the power system to which it is connected. That also includes neighbouring power plants. In many cases the transmission system operator has much more flexibility to define operational philosophy focused on maximizing fault infeed and avoiding unwanted resonances, thus consequently improve the entire system robustness. Adjustment of operational scenarios within the grid is rather considered as corrective mitigation measure to address grid expansion or connection of new power plants. One of possible mitigation measures applied within the power system is inter-tripping to avoid contingency scenarios leading to converter instability.

Additional passive filter

Contrary to described earlier operational measures and control adjustment, a high-voltage passive filter can be added to alter the resonance frequency of the power grid at the point of interest. There is a large variety of passive filters being able to improve damping within harmonic frequency range. Installation of a passive filter is considered as preventive mitigation measure. It is much easier to add the passive filter during the design phase. Moreover, passive filters provide extra damping which together with stability enhancement can also reduce transient overvoltage and harmonics within specified frequency range.

Additional active power electronic equipment

The control interaction is a phenomenon caused by the active participation of power generation unit converters, which can occur with e.g. purely damped plant, weak grid, resonant network. This is because the converter control loops play an important role in defining the overall impedance of power plant. The impedance of each power plant can be reshaped by the addition of supplementary shunt-connected converters, such as a static synchronous compensators to provide extra resistive damping at the hazardous resonance frequency via a damping controller. As this type of mitigation requires additional equipment it is considered as preventive mitigation measure. However, static synchronous compensator are more often installed with renewable-based power plants and additional damping can be provided on already existing assets, thus in that case active damper one can consider also as corrective instability mitigation technique. Moreover, active damper control scheme can be adjusted on already operating assets to address potential power grid topology changes.

Power system voltage stiffness increase

In conventional power systems characterized by extensive use of synchronous machines the short-circuit ratio is typically calculated to measure the strength of the grid to which converter-based power plant is connected. Any attempt to improve short-circuit ratio by increasing the fault level at the power plant point of common coupling can improve the stability. It can be achieved by (i) avoiding severe contingency operation, (ii) improving transmission system capabilities by adding e.g. extra lines, (iii) installing synchronous condensers characterized by short-circuit current contribution. Increasing the fault level can improve the system robustness within the low-frequency range as it reflects the system impedance at the fundamental frequency. This mitigation measure is preventive because can be much easier initiated during the planning and design phases. Due to the complexity in power system modifications to improve short-circuit ratio, it is difficult to think about it as a corrective measure, unless one could easy avoid severe contingency operation leading to low short-circuit ratio. For modern power systems characterized by extensive use of power electronic equipment the voltage stiffness increase can be achieved by dedicated control loops if grid-following converters or use of grid-forming converters to improve the stability.

Categories
Measurements

PhD Course on  Harmonics in Power Electronics and Power Systems

Description:
This course provides a broad overview of power system harmonic problems, methods of analyzing, measuring and effectively mitigating them. Several extended simulation and data processing tools, among others DIgSILENT PowerFactory, Matlab/Simulink or LabVIEW are used to assess and study the harmonic distortion at different points of power networks.
The results of analytical investigation and simulations are validated against measurements applying sophisticated data processing techniques. Furthermore, deep understanding of hardware considerations regarding har- monic measurements in harsh industrial environment is given, using specialized equipment, for in- stance GPS-synchronized measuring instruments.

The course covers the following topics:

  • Power Quality definitions. Generation mechanism of power system harmonics. Harmonic indices.
  • Voltage vs. current distortion as well as parallel vs. series resonance in modern power systems. Point of Common Coupling (PCC).
  • Sources and effects of harmonic distortion.
  • Harmonic measuring instruments and measuring procedures in LV, MV and HV networks.
  • Mathematical tools and theories for analyzing distorted waveforms. Signal processing and uncertainty analysis.
  • Modelling of classical power system components. Harmonic analysis.
  • Modelling of grid-connected converters for harmonic analysis purposes and their application in modern power systems including e.g. offshore wind power plants.
  • Harmonic load-flow, frequency scan and time domain simulations. Linear and nonlinear analysis techniques.
  • Steady-state harmonics vs. harmonic stability. Small-signal representation, sequence and frequency coupling.
  • Software tools for harmonic analysis.
  • Precautionary (preventive) and corrective (remedial) harmonic mitigation techniques. Passive and active line filters. Filter design.

Organizer: Professor Claus Leth Bak
Lecturers: Christian Frank Flytkjær from Energinet and Łukasz H. Kocewiak from Ørsted

Harmonic current of 6-pulse rectifier supplying a resistive load
Figure 1 Harmonic current of 6-pulse rectifier supplying a resistive load.