Coal, Renewables

The Western Grid Can Weather Disturbances Under High Renewable Penetrations

Issue 1 and Volume 119.

By Kara Clark, National Renewable Energy Laboratory, and Nicholas W. Miller, Miaolei Shao, Slobodan Pajic, and Robert D’Aquila, GE Energy Consulting

A new report finds that, with good system planning, sound engineering practices, and commercially available technologies, the Western Interconnection can withstand the crucial first minute after grid disturbances with high penetrations of wind and solar on the grid.

The report by the U.S. Department of Energy’s National Renewable Energy Laboratory and GE Energy Consulting is titled The Western Wind and Solar Integration Study Phase 3 (WWSIS-3)-Frequency Response and Transient Stability Study.

Large-scale transient stability and frequency response are critical to grid reliability, particularly for the Western Interconnection, which has a long history of dynamic performance constraints on its operation. The new report specifically addresses the dynamic performance of the Western Interconnection with high penetrations of renewable energy. The report examines a range of scenarios with instantaneous renewable energy penetrations of up to 53 percent (see Table 1), under both light spring and heavy summer load conditions. For each scenario, it analyzes grid performance in the tens of seconds following a large grid disturbance, such as a loss of a large power plant or major transmission line.

table 1

For modeling the renewable energy systems, all new wind plants and utility-scale photovoltaic (PV) plants were modeled as asynchronous machines with voltage regulation and low-voltage ride through (LVRT), while concentrating solar power (CSP) plants were modeled as synchronous machines without governor response.

All new distributed PV was modeled using the WECC composite load model.

Frequency Response After Generation Loss

Frequency response is the overall response of the grid to large, sudden mismatches between generation and load. The primary concern is that the minimum frequency, or nadir, during design-basis disturbances should not cause under-frequency load shedding (UFLS), which means dropping customers from the grid. In the West, the first stage of UFLS is normally at 59.5 Hz. Based on standards developed by the North American Electric Reliability Corporation (NERC), the Western Interconnection must also comply with a frequency response obligation (FRO) of 840 MW/0.1 Hz, which means that the power output from all generators should increase by 840 MW for a frequency drop of 0.1 Hz.

Without special operation or controls, wind and solar plants do not inherently participate in the regulation of grid frequency. By contrast, synchronous machines always contribute to system inertia, and some fraction of the synchronous generation in operation at any point has governor controls enabled. When wind and solar generation displaces conventional synchronous generation, the mix of the remaining synchronous generators changes and has the potential to adversely impact overall frequency response.

This analysis focused on spring conditions when the loads are light, because the relatively low level of synchronous power generation may present a challenge for frequency response. It also focused on the single largest design-basis generation outage in the Western Interconnection: the trip of two fully-loaded Palo Verde nuclear power station units for a loss of about 2,750 MW.

The subsequent frequency excursion is severe, as shown in Figure 1, but in all cases, the frequency nadirs avoid UFLS relay action, which begins at 59.5 Hz. The frequency nadir is 59.67 Hz in the base case (blue line), 59.65 Hz for the high renewable case (green line), and 59.61 Hz for the extremely high renewable case (red line).

1 Frequency response to the loss of two Palo Verde units under light spring system conditions.
figure 1

In addition, the interconnection-wide frequency response meets its obligation (840 MW/0.1 Hz) in all three cases. However, portions of the system that rely primarily on thermal generation tend to fall short of meeting their approximate FRO with their own generation resources, especially in the case with a high mix of renewable energy. This occurs because that thermal generation was displaced by wind and solar, which do not provide frequency response unless equipped with specific controls. Other regions, particularly the Northwest, far exceed their approximate FRO due to high levels of responsive hydropower.

Renewable Generators Can Contribute to Frequency Response

Despite the encouraging results of the frequency-response evaluation, WWSIS-3 also examined ways that renewable generation sources could contribute to frequency response and grid stability. Specifically, the report examined frequency-responsive controls on wind and utility-scale PV plants.

For wind plants, one option examined was inertial control, which involves extracting additional energy from the wind plant for a short period of time, drawing on the momentum of the spinning blades and slowing them down momentarily. The downside of this approach is that power production subsequently dips as the turbines regain speed. The study found that inertial control provided a substantial improvement in keeping the frequency nadir above the UFLS, but the energy-recovery period tends to stretch out the frequency depression over time.

Governor controls involve intentionally curtailing wind plant output, holding this extra power production in reserve for frequency response. In this case, a 5 percent curtailment was applied only to the new wind plants, creating an overall curtailment of only 1 percent of the total wind production. Using a governor or active power control alone greatly improved the settling frequency–the frequency that the system ends up at after the transient response-and the frequency response, but had little impact on the nadir. However, the combination of both the governor and inertial control improved the frequency nadir, settling frequency, and frequency response. Note that the wind governor controls were set to emulate those on conventional generation. Both the wind governor and inertial controls could be made more aggressive.

In addition, WWSIS-3 examined governor controls for new utility-scale PV plants (again requiring a 5% curtailment), and it found such controls also greatly improved frequency response.

2 Frequency response to two Palo Verde unit trips for high mixes of renewable energy under light loads, with three combinations of frequency controls on wind plants.
figure 2

Improving Transient Stability

The Western Interconnection has a long history of constraints due to transient stability limitations that vary depending on system characteristics, such as the level of electricity demand (e.g., peak summer load), the amount of power flowing on the transmission system (e.g., heavy flows on critical paths), and the location of generating plants in operation (e.g., remote from population centers). Large penetrations of inverter-based, or non-synchronous, wind and solar generation may substantially alter the system’s stability, so this analysis focused on heavy load conditions and outages that stress major interfaces in the West.

One of the major vulnerabilities is in the power flows from the Northwest to Southern California, because a limited number of pathways exist. For instance, the California-Oregon Interface (COI) is sensitive to disturbances under high-flow conditions; the loss of another north-south route, the Pacific DC Intertie, requires some generation to be tripped offline to recover. For the base case, system performance degrades significantly with the loss of the Pacific DC Intertie under high power flows on the COI, and this dynamic is not fundamentally changed in the case with a high mix of renewable energy.

To help address this problem, WWSIS-3 examined the use of governor response on all new CSP plants in the West. As shown in Figure 3, this enabled the system to recover without having to trip any generation offline. One reason that the CSP plants provided such a benefit is that most of them are located in Southern California, south of the COI, so they provide crucial power to that area when it is cut off from power sources in the Northwest.

3 Voltage response to a loss of the Pacific DC Intertie, without tripping generation, with and without CSP governor controls.
figure 3

Addressing Coal Displacement and Weak Grid Concerns in Wyoming

As demonstrated by the transient analysis, high penetrations of renewable energy on the Western Interconnection have a direct impact on how the system functions, simply because the geographic locations of the main power sources will change. Specifically, a high mix of renewable energy results in decreasing coal production in the Desert Southwest (Arizona, Nevada, New Mexico, Colorado) and in the northeast section of the interconnection, which includes Idaho, Montana, Utah, and Wyoming.

Because that northeast region currently features significant power production from coal, the high-mix scenario results in more than an 80% reduction in coal production for this particular condition in the spring, when loads are light. The transmission system in that region was designed based on the size and location of the large coal power plants, which thus became critical nodes in the network. As a result, the transmission system operators have historically counted on those plants to provide the voltage and reactive power support needed for reliable operation. Displacement of those central plants by more dispersed wind and solar generation results in those nodes being poorly supported. Not surprisingly, local voltage and thermal problems occur, and good planning practices need to be followed.

However, despite such challenges, WWSIS-3 found that under the high-mix scenario, an extra-high-voltage fault at Aeolus-a planned transmission line in the heart of the high-wind area in the northeast region-did not lead to system instabilities. Of course, the stress on the system becomes larger under the “extreme” renewable energy case, which causes a 90% reduction in coal commitment from the base case. The rapid voltage collapse and system separation during the fault, as shown in Figure 4, is representative of so-called “weak grid” issues.

Systems with very high levels of inverter-based generation are challenged to provide fast, confident control during faults and other disturbances. No commercially available wind or utility-scale PV generation is capable of operation in a system without the stabilizing benefit of synchronous machines. Therefore, some mitigation was required. The conversion of some coal plants to synchronous condensers and the addition of mechanically switched shunt compensation successfully stabilized the system for the Aeolus fault with the conservative load and wind plant modeling used. The synchronous condenser conversion works well to stabilize the system, which recovers in an orderly fashion when the fault is cleared (see Figure 4).

4 Bus voltage for a fault at Aeolus under varying penetrations of renewable energy, with and without synchronous condenser conversion
figure 4

Conclusions

WWSIS-3 did not identify any fundamental reasons why the Western Interconnection cannot meet transient stability and frequency response objectives with high levels of wind and solar generation. However, good system planning and power system engineering practices must be followed. At a minimum, local voltage and thermal problems will inevitably require some transmission system improvements.

Mechanisms are also needed to allow balancing authorities to both share frequency-responsive resources and make sure that they have adequate frequency-responsive resources within their control. From a transient stability perspective, the system appears to tolerate substantial displacement of thermal generation. However, care will be needed in the event that portions of the system are driven to a near-zero commitment of coal plants.

The fact that the Western Interconnection can handle such high penetrations of renewable energy bodes well for coming years, as our nation shifts to cleaner, lower-carbon sources of energy to address climate change, local air pollution, and other concerns. It is also encouraging to see how renewable energy can contribute to frequency and transient response, resulting in a more reliable power grid.

The full report is available for free download at: http://www.nrel.gov/docs/fy15osti/62906.pdf.

More Power Engineering Issue Articles
Power Engineerng Issue Archives
View Power Generation Articles on PennEnergy.com