Mattewos Tefferi, Nick Nakamura, Brad Barnes, Nenad Uzelac
©SHUTTERSTOCK.COM/URBANS
Power systems are experiencing a significant transformation with the implementation of emerging technologies to face 21st-century challenges such as decarbonization, digitization, and decentralization. The penetration of renewable energy is increasing worldwide, and initiatives such as distributed energy resource (DER)-based microgrids play a vital role in generating electrical power with fewer environmental impacts.
The operation of a microgrid depends on the successful integration of DERs, relying on several factors, such as hosting capacity, anti-islanding, synchronization, and power quality. Modern renewable energy sources use power electronic systems, such as rectifiers, dc/dc converters, and dc/ac inverters, that are susceptible to emitting power quality phenomena into the network, such as voltage instability, harmonics, frequency instability, and flicker. The continuous increase in switching frequencies resulting from these power electronics technologies has led to the emergence of new emissions in the range of 2–150 kHz (supraharmonic frequency), outside the traditional frequency range for power quality. Supraharmonics have the potential to disrupt network operations by damaging capacitors, disrupting communications, degrading dielectric insulation, and misoperating relays/controls. These symptoms can negatively affect the operation of street lighting controls, household dimmers, semiconductor manufacturing equipment, medical scanners, security systems, and transportation controls. Traditional technologies used on the grid today may not have the capability to properly measure and detect this new emerging supraharmonic threat to the electric power system reliability.
This article presents measurement observations of DER impact on voltage and frequency stability, utilizing state-of-the-art technologies available today. In the first part of the article, seven test scenarios were performed in a microgrid test bed to intentionally introduce disturbances and measure the resulting impact. In the second part, an unintentional event occurred in the microgrid, and the resulting measurements were recorded and evaluated.
The study was performed in a medium-voltage (MV) microgrid with natural gas generators, battery storage, voltage regulators, solar panels, and a wind turbine. The measurement instrumentation in the microgrid includes a set of capacitive voltage divider (CVD) sensors (0.5 class) certified to IEC 60044-7 and a power quality analyzer certified to IEC 61000-4-30 Class A for data collection and analysis of the measurement data.
Instrument transformers (ITs) are key technologies that enable applications for the metering, protection, and control of modern power grids. Initiatives such as DERs and optimizing energy efficiency and resiliency are driving a need for high-precision sensing technologies that can effectively monitor the state of the modern grid. Low-power voltage transformers (LPVTs) are typically based on either CVDs or resistive voltage dividers (RVDs) and are used to step-down the primary MV to an appropriate low-voltage signal measured by an intelligent electronic device (IED), such as a meter or controller.
A CVD sensor consists of a series of two capacitive arms: a primary arm with a low capacitance value, in the order of a few pF, and a secondary arm with a high capacitance value, typically hundreds of nF. On the other hand, an RVD sensor comprises a high primary resistor, in the order of 10–100 MΩ, and a low secondary resistor, typically 10–100 kΩ. Both technologies involve connections to the primary voltage, and the secondary voltage is then measured across the secondary arm.
A key characteristic of CVD sensor technology used in DER applications is the dynamic measurement range and ability to measure and detect supraharmonic (2–150 kHz) phenomena. LPVTs must maintain accuracy class over a wide band of a frequency range. The frequency characteristics of CVD and RVD test samples via the frequency-sweeping method are shown in Figure 1. A constant voltage was applied, and the output voltage was monitored for both technologies. The cutoff frequencies were 7 kHz and 40 kHz for RVD and CVD, respectively. The high cutoff frequency of CVDs makes it suitable for detecting high-frequency supraharmonic phenomena up to 40 kHz.
Figure 1. Frequency response of CVD and RVD sensors.
Figure 2(a) represents a simplified scheme of the microgrid test bed under study. The DERs in the microgrid include natural gas generators, battery storage, solar panels, and a wind turbine. The microgrid services either the on-site research facility and an electric vehicle station or a small distribution feeder serving approximately 200 residential and small commercial customers. The battery energy storage system (BESS) and natural gas generators both have grid-forming capability. When islanding only the on-site research center and vehicle-charging station, the BESS is used as the grid-forming asset. When islanding the distribution feeder, the natural gas generators act as the grid-forming asset. In either islanding scenario, the renewable assets are allowed to produce with their output level managed by the microgrid control system. All assets are available for use when the system is grid tied. Figure 2(b) shows the renewable energy sources and field installation of the CVDs at the microgrid.
Figure 2. (a) Simplified scheme of the microgrid test bed network and location of measuring devices. (b) Renewable energy sources at the microgrid and field installation of voltage sensors. MVA: mega volt amperes.
Seven test scenarios were performed with the various DER sources in the microgrid (Table 1). These experiments were performed to intentionally introduce disturbances and measure the resulting impact with two different measurement instruments, including low-power CVD sensors and a power quality analyzer. A view of the voltage magnitude, frequency, conducted emissions (supraharmonic frequencies), total harmonic distortion (THD), and flicker was monitored over this time period.
Table 1. Testing scenarios in microgrid.
Based on the data observed, power quality-related observations were noted in test scenarios 4, 6, and 7. The following analysis will focus on islanded conditions.
Voltage fluctuations are systemic variations of the voltage envelope or random voltage changes (Figure 3). Industry standard IEC 61000-4-30 indicates that voltage dip (sag) thresholds are typically in the range of 85%–90% of the nominal voltage, while voltage swell thresholds are typically greater than 110% of the nominal voltage.
Figure 3. Voltage and frequency fluctuations observed during test scenarios 1–7. Avg.: average.
While no voltage deviations exceeded 10% of the nominal 7.2-kV line-to-ground system voltage, voltage instability was observed with minimum values of 6.9 kV during islanding all DERs and 6.84 kV during islanded and isolated solar scenarios. The voltage magnitude behavior exhibited fluctuations and less stability when islanded and stabilized when tying back into the grid on all phases.
Frequency deviations of +0.2 Hz were observed for 10 min during islanding all DERs and +0.5 Hz during periods with solar inverters islanded and isolated. The highest frequency deviations of +0.6 Hz occurred during periods with the wind turbine islanded and isolated.
Supraharmonics are harmonic distortions in voltage and current waveforms in the frequency range of 2–150 kHz. The continuous increase in DERs has resulted in a proliferation of inverter-based power electronics that are subject to switching frequencies in the supraharmonic frequency range. Additionally, these distortions are found in nonlinear loads associated with variable-frequency drives, electric vehicle chargers, LED controllers, and uninterruptible power supplies (UPSs). These sources are subject to exhibiting symptoms such as thermal stress on the connected equipment, insulation stress on cables, premature power supply failure, IED misoperation, and lighting control malfunction.
Industry standard IEC 61000-4-30:2015 Ed3 (informative) provides guidance on how to measure these supraharmonic distortions (conducted emissions) in the 2–150-kHz range. The power quality analyzer referenced in this study measures conducted emissions in 2-kHz segments with minimum, average, and maximum magnitudes of the RMS voltage in each segment. Other standards that reference conducted emissions include IEC 61000-2-2:2002, with a focus on compatibility levels for voltage distortion and emissions, and CISPR-16, with a focus on providing methods for measuring high-frequency radio disturbances and immunity greater than 9 kHz.
While the potential for supraharmonic phenomena exists in a DER-impacted grid, there are limited references documented in industry standards or guides for measuring supraharmonics in MV or as a complete measurement system including sensors, cables, and IEDs. The measurement results published herein are subject to uncertainty in the measurement system used in the test bed demonstration. The measurement instruments used in the test have been tested independently at voltage supraharmonic levels with the power quality analyzer compliant with IEC 61000-4-30:2015 and the CVD sensors tested to frequency cutoff, as identified in Figure 1.
Figure 4 presents the supraharmonics during the test scenarios. Supraharmonic measurements were observed throughout the islanding transition and islanded isolated solar and wind experiments. The graph on the left represents the average conducted emissions voltage measurements observed, and the table on the right represents the average voltages observed at four frequencies in the supraharmonic range, ranging from 4 to 16 kHz. The maximum values were observed when the solar inverters and wind turbines were islanded, suggesting that they are the main sources of MV supraharmonic distortion.
Figure 4. Supraharmonics observed during the testing scenarios. Freq.: frequency.
Harmonic distortion in DERs is caused by nonlinear devices (nonlinear loads) when the current is not proportional to the applied voltage. As the integration of DERs into the grid advances, various harmonic distortion limit criteria are implemented to ensure that the voltage and current waveform are compatible with the grid. IEEE 519, IEEE 1547-2018, and IEC 61000-3-2 standards impose that the voltage THD must not exceed 5% at MV levels.
Figure 5 represents the THD observed during isolated and islanded solar and isolated and islanded wind turbines. The THD is observed up to 6.6%, exceeding the limits identified in the standards.
Figure 5. THD values observed during the testing scenarios.
Flicker is defined in the IEC 61000-4-30 standard as an impression of visual discontinuity induced by a light stimulus whose luminance or spectral distribution fluctuates with time. Light flicker phenomena appears when there is a fluctuation of voltage. The IEC 61000-4-15 standard establishes a voltage signal as the input, and the measurement procedure reproduces the response of the human vision system by precisely characterizing real flicker perception. Instantaneous flicker perception (Pinst) is given in perceptibility units, where a unit value defines the reference human flicker perceptibility threshold, which means that such a level of flicker would be perceived by 50% of the population. However, this perception does not mean irritation and, therefore, cannot be directly related to customers’ complaints.
To represent the irritation, the flicker meter integrates the flicker perception Pinst over two types of flicker. The short-term flicker (PST) is a statistical analysis of Pinst after 10 min, and the long-term flicker (PLT) is the mean value of Pinst over the previous 2 h, both synchronized to a real-time clock. The value of PST shall not exceed 1.0, and the value of PLT shall not exceed 0.65. Figure 6 represents the flicker values measured during the island transition and the islanded and isolated wind and solar experiments.
Figure 6. Short-term flicker (PST) and Long-term flicker (PLT) observed during the test scenarios.
During the test bed demonstration, several months after the simulated test experiments, an island event was observed that lasted for a duration of approximately 45 min and resulted in several power quality observations. The BESS is the grid-forming asset with an imposed maximum state of charge. The microgrid control system is designed to temporarily curtail solar and wind DERs when in island mode if the energy production exceeds demand and the BESS is at its imposed maximum state of charge. Once the BESS state of charge drops below a prescribed threshold, the DER curtailment will be released. If communication to the DER source fails and the curtailment signal is not acknowledged, the microgrid control system will open the DER interconnection switch to isolate the asset from the microgrid.
During an islanding event, the microgrid control system was unable to communicate with one of the five inverters that make up the solar DERs. This resulted in a looping sequence that involved isolating the entire solar DER due to the loss of communication, the BESS discharging to a state below its threshold, and closing the solar DER back into the island. Based on the sequence of events, the most likely cause of the voltage and frequency instability observed is likely due to the switching operation and resulting transformer inrush current, an overpower production issue that drove the BESS inverter into an unstable region, or a combination of both.
Voltage and frequency fluctuations were observed for approximately 45 min before stabilizing, as observed in Figure 7. Voltage dips (sags) were observed below the typical 10–15% threshold at 5.76 kV on a nominal 7.2-kV system.
Figure 7. Voltage and frequency fluctuation during an island event.
Figure 8 represents the supraharmonics measured throughout the duration of the islanding event. The graph on the left represents the average conducted emissions voltage measurements observed, and the table on the right represents the average voltages observed at four frequencies in the supraharmonic range, ranging from 8 kHz to 24 kHz. These observations demonstrate that the emission behavior during the event showed maximum values that are significantly higher than what was observed in the test scenarios during islanding.
Figure 8. Supraharmonics observed during the island event.
Figure 9 represents the THD observed during the islanding event. Maximum values were observed up to 20.85%, exceeding the 5% limits identified in the standards.
Figure 9. THD observed during the island event.
Figure 10 represents the flicker values measured during the islanding event. The Pst values were observed >1.0, above the thresholds published in standards. During the event, it was reported that flickering lights were observed at the onsite research center.
Figure 10. Short-term flicker (PST) and Long-term flicker (PLT) observed during the island event.
The microgrid test bed demonstration with the power quality analyzer paired with high-performance CVD sensors proved that DERs are susceptible to generating power quality phenomena such as supraharmonics, voltage instability, THD, and flicker. The test scenarios performed yielded power quality-related observations in three of the seven experiments. Islanded conditions demonstrated instability with voltage fluctuations, THD, and evidence of supraharmonic frequencies in the 4–16-kHz range.
In addition to the test scenarios, an unintentional event was measured and recorded with the CVD sensors and power quality analyzer. This event involved an islanded condition with a sequence of switching DERs that resulted in approximately 45 min of power quality issues. During this event, observations were recorded including voltage sags at 5.76 kV, a frequency of 60.3 Hz, flicker Pst >1.0, THD up to 20.8%, and supraharmonic frequencies up to 24 kHz. Table 2 summarizes the power quality issues observed during the three test scenarios and the microgrid islanding event.
Table 2. Test scenarios and event observations summary.
These measurement observations demonstrate that DERs do have an impact on grid power quality and that supraharmonic frequencies are present beyond what traditional technologies can measure at the MV level. Traditional transformer technologies (designed based on magnetic induction) may have frequency cutoff measurement limitations that inhibit their ability to measure supraharmonic frequencies. Traditional IEDs may have frequency measurement limitations if they are designed to measure up to typical industry guidelines at the 50th harmonic (3 kHz). While the impact of supraharmonic activity on MV grid reliability is not thoroughly understood, this study demonstrates that DERs do generate them and that they can be measured with capable sensors and power quality analyzers. The CVD sensor and power quality analyzer system applied in this microgrid test bed have demonstrated supraharmonic frequency measurement of 4–24 kHz that may be limited or undetectable with traditional measurement systems.
M. H. J. Bollen, Understanding Power Quality Problems: Voltage Sags and Interruptions. Piscataway, NJ, USA: IEEE Press, 2002.
S. T. Y. Alfalahi et al., “Supraharmonics in power grid: Identification, standards, and measurement techniques,” IEEE Access, vol. 9, pp. 103,677–103,690, Jan. 2021, doi: 10.1109/ACCESS.2021.3099013.
Electromagnetic Compatibility (EMC) - Part 4-30: Testing and Measurement Techniques - Power Quality Measurement Methods, IEC 61000-4-30, 2015.
IEEE Guide for Identifying and Improving Voltage Quality in Power Systems, IEEE Standard 1250-2018 (Revision of IEEE Standard 1250-2011), Nov. 2018.
IEEE Recommended Practice for Measurement and Limits of Voltage Fluctuations and Associated Light Flicker on AC Power Systems, IEEE Standard 1453-2004 (Adoption of CEI/IEC 61000-4-15:1997+A1:2003), Mar. 2005.
IEEE Recommended Practices and Requirements for Harmonic Control in Electric Power Systems, IEEE Standard 519-1992.
IEEE Standard for Interconnecting Distributed Resources with Electric Power Systems, IEEE Standard 1547-2003, 2003.
Mattewos Tefferi (mtefferi@gwelec.com) is with G&W Electric Co., Bolingbrook, IL 60440 USA.
Nick Nakamura (Nick.Nakamura@powerside.com) is with Powerside, Alameda, CA 94501 USA.
Brad Barnes (BBarnes@ameren.com) is with Ameren Illinois, Decatur, IL 62526 USA.
Nenad Uzelac (nuzelac@gwelec.com) is with G&W Electric Co., Bolingbrook, IL 60440 USA.
Digital Object Identifier 10.1109/MELE.2023.3264929
2325-5897/23©2023IEEE