What is causing interconnection delays for solar EPC firms in California and what can industry stakeholders do to fix the problem?
Thanks in large part to the recent influx of solar distributed energy resource (DER) generation, the California utility grid is becoming saturated much faster than anyone ever expected. Saturation in this context refers to the maximum solar capacity beyond which the electric power system becomes unstable, which varies circuit by circuit and also has holistic implications at the system level. Investor-owned utilities (IOUs) currently lack many of the data sets and analytical tools needed to properly assess the impact of DER saturation. As a result, many projects are languishing in a seemingly endless cycle of interconnection review studies and grid upgrades. In this article, I explore how we got here, illustrate some of the interconnection challenges and consider some potential remedies.
In spite of the recent rise of DER generation—which includes demand management as well as distributed energy storage and variable renewable resources—the basic architecture of California’s electric power system remains much the same today as it was 100 years ago: a relatively small number of centralized generation sources push power out to electric consumers across a vast long-distance network of high-voltage transmission lines and medium-voltage distribution lines.
How Did We Get Here?
Though this system has served California fine to date, it is not particularly well suited to meeting the state’s future needs, which are changing due to shifts in environmental and energy policies. More Than Smart is a nonprofit organization whose mission is to help policymakers and industry stakeholders transition to a 21st-century electric power system. The authors of its report “A Framework to Make the Distribution Grid More Open, Efficient and Resilient” (see Resources) note: “There is a recognition that California is at a crossroads with respect to the future role of the electric system generally and the distribution system specifically.”
The rise of virtual net metering. With the implementation of net energy metering (NEM) and California’s Rule 21, which details interconnection, operating and metering requirements for DER projects, the state’s IOUs enabled customers to generate renewable power for use in offsetting a specific load. The interconnection method used to apply for the NEM tariff was typically a line- or load-side tap, landing behind an existing meter. In general, this interconnection did not create a problem for the local utility system because the customers consumed most, if not all, of the generation at the source, and backfeed across the point of interconnection was minimal.
However, in 2014, the advent of net-energy metering aggregation (NEMA) changed the dynamic because it permitted load offset using virtual net metering. Generation customers can now create power anywhere on their property and export it directly back to the electric grid. The utility then nets that generation against their aggregated electrical consumption over one or more meters. The customer pays the difference or, in the instance of a surplus, receives a credit to offset future consumption.
On paper, this seems straightforward, and it has allowed customers to install larger solar projects to offset multiple meters. The problem this creates for IOUs is that, while the load is virtually offset, they must study a stand-alone generation system as an exporting facility. While the rules on how the utility compensates the customer differ from those for a traditional exporting facility, the overall impact on the utility distribution system is the same as if the facility were an export project.
In addition, the advent of NEMA promoted the use of the NEM tariff on a much broader scale than previously anticipated. The benefits of net metering are no longer limited to small roof-mounted residential systems; the door is now open for much larger ground-mounted projects all the way up to 1 MWac to interconnect under the NEM tariff. As shown in Figure 1, the rate of nonresidential NEM interconnections across all IOUs has grown 246% since 2014.
Integrating increasing amounts of DER generation is an inherently challenging project. The authors of the Electric Power Research Institute (EPRI) report “The Integrated Grid: A Benefit-Cost Framework” (see Resources), note: “The question is about the ways in which DER interacts with the power system infrastructure. The formula for this answer has multiple dimensions. Beneficial and adverse circumstances can arise at differing levels of DER saturation. The interaction is dependent on the specific characteristics of the distribution circuits (design and equipment), existing loads, the time variation of loads and generation, environmental conditions, and other local factors. Benefits and costs must be characterized at the local level and the aggregated level of the overall power grid.”
With this in mind, we have to expect long interconnection timelines with large non-NEMA projects, as reliability engineers must extensively model the impact these systems have on the electric power system. However, very large PV systems can connect directly into the transmission system, bypassing the distribution system altogether. Though NEM and NEMA projects initially experienced shorter engineering review timelines, these projects are now hitting a wall in areas where IOUs deem the existing utility infrastructure inadequate to support additional DER generation.
In effect, a large portion of the electric power system is not designed for back feeding from the distribution systems into the transmission level. The antiquated protective devices at the substation and transmission level are designed for power flowing to the distribution system, not from it. As a result, increasing numbers of relatively small DER projects fail to pass the Rule 21 fast-track interconnection review process, and then they must undergo detailed studies with significantly longer timelines.
During a recent distribution-level substation walkthrough, I had an experience that offered a microcosm of the situation we are in. The purpose of the visit was to help a client understand the IOU’s proposed upgrades related to direct transfer trip (DTT), a protection scheme that manages unintentional islanding—and that can do so faster and more reliably than the anti-islanding protections in an inverter. (Islanding refers to a situation where an inverter-based distributed generator continues to energize a power-system circuit.) The utility’s protection engineers were concerned that an additional 1 MW of distributed generation (DG) could back feed the substation and support an unintentional island.
During this walkthrough, the person charged with managing the substation on a day-to-day basis pointed to a voltage regulator and said: “That’s going to be a problem.” This statement confused all of us until the substation manager explained that the voltage regulator had been installed 70 years ago. This meant it had analog controls, which could not accept the digital commands that utility operators need to send to a voltage regulator to enable DTT. Admittedly, I initially thought the substation manager was exaggerating the age of the equipment, but when we walked over to the voltage regulator, he pointed out the year printed on the nameplate: 1946.
Though it is just one piece of hardware, this voltage regulator is emblematic of the challenges associated with transitioning a 20th-century power system into a 21st-century infrastructure. Some of the technology that the utility employs, while perfectly functional, is antiquated. When the utility installed that voltage regulator back in 1946, nobody could have dreamed that a solar power plant would necessitate its replacement in 2016. This single voltage regulator serves hundreds, if not thousands, of customers. Taking it offline, while manageable, requires a substantial amount of coordination within the utility system to ensure no loss of service. The system operator will need to take lines out of service, reroute feeders and so forth. There is no room for error in a system where reliability is of the utmost importance. To manage these risks, the utility typically conducts these change-outs when the load is at a minimum so the potential impact is as low as possible, which in this case is between the months of November and February.
In this particular case, the IOU conducted a detailed study related to the proposed project in March 2016 per Rule 21. Since the IOU identified the need for specific substation upgrades, we requested that the utility place the project in the 2016–17 upgrade window. However, projects in the queue from the previous year had reserved all of the spots in this year’s upgrade window. As a result, the upgrades required to interconnect this project will have to wait until the 2017–18 upgrade window.
This sort of delay stacking is directly due to DER saturation, and antiquated equipment compounds the problem. Had there been enough load to offset the combined generation, the vintage voltage regulator could have gone right on functioning into perpetuity. Instead, what would normally have been an interconnection process of 3–6 months became a project of 18–24 months. It is important to keep in mind that we are not talking about a 50 MW utility-scale project, but rather a NEMA project with a generating capacity of less than 1 MW.
These timelines impact system owners by delaying their ability to realize returns on their investment. Without these returns, loan terms are subject to higher rates and the financial viability of a project drops significantly. As a result, many owners choose to walk away from projects rather than wait out a substation upgrade, which has a crippling financial impact on the solar industry in California and severely slows down DER interconnection.
Amanda Johnson is the utility interconnection manager at JKB Energy, a solar integration firm specializing in commercial and agricultural solar projects in California’s Central Valley. Johnson has firsthand experience with interconnection delays. “Many of our customers are subject to substation upgrade requirements when building solar projects that are close to the upper end of the 1 MW limit for a NEM or NEMA project,” she says, “despite the fact that these customers pull the same amount of energy from the grid throughout the year.” Whenever this happens, interconnection timelines increase to allow for the necessary engineering reviews and equipment upgrades.
According to Brad Heavner, CALSEIA’s policy director: “Trend lines for large and small systems have been moving in opposite directions. While the California IOUs have done a great job of automating the interconnection process for standard rooftop systems, new roadblocks keep emerging for large systems. We used to hear complaints about 4-month delays turning into 9-month delays, but we are now hearing about 2-year delays.”
Part of what makes these scenarios so commonplace is an outdated way of approaching distribution network design and upgrades. Utilities can no longer stop at the transmission level in envisioning the utility grid as a network. For elevated levels of DER penetration to become sustainable, the network platform mentality must percolate down to the distribution level.
Think of the transmission system as a freeway designed for one-way traffic flow, where substations are off-ramps to the smaller feeder roads that represent the distribution system. With the rise of DER generation on these distribution circuits, traffic suddenly wants to flow back onto the freeway. This two-way traffic flow disrupts the system, causing congestion as well as reliability and safety concerns. To mitigate these problems, the system operator must begin an extensive infrastructure upgrade process to convert each of these off-ramps into combination on- and off-ramps. The process is costly and time-consuming; we have all experienced how freeway upgrades tend to make traffic worse before it gets better.
This is analogous to the interconnection challenges in California today, with one big traffic jam of DER generation projects all trying to get on the same freeway at the same time. This situation leads to frustrated customers and impatient solar companies. It is unfair, however, to place the blame for the problem on the IOUs. Increased participation in the NEMA tariff program has resulted in an ever-increasing number of applications to interconnect. Compounding the problem, the amount of DER generation simply overwhelms the existing distribution system. When I asked CALSEIA’s Heavner which specific area seems to be the most congested, he responded: “California.”
Fresno County, a major hub in PG&E’s utility system, is a good case study for this phenomenon. As shown in Figure 2, Fresno County is a hotbed for nonresidential solar interconnections.
The rate of interconnection for nonresidential NEM projects in Fresno County has grown more than 250% since 2014. Current interconnection processes cannot keep pace with the upgrades required to interconnect these projects in a relatively short time frame. This is simply a case of too much solar, too soon.
While California’s environmental and energy policies envision integrating more than 15 GW of DER generation into the state’s electric power system, these projections may underestimate the grid transformation that is underway. Because of the scale and the capital-intensive nature of the grid investment required to meet these goals, stakeholders must plan and invest wisely. The More Than Smart report elaborates: “As distribution infrastructure is largely depreciated over several decades, investments in this decade may need to be useful to 2040. The implication for California is that the current annual utility distribution investment of nearly $6 billion is effectively a 25-plus year bet on a future [that] will likely be quite different than we can imagine today.”
Creating a sustainable 21st-century grid requires mixing and matching new data with emerging technologies. While many potential solutions are on the horizon, there is no silver bullet. The end solution will probably come from all of the stakeholders, each working to solve small pieces of the overall puzzle. When used in conjunction with rapidly advancing smart grid communications, this combined contribution will have a substantial impact on the overall problem of excessive DER congestion.
Better modeling can generate the data needed to identify weaker areas of the utility grid and incentivize solar DER development in the appropriate areas. R&D roadmaps should include allocations to understand new phenomena such as islanding, and IOUs and regulators should implement solutions to these problems in a reasonable and targeted manner. Smart inverters have the potential to provide the level of control that IOUs require to curtail and prevent overgeneration. Energy storage systems can provide customers with an unprecedented ability to control when and how their systems produce energy.
Given the scale of the existing power system and the investment necessary for its modernization, the authors of the More Than Smart report argue that distribution planners should start by evaluating the existing system and developing a baseline model of its capabilities. Once this exercise is complete, distribution planners and reliability engineers can stress-test this baseline model against a set of future scenarios. By identifying capability gaps in this manner, stakeholders can determine the grid upgrades that provide the best return on investment.
The More Than Smart report explains: “Analysis today requires both the traditional power engineering analysis, as well as an assessment of the random variability and power, flows across a distribution system. Such an analysis would include real and reactive power flows under a variety of planned and unplanned situations across a distribution system, not just a single feeder. Evolution to a more network-centric model for a distribution system to enable bidirectional power flow underscores the need for a fundamental shift in planning analysis.”
To facilitate this shift, California Assembly Bill 327 calls on IOUs to develop distribution resource plans that identify optimal locations for DER deployment, as well as ways to optimize the value of these resources. As part of these efforts, IOUs and other stakeholders are participating in working groups, which More Than Smart is facilitating, to establish two new planning tools: locational net benefit analysis (LNBA) and integrated capacity analysis (ICA).
Laura Wang, project director for More Than Smart, explains: “We envision that the IOUs will use both tools [LNBA and ICA] to meet the objective of Assembly Bill 327, which asks IOUs to determine optimal DER locations on their distribution systems. Developing these tools is expected to be an iterative process. IOUs will continue learning from the implementation process, the working group will refine the methodology as smart inverters become standard, and the [California Public Utilities Commission (CPUC)] makes a final decision on how these tools should be used.”
One challenge to overcome is that there is no consensus on when and how to compensate solar DER producers for avoided costs. According to a February 1, 2017, CPUC memo: “[The LNBA Working Group emphasizes] that the LNBA addresses the narrow question of evaluating DERs in single locations against certain distribution upgrades that are already in IOU distribution system plans, and should not be construed as the advancement of a comprehensive, location-specific utility avoided–cost calculator that could be used to proactively identify high-value locations for DER deployment.”
The ultimate goal of the LNBA is to identify the locational value of DER, ideally at a granular level based on a model of the entire electrical system. While a lot of work remains to be done, the LNBA tool will one day determine how DERs are compensated in California. Whereas IOUs can currently react to power quality issues only after they become a problem, the LNBA will eventually allow IOUs to identify problem areas in advance and promote specific DER functionality in certain markets by setting a premium price for these services.
The ICA, meanwhile, is unlike any utility analysis ever attempted and thus will undergo slow and methodical development. At the conclusion of this process, the ICA will provide each IOU with the ability to quickly study any part of its grid and will allow stakeholders to swiftly assess a proposed interconnection location. The first objective is to identify how much DER generation developers can add at any interconnection point on the distribution system. The second is to bring DER generation into utilities’ annual planning of the distribution system by identifying the best sites for future DER development.
These analyses represent a paradigm shift in the way IOUs view distributed generation, as they will transform solar DER projects from a liability to an asset on heavily congested portions of the utility grid. The LNBA will ultimately determine future NEM rates. The ICA will transform a reactive interconnection process—wherein developers are subject to the long review timelines in the Rule 21 tariff—into a proactive process. Gone will be the days of waiting for months only to find out that your proposed system location is on the worst possible distribution line in an IOU’s service territory.
IOUs need to devote greater resources to understanding new problems that arise from expanded transmission network penetration. One example of a relatively new, and often misunderstood, the problem is islanding. With the implementation of UL 1741, which requires all certified inverters to shut down within 2 seconds of grid de-energization, reliability engineers generally agree that there is no need to worry about an island forming and continuing in perpetuity. What is less clear is the extent to which they need to worry about the impacts of a temporary short-duration island.
Overcurrent protective devices manage fault conditions by sensing when a sharp surge in current hits a circuit over a brief period. The problem is that if a fault occurs during a temporary island, the circuit has already been interrupted; therefore, there is no device available to clear the excess current. While a few extra seconds of inverter operation might not seem like a long time, it is more than enough time to generate overcurrent conditions that could damage electrical equipment and negatively affect power quality.
The safety and reliability tests in Screen P of the Rule 21 engineering review analyze the potential for temporary islanding. (See “Rule 21 Engineering Review Process,” pp. 32–33.) These tests are relevant to the interconnection discussion because many projects that fail Screen P face longer interconnection timelines. This is so the IOUs can design and implement DTT, SCADA visibility reclosers or other mitigation techniques to negate the threat of extended run-ons at the substation or transmission network level. The issue of temporary islanding begs further analysis. We need to understand not only the conditions required for a temporary island to occur but also the potential implications.
In an August 2016 project report for the California Solar Initiative (see Resources), General Electric Energy Consulting recommends five updates to the PG&E interconnection process:
To date, PG&E has only implemented items 4 and 5 at any level. This situation highlights a larger problem: IOUs are mandating major upgrades based on an incomplete understanding of new phenomena. In addition, the IOUs mandate these upgrades with no oversight from the CPUC. For its part, the CPUC has no technical personnel on staff and therefore lacks the resources to adequately assess and review such issues.
In initial review: Raise the screening limit from 15% peak load to 60% of the estimated simultaneous load; the estimated simultaneous load will be based on conversion factors as was defined and implemented in Task 2 Report: Statistical Analysis of PV Generation and Load Balance.
In supplemental review: Keep the existing minimum daytime load screen when SCADA data is available, and allow 80% of the estimated simultaneous load by maintaining the power factor of the section below 0.98 inductive.
In detailed review: Allow up to 105% of simultaneous load by detuning circuits to maintain the power factor between 0.95 and 0.98 inductive, to address islanding concerns if needed.
In protection requirements: Modify the Direct Transfer Trip exemption bulletin to enable the quick interconnection of certified inverters rated less than 1 MW if there are no significant machine-based generators on the island.
In protection requirements: Eliminate reclose blocking for all certified inverters by lengthening reclose time on high-penetration feeders to 10 seconds.
A California bill signed into law in September 2016 establishes an expedited review process for such circumstances that includes an independent engineering expert. However, the state legislature has not approved the funds needed to enact this bill, so the IOUs will continue to have unchecked power to implement such policies.
For their part, IOUs say that they make many of their policies, such as anti-islanding, with an abundance of caution. However, the CPUC should encourage IOUs to take a closer look and analyze whether their determinations are appropriate or whether they have gone overboard because they lack the data for an informed decision. In addition, the solar industry must request, encourage and support technical policy review when new data supports a revised approach.
Most UL 1741–certified inverters are capable of much more than just producing kilowatts. Technicians can program them to absorb or create reactive power and to play an active role in power-factor correction. In a more traditional sense, DER projects can also supplement existing power plants during periods of high demand. The missing piece of the puzzle in the current interconnection model is a lack of understanding as to where we need these services and under what circumstances we should implement them.
On utility-scale projects, dedicated substations provide a secure, reliable means for utility visibility, control and curtailment of PV systems. This is not the case on smaller DER systems because they typically do not connect through a substation. As a result, IOUs have no visibility into the quality or quantity of power those DER systems are producing. At a basic level, an IOU cannot even confirm whether the system is producing power at all. Smart inverters can help bridge the gap and allow IOUs some basic functional control of and visibility into DER systems.
Seizing on this opportunity, in 2013 the CPUC, IOUs, and solar industry stakeholders convened a working group to explore the role that smart inverters can play in easing grid congestion. In a June 2016 Solar Builder article (see Resources), Brian Lydic, senior standards and technology engineer at Fronius USA, explains: “Seeing the need for not only frequency tolerance but grid-supportive functions in general, the California Public Utilities Commission and the California Energy Commission convened the Smart Inverter Working Group (SIWG) in early 2013 to start developing recommendations of technical requirements for inverter-based DER in California.”
A recent result of the SIWG is the implementation of UL 1741 Supplement A (SA). This supplement is a step toward allowing IOUs the visibility and control they need to handle high levels of DER penetration. Beginning on September 8, 2017, all California IOUs will require the design of new Rule 21 solar applications around inverters certified to UL 1741 SA. This supplement specifies an enhanced testing protocol that UL describes as an “advanced inverter grid support utility-interactive test plan” that addresses anti-islanding (with advanced features active during test), low- and high-voltage ride through, low- and high-frequency ride through, a must-trip test, ramp rate (normal and soft start), specified power factor, volt and VAR modes, and optional tests including frequency watt and volt watt. These functions have the potential to turn highly congested areas of DER from a burden to a blessing.
Additional controllability is theoretically a great asset in the context of managing overall grid congestion, but the reality is that changing operational parameters will result in lower returns on investment. The CPUC needs to explore how and when to compensate customers for grid support functionality. Lydic concludes, “In general, any of these changes will allow for higher-penetration levels and thus benefit the PV industry, but care must be taken that revenues are not unduly affected.”
Security is another area of concern, especially with regard to communication and control protocols. Tying inverters or any portion of a DER system into the utility SCADA system immediately opens a channel for hacking and potential security breaches. A safe, reliable communication system can help mitigate these risks. Any smart inverter certification program must specify a simple, widely implementable communication protocol. Ideally, the circumstances that activate grid-stability functions (and possibly reduce production) in San Diego will be the same as those in Sacramento. The CPUC should clearly define how IOUs exercise smart inverter functions and ensure consistent implementation of these standards throughout California.
Energy storage systems
As shown in Figure 3 (p. 30), increased solar generation capacity is changing California’s daily power production curve. A 2013 California Independent System Operator (CAISO) report first identified the duck curve. This term describes the shape of the daily power production curve due to periods of significantly lower electrical demand in the middle of the day followed by a steep ramp-up in the afternoon and early evening. This profile stands in stark contrast to the traditional two-peak bell curve model of power consumption, where power peaks during the midday hours, drops a little and then ramps back up during the early evening.
A side effect of identifying the duck curve is that IOUs are asking the CPUC for a change in the time of use (TOU) periods, allowing them to charge more for energy during the ramp-up period in the late afternoon rather than in the middle of the day. This change in TOU periods will result in a significant reduction in the value of power for NEM customers but will help IOUs reduce peak afternoon demand.
Energy storage is another way for utilities to flatten the duck curve. Storage can both pull up the trough of the curve (the duck’s back) and push down the peak of the curve (the duck’s tail). It is possible to deploy both large-scale and distributed-scale energy storage to address California’s duck curve. On a large scale, for example, the utility could divert excess solar generation into utility-scale energy storage systems at the substation or subtransmission level. These front-of-meter storage assets can reduce the overall impact of DER backfeeding onto the transmission network and allow IOUs to store reserves for use during the late afternoon or early evening ramp-up periods. A complementary option is to use smaller-scale energy storage systems at the customer level. Solar-plus-storage systems are a perfect fit for these behind-the-meter applications.
Advances in power system analyses, such as the ICA tool, are going to prove crucial for identifying the best areas in which to apply distributed energy storage resources. An IOU could install a large battery bank, for example, and then incentivize solar development in that area. Essentially, the IOU would be creating a giant energy storage reservoir and asking the solar industry to help fill it. The overall effect would be greater solar development and a smoother, more stable load profile, creating a win-win scenario for IOUs and solar developers alike.
No solar developer wants to tell customers that their small ground-mounted system could potentially have a 3-plus–year delay in interconnection. Sometimes this delay does not present itself until the developer is already 6 months into the process. Customers often do not understand or have the patience for such setbacks. However, areas of the distribution grid that were once prime DER locations have now become saturated, resulting in nightmare scenarios for developers and customers alike.
Johnson at JKB Energy shares an example: “Not only do many customers experience substation upgrade requirements, but also some have had to upgrade the same substation more than one time. On multiple occasions, existing customers have sought to install additional solar capacity on a distribution circuit they were already interconnected to only to face additional upgrades—in some cases, within 12 or 18 months of the previous substation upgrade.” She continues: “Instead of mandating that the IOUs anticipate future solar generation capacity, the CPUC requires prompt distribution upgrades based on the existing interconnection queue. The lack of pre-emptive upgrades on the utility substation infrastructure has caused significant impacts to customers’ bottom lines.”
Developers and EPC firms should do everything in their power to set reasonable and accurate interconnection timelines, which may mean informing customers that the path to interconnection could be a long and arduous journey. One useful tool in setting expectations is the revamped Rule 21 pre-application report. For a nominal fee, companies can get access to grid minimum loading data, utility equipment sizes, utility equipment ratings and other real-time site-specific utility data. With some technical understanding, they can use these data to identify distribution and service level upgrades prior to submitting an interconnection application. The data contained in the expanded pre-application can be crucial in the early determination of the overall interconnection cost and timeline.
Engagement. When I speak to colleagues, many of the same topics come up across the industry: new tracking technologies or magic widgets that boost production, innovative ideas on how to save a buck in wire management and so on. Many people simply do not understand that grid congestion is a major issue. What good does a fancy new tracking system with impeccably concealed wiring do if you cannot plug the project in?
To develop and advance solutions to California’s interconnection problems, we need more industry engagement. The more we debate, write about and analyze these obstacles, the faster we can overcome them. Companies that participate in the discussion have a distinct advantage because they will understand the coming changes. They can proactively develop the business mechanisms to get the most out of future innovations while everyone else is just playing catch-up.
Kenneth Sahm White is the economics and policy analysis director at the Clean Coalition, a nonprofit working to expedite the transition to a 21st-century energy system. White explains: “We are actively engaged in regulatory policy development through official proceedings and working groups related to procurement, interconnection, full valuation, planning, pilot programs, and development of markets, tariffs and compensation for the range of services that distributed energy resources can provide to customers, utilities, and transmission operators. We advocate for policies and programs that typically overlap with growth and cost reduction in all renewable energy industries.”
Groups such as the Clean Coalition and CALSEIA are instrumental in consolidating the solar industry’s concerns and presenting a clear and effective message to the appropriate parties at the CPUC or in the IOUs. These groups also provide training and discussion forums to develop proposals for policy changes that impact the entire industry. Heavner emphasizes the importance of working together as a unified industry via groups like CALSEIA: “We need everyone to join forces to take solar to the next level.”
Tim McDuffie / CalCom Solar / Visalia, CA / calcomsolar.com
Electric Power Research Institute, “The Integrated Grid: A Benefit-Cost Framework,” February 2015
General Electric Energy Consulting, “California Solar Initiative Final Project Report: Quantification of Risk of Unintentional Islanding and Re-Assessment of Interconnection Requirements in High Penetration of Customer-Sited PV Generation,” August 2016
Lydic, Brian, “How California’s Rule 21 Inverter Requirements Expand Grid Capacity and Limit Energy (Revenue) Generation,” Solar Builder, June 28, 2016
More Than Smart (previously the Greentech Leadership Group), “A Framework to Make the Distribution Grid More Open, Efficient and Resilient,” August 2014