Rated Power Output Can Affect Grid for Good or Bad
By Jens Schoene, Doug Houseman and Sean Morash, EnerNex
Distributed generation (DG), such as distribution-connected photovoltaic (PV) generation, presents a number of challenges to distribution system planning and operations. Dispersed generation sources connected to the distribution system disrupt the traditional paradigm that the substation is the sole source of power and short-circuit capacity. In addition, PV is inherently variable, adding further complexities to distribution system integration. At the same time, due to its recent popularity resulting in large-scale deployment in many distribution systems, PV offers opportunities for utilities that, if leveraged, can help optimize grid performance.
The potential for PV to negatively or positively affect the grid or do both depends on the rated power output of the installed PV unit. Utilities must understand this in order to guide distribution system planning studies, interconnection requirements and grid modernization efforts.
This article discusses which potential effects can be traced to the size of the PV installation and which effects are primarily dependent on the electrical characteristics of the system to which the installation is being connected. Because both matter, a simple assessment that is solely based on PV size is meaningless.
PV’s Negative Impacts on the Distribution Grid
The detrimental impact that PV potentially has when installed in large numbers on an unprepared grid includes:
1. Over-voltages on distribution feeders that are outside permissible limits and/or cause tripping of PV inverters
2. Voltage sags and swells
3. Flicker caused by the variability of DG-generated power and inrush currents due to cloud movement
4. Reverse fault current flow causing existing protection schemes to be uncoordinated or possibly non-functional
5. Increased wear on utility equipment such as voltage regulators and capacitor switch tap-changers
6. Unintentional islanding
8. Real and reactive power phase imbalances
Many of these issues were seen in the results of a utility survey EnerNex conducted recently. The results were presented at Centre for Energy Advancement Through Technological Innovation (CEATI) International in Montreal, Canada in August 2015.
Uutilities with which EnerNex has worked also have experienced mitigating issues due to their high PV penetration on distribution feeders. These utilities already are experiencing more frequent operation of voltage regulators on their high-PV penetration feeders leading to greater maintenance requirements and PV inverters disconnecting frequently due to overvoltage conditions.
Simply adding PV to the system without preparing for the consequences will almost certainly result in an escalation of the PV-caused issues that already exist on the feeder, and in the emergence of additional issues once even higher PV penetration levels are reached.
Leveraging PV for Grid Support
The distribution system can receive a number of potential benefits from inverter-based DG, such as PV. Inverters can facilitate the introduction of advanced control and communication technology. Much industry discussion has centered on so-called “smart inverters” that provide advanced capabilities that can be used to optimize grid performance. These capabilities can be activated through autonomous control such as built-in Volt/VAR curves that determine active and reactive power injection/absorption based on the condition at the PV terminal. In addition, and even better, is distributed control in which a number of PV systems are controlled in a concerted fashion to achieve a pre-determined optimization goal (such as flat voltage profile, overloading prevention, shaping of load curve, etc.).
Such voltage control capability could not be implemented to support distribution systems and mitigate DG impacts in the U.S. until IEEE Standard 1547 was amended in 2014. This is because other voltage regulation equipment is not ready to be coordinated with smart inverters. In fact, voltage might not even be measured along a distribution circuit. This creates no basis to control the resource, not to mention to remunerate it.
Fast, low cost monitoring and communications systems are becoming available, however, and can form the basis for control systems that use both traditional distribution equipment and solar inverters. To make a business case for (or against) such grid modernization investments, evaluations that accurately assess both the PV-caused negative effects and PV-provided benefits are needed.
High Penetration PV and Reverse Power Flow
Increasing the size of residential-scale PV from small units that have a rated power of a few hundred watts to large units that can supply up to tens of kilowatts has a number of technical implications. Two scenarios that should be closely considered are:
1. The scenario at which PV production exceeds the load demand at the residence but the net power flow through the line conductors is always less than the net flow in the absence of PV.
2. The scenario at which PV production reaches a level at which the line conductors carry more power than they would carry in the absence of PV.
In scenario No. 1, localized reverse power flow occurs but, due to its low magnitude, there are little to no negative impacts on grid operation and even some positive effects, such as reduced losses, reduced line/transformer loading and a flatter voltage profile along the feeder.
Scenario No. 2 is of more concern for grid operation because increased power flow will augment losses and line/transformer loading and could cause overvoltage conditions. The graph on page 25 was created from solar production and load consumption data recorded in Michigan for an entire year. It illustrates the frequency of occurrence of each of these two scenarios for various PV sizes. The solar production data was normalized and then scaled based on assumed PV array sizes ranging from 500 W to 30 kW. This creates two scenarios. One in which the entire energy produced by the PV systems is always consumed by local loads and reverse/increased power flow never occurs (for small array sizes). The other is when PV systems sometimes overproduce energy resulting in reverse or increased power flow. For instance, PV sizes of up to 1 kW rarely caused reverse power flow and never caused increased power flow, while a PV size of 5 kW causes reverse power flow 29 percent of the time and increased power flow (in addition to reverse power flow) 17 percent of the total hours within a year. In other words, with a relatively small 5 kW PV size, there is potential for negative impacts (e.g., increased losses, voltage problems, equipment overloading, etc.) for 1490 hours each year.
PV Caused Transient and Power Quality Problems
While overproduction caused by large PV sizes is a strong indicator for certain types of PV impacts, overproduction is less of an indicator (or none at all) for other types of impacts. For instance, system characteristics such as line lengths, resonance conditions and strong/weak source, play a major role for the existence or absence of PV-caused power quality problems such as voltage fluctuation, transient overvoltages (TOVs), harmonics, and more. In addition, the size of the PV will still have an impact. For example, a 500 kW commercial-size PV system would cause a more pronounced voltage sag if it is suddenly shaded by a quickly moving cloud than would a 5 kW residential PV system. Determining if this voltage sag is actually of concern is more complex, however, because its severity depends on many factors. In this example, neither of the voltage sags caused by the large and small PV systems would likely be noticeable if the PV systems are connected to an urban feeder supplied by a strong source. It might be noticeable, however, on a typical rural feeder. Detailed engineering studies are required to evaluate these kinds of impacts.
Summary and Conclusion
It is important to avoid PV-caused increased power flow as it can cause excessive losses, overloading conditions and voltage violations. The presence of increased power flow can be readily determined for different PV sizes from PV production and load consumption data. While increased power flow is an indicator for PV-caused problems, evaluating whether or not these problems actually occur requires additional steps, which should be relatively straight forward to develop by determining correlations-increasing the PV size by “x” will increase the power flow through the line by “y”, which will in turn increase the losses through the same line by “z”. On the other hand, the evaluation of transient and power quality effects and whether any infrastructure investment is needed requires detailed computer simulations, which are complex and nontraditional in that they need to account for the secondary system (from the distribution transformer to the PV installation), explicitly represent each PV unit (no aggregation) and load (aggregated to the customer level), and, finally, simulate an entire year with time steps of one minute or less.
Authors’ Note: Our paper titled “Photovoltaics in Distribution Systems – Integration Issues and Simulation Challenges,” which was presented at the 2013 IEEE/PES General Meeting in Vancouver, Canada, elaborates on these simulation challenges.
Jens Schoene, Ph.D., is the director of research studies at EnerNex. His areas of expertise are the transient and harmonic analysis of power systems, distributed generation interconnection studies, induction studies, lightning studies and arc flash studies.
Doug Houseman, is vice president of technical innovation at EnerNex. Houseman works with clients all over North America and Australia on issues related to smart grid/metering/homes and other related issues. He also works with regulators, utilities and vendors in the market to help move the industry to the next generation grid.
Sean Morash, is a consultant at EnerNex. Morash produces solutions through research based on a working knowledge of smart grid-related applications, including communication technologies and protocols, advanced sensing and control, renewable energy, electrical, mechanical and information systems integration, enterprise information architecture, cyber security, information modeling and related disciplines and methodologies.