Wind Farm Discoveries that Impact the Grid

Presented By:
Eric Hope PE
Troy Ryan
Leeward Renewable Energy
TechCon 2024


This paper will discuss the various impacts of wind generation on the grid. Topics include Wind Turbine Generator (WTG) interconnection fundamentals, voltage and frequency support, power quality, and wind turbine generator performance. The paper will also discuss Wind Turbine Generator operational experiences.


Interconnection of wind generation with the grid presents unique challenges compared to traditional thermal generation. This paper presents wind turbine generator (WTG) interconnection fundamentals, voltage and frequency support, power quality and reliability, and wind turbine generator performance. The paper will also discuss Wind Farm operational experiences.

1. Wind Turbine Generator Interconnection Fundamentals

Traditional thermal power plants typically consist of one or more large generating units in the range of 50MW for small, natural gas, peaker units to over 1,000MW for nuclear generating units. These units generate electricity at the medium voltage level – typically 13.8kV to 25kV. The synchronous generator is connected directly to a generator step-up (GSU) transformer, located only a short distance away, via an isolated phase bus duct. The GSU steps up the voltage to the grid voltage level – typically in the range of 115kV to 500kV. The high-voltage side of a generator step-up transformer is then connected directly to the transmission switchyard via a generator tie line. Figure 1 below shows an example single line diagram of the main power circuits of a traditional thermal generating unit.

Figure 1 Example Thermal Generation Single Line Diagram

In contrast, a utility-scale wind farm consists of many small generating units in the range of 1-5MW for terrestrial wind farms and 10-15MW for offshore wind farms. These units are spread over a large geographic area to optimally capture wind energy. This requires an extensive network of medium voltage cables, called the collection system, to electrically connect the individual generating units and transmit power to the transmission grid. In a typical utility-scale wind farm, the individual wind turbine generators (WTGs) generate power at the low voltage level – typically 600-900V. The generators will either include an integral GSU transformer at the top of the tower (called the nacelle) or an external pad-mounted transformer will be employed as a GSU. The GSU transformer steps-up the voltage to the medium-voltage level – typically 34.5kV for terrestrial wind farms. Then, a large array of medium-voltage cables connects the individual wind turbines back to the collection substation, which is interconnected at grid voltage. The turbines are typically daisy-chained together in groups, forming individual feeders that terminate on a medium-voltage circuit breaker in the collection substation. In the collection substation, the feeders are bussed together using multiple circuit breakers, and the voltage is stepped-up to the transmission voltage level via a main power transformer (MPT). This transformer may also be referred to as a main GSU or Station Transformer. Oftentimes, the collection substation will include reactive power compensation in the form of switched shunt capacitor banks, switched shunt reactors, and/or dynamic reactive power compensation devices (e.g. STATCOM or SVC) at the medium-voltage level (e.g. 34.5kV). The wind farm interconnects to the transmission grid via a generator tie (gen-tie) line. The tie line could be a short transmission line over the substation fence to interconnect to a neighboring transmission substation. However, oftentimes wind farms are in remote areas, which requires a longer gen-tie line (e.g.10-20 miles or longer). Figure 2 below shows an example single line for a utility-scale wind farm.

Figure 2 Example Utility-Scale Wind Farm Single Line Diagram

The vast array of collection system cables, multiple voltage transformations, and long gen-tie lines, introduce voltage drop, capacitive charging current, and real and reactive power losses into the system. This presents unique interconnection challenges for wind farms compared to traditional thermal generating units. The most significant challenges include voltage and frequency support, maintaining acceptable power quality, and good wind turbine generator performance during grid disturbances.

2. Voltage and Frequency Support

Wind Farms provide voltage support to the transmission grid by supplying or absorbing reactive power based on a voltage drop characteristic. Voltage control is performed by the inverter connected wind farm power plant controller, which measures the grid voltage and sends reactive power commands to individual wind turbine generators as well as sending switching commands to the shunt capacitors and reactors in the collection substation (if equipped). Alternatively, the wind farm may be operated in reactive power control mode, in which the grid operator sends reactive power commands to the wind farm. In either control mode configuration, the grid operator will specify a setpoint (including the target voltage or reactive power setpoint). Generally, the wind farm must respond within 10-minutes to a setpoint change.

FERC Order 827, which was released in 2016, mandates that all non-synchronous generation have the capability to provide and absorb reactive power within the power factor range of 0.95 leading to 0.95 lagging at the high voltage bus of the generator substation. Additionally, several transmission owners and system operators (e.g. ERCOT) may have more stringent requirements. The reactive power capability must be dynamic, meaning that the reactive power output can be commanded to be anywhere within the range. FERC 827 allows dynamically switched shunt capacitors and reactors to be utilized to offset reactive power losses in the wind farm.

Modern Type 3 (doubly-fed induction generators) and Type 4 (full-converter) wind turbine generators utilize power electronic converters, which enable the turbines to provide or absorb reactive power. Typical wind turbine vendor offerings include the reactive power capability in the range of 0.95 leading or lagging to 0.87 leading or lagging. Additionally, many turbines offer the capability to provide reactive power output when the turbines are not generating. Figure 3 shows an example wind turbine generator reactive power capability for a Type 4 wind turbine generator.

Figure 3 Example Wind Turbine Generator Reactive Power Capability Curve graph

While the wind turbines have capability to provide and absorb reactive power, the collection system cables and power transformers introduce reactive power losses and cable charging current. Depending on the wind turbine reactive power capabilities and collection system design, additional reactive power compensation may be required to offset the reactive losses incurred in the collection system. Switched shunt capacitors are typically employed to offset reactive power losses that occur at high levels of generation output. Switched shunt reactors are utilized to offset cable charging current at low levels of generation output (if the turbines lack capability to provide reactive power under low/no generation output. Dynamic reactive compensation (e.g. STATCOMs or SVCs) may be needed if additional dynamic reactive compensation power is needed. However, given the capability of modern wind turbine generators, these instances are rare. A reactive power study, which simulates the power flow in the wind farm under a variety of generation and grid conditions must be conducted to determine if additional compensation is needed. The reactive power study determines the reactive power capability of the wind farm from a system level at the point of interconnection with the transmission grid to verify compliance with the grid codes. Figure 4 shows an example wind farm reactive power capability curve produced from a reactive power study along with the FERC 827 requirements. The figure illustrates the need for a switched shunt capacitor bank to offset reactive power losses at full power output.

Figure 4 Example Wind Farm Reactive Power Capability Curve graph

In addition to voltage support, wind farms also must have the capability to provide primary frequency response to the grid, as mandated by FERC 842. Frequency support is accomplished by reducing power output for overfrequency events and increasing power output for underfrequency events. Note that most wind farms do not have the capability to increase power output unless they are in a curtailed state, so the response to under-frequency events is not required unless the wind farm is generating in a curtailed state as controlled by the grid operator. In a typical wind farm design, the power plant controller measures the frequency of the power system and adjusts the power commands to the individual turbines if the system frequency deviates outside of the mandated dead band – ±36 mHz per FERC 842 or ±17 mHz for the ERCOT region. Figure 5 shows the grid frequency measurement captured at a wind farm in ERCOT over the course of several days. The figure shows numerous grid frequency excursions outside of the ±17 mHz dead band for which the plant would need to respond.

Figure 5 Wind Farm Frequency Measurement with ±17 mHz control dead band graph

3. Wind Farm Harmonics

Modern wind turbine generators are designed to be compliant with harmonic emissions standards (e.g. IEEE 519 in North America) and are therefore not a significant source of harmonic current distortion. Wind turbine generators are equipped with internal harmonic filters to inhibit harmonic currents generated by the power electronic converters in the WTG from flowing into the grid.

Though the wind turbine generators are not significant sources of harmonic distortion, the capacitance of the collection system cables and switched shunt capacitors in the collection substation present a potential for harmonic resonance condition with the main power transformer. The main power transformer is primarily inductive and will have an effective ac impedance of jωL, where j=sqrt(-1), ω is 2*π*frequency, and L is the equivalent inductance of the transformer. The shunt capacitance of the substation capacitor banks and collection system cables has an impedance of -1/(jωC), where C is the total capacitance. At some frequency (ω), the two impedances will be equal but opposite and sum to zero. This creates a short-circuit condition at that frequency. Figure 6 below shows a simple circuit diagram for a wind farm, demonstrating the potential for harmonic resonance. If the frequency lines up with background voltage harmonics on the grid, the wind farm has the potential to pull in a large amount of harmonic current at that frequency in the grid, resulting in high levels of voltage distortion in the collection system and additional thermal heating of equipment, potential for blown fuses or nuisance protection trips, and potential damage to balance of plant equipment.

Figure 6 Simple Wind Farm Circuit Diagram Showing Effects of Harmonic Resonance diagram

A harmonic study is typically performed during the wind farm design to identify potential harmonic resonance issues and confirm the project is compliant with IEEE 519. The study will include harmonic frequency scans for a variety of system configurations to identify potential harmonic resonance conditions. An example frequency scan is shown in Figure 7 below. A harmonic distortion analysis is also performed as part of the study to determine if harmonic resonance conditions identified in the frequency scans could manifest into harmonic distortion issues. If the harmonic study identifies a potential issue, then potential mitigation options (e.g. harmonic filtering) will be identified in the study. However, due to the high levels of uncertainty in harmonic study modeling, most wind farm developers take a “wait and see” approach to employing harmonic mitigation. The harmonic study results are first verified by performing power quality monitoring once the wind farm is in operation to confirm if a harmonic distortion issue is present. Oftentimes, a desktop harmonic study will identify potential resonance issues, but the issue will not manifest in the field. Harmonic monitoring also provides data to benchmark harmonic study models. Harmonic resonance issues are usually self-revealing as they will typically result in a protection trip of the capacitor bank and/or wind turbine generators.

Figure 7 Example Harmonic Frequency Scan Showing Resonance graph

4. Operational Experiences

Wind farm interconnection agreements typically require a quick response to voltage or power factor adjustments- usually within 10 minutes. This demands well-maintained and functioning static (switched shunt capacitors and reactors) and dynamic (e.g. STATCOMs, DVARs) reactive power equipment. Occasionally, issues may arise and take months to resolve. Therefore, it is essential to maintain a strong relationship with the equipment manufacturers for technical support, especially in reactive power systems. Long warranty periods and long-term service agreements for these systems is considered beneficial for sustained reliability.

First Operational Experience

The first operational example involves two sites – Site 1, with an operational history of over three years, has a capacity of approximately 300 MW and Site 2, with a capacity of around 200 MW, is newly commissioned. These sites are connected via a 70-mile, 230 kV gen-tie line leading to the Point of Interconnection (POI), including a 12-mile line through another 230kV switchyard. Each site has its own reactive power controller and reactive power compensation equipment. Site 1’s reactive compensation is overbuilt with both static compensation (capacitor banks) and dynamic compensation (DVAR). For three years of operation, Site 1 was successful in meeting the utility-required power factor at the POI, including meeting the 10-minute response time for changes.

An issue emerged when Site 2 began generating power at full capacity. Protection trips occurred in most of Site 1’s towers due to undervoltage conditions. Site 2’s operation at the POI required new 230kV capacitors, which were commissioned but failed to function adequately. This issue occurred partly because Site 2 had underbuilt its reactive power, based on the assumption that Site 1’s overbuilt capabilities would compensate.

The solution involved implementation of a shared Master Controller to coordinate the reactive power output of both sites to meet the set point at the POI. The master controller helped avoid conflicts in reaching the set point. Additionally, adjustments were made to the secondary of the turbine pad-mounted step-up transformers on some turbines through a no-load tap change to provide additional voltage margin at the turbine terminals. In the long term, maintaining the functionality of all reactive power systems is crucial. Before commissioning Site 2, it was possible to have a few capacitor banks or even a DVAR offline. However, when Site 2 was installed, all the reactive compensation equipment needed to be online and functioning properly to meet the reactive power requirements at the POI.

The recommendation from this case study is to overbuild static and dynamic reactive power capabilities to ensure sufficient capability to allow for planned maintenance and forced outages. Additionally, it is recommended to coordinate voltage/reactive power controls for generation sites sharing a gen-tie line.

Figure 8 Single Line Diagram Showing Capacitor Banks and Cap Yard image

Second Operational Experience

Another operation example involves a winter visit to a northeastern substation, it was noticed that snow was melting around the grounds in the capacitor area. An inspection revealed that approximately 400 Amps were flowing to ground through the ground wires in this area. Consequently, the capacitor bank was taken out of service and de-energized to identify the issue.

The investigation found that the harmonic reactors, which were placed in front of the capacitors in the circuit, had been installed incorrectly, being positioned too close to each other. Figure 9 shows the installation of the reactors. This improper installation caused the aluminum I-beams, used to support the reactors, to be trapped in the magnetic field. This, in turn, induced current in the I-beams, which was then running to ground.

A further complication arose as the studies justifying the use of the harmonic reactors could not be located, requiring a new study to be completed. This testing revealed that the reactors, when operational, exacerbated the harmonics issue. To address the problem, the capacitor bank was re-tested for performance after the removal of the reactors from the circuit. The issue was ultimately resolved by removing the reactors from the capacitor bank. Since this intervention, the capacitor bank has been functioning without further issues.

As a recommendation, it is advised that when installing reactors, one should strictly adhere to the manufacturer’s recommendations for both design and installation. Additionally, harmonic monitoring should be performed prior to the installation of harmonic filtering equipment to confirm mitigation specified in the desktop harmonic study is needed.

Figure 9 Harmonically Tuned Reactors as Installed in the Field image


1. FERC Docket No. RM16-1-000; Order No. 827, “Reactive Power Requirements for Non-Synchronous Generation”, issued June 16, 2016.

2. FERC Docket No. RM16-6-000; Order No. 842, “Essential Reliability Services and the Evolving Bulk-Power System—Primary Frequency Response”, issued February 15, 2018.

3. IEEE 519-2022, “IEEE Standard for Harmonic Control in Electric Power Systems”

Join our email list

We use cookies to give you the best online experience. By using this website you agree with our cookie policy.