Distributed Energy Magazine

Power Quality of Microgrids

An impact study of voltage regulation, fault recovery, frequency, and harmonics

  • Email This Post Email This Post

Credit: iStockphoto
The electric power grid provides the lifeblood of modern living. It’s not surprising, therefore, that many systems on the grid are benefitting from modernization activities, including the overall design of grids. Some amazing projects are incorporating microgrids today. There have been microgrids built for smart villages, to electrify developing regions with solar power, inverters, and batteries. There are microgrids for military and mining outposts. Campus environments have utilized their own internal generation (renewables and cogeneration) to provide emergency power to several buildings. Municipalities and utilities are implementing microgrids to provide resiliency to essential services.

Emergency Backup Power and Microgrid Systems
Emergency backup power systems have long been in place anywhere there is a need to protect from consequences of a power outage—whether it be a hospital, factory, data center, emergency response, or a stock exchange. Electric power systems on airplanes and ships are systems with special, self-contained systems. In a sense, all of these systems are also microgrids. However, the driving force with newer microgrid projects is the desire to integrate renewable resources, fuel cells, and energy storage systems in new configurations. And the rapidly falling cost of solar panels is giving rise to new opportunities for buildings, campuses, and communities.

Are you subscribed to Distributed Energy magazine? Click here for a free subscription!

Use Cases for Microgrids
Microgrids are not the least expensive way of configuring a power system, unless there is no larger grid available. On islanded power systems, there are similar challenges to microgrids: generation supply, supply/load balance, controls, voltage and frequency regulation, and inertia to ride through system disturbances, among others. Now, for projects also connecting to a larger (utility) grid, the microgrid comes with additional costs: synchronization controls to connect and disconnect, generation, configuration of loads, energy storage, and microgrid controller. These additional costs are justified with an economic business case. Often, these business cases will rely on an improvement of reliability, resiliency, or power quality.

Resiliency, Reliability, and Power Quality
These terminologies sound the same, but distinguishing between them is important. In simple terms, resiliency is the ability of the power system to recover from some disaster such as a severe weather event. Residential customers care the most about resiliency because they may live some distance from the substation, and care when the power is going to be restored to their refrigerator so that they don’t lose food. As an outage extends to several days, medical care and other issues arise, and impact all types of customers. Reliability is the number and duration of annual outages affecting a consumer. A commercial facility will generally locate closer to the substation and population center, and so it expects only a few outages per year. Power quality is the overall ability for the electrical power system to do its job. It covers transient, momentary, and temporary events like switching transients, voltage sags, momentary outages, light flicker, and harmonics. Industrial customers may be very concerned about power quality as it affects their productivity. Even a voltage sag lasting for 0.1 second can cause loss of production for hours.


A microgrid will almost always improve the reliability and resiliency to its protected load—the microgrid system itself. But the power quality impacts are not so clear—what about harmonics, frequency and voltage regulation, voltage dips, or flicker? Does a microgrid always improve these aspects?

Add Distributed Energy Weekly to your Newsletter Preferences and keep up with the latest articles on distributed power, fuel cells, HVAC options, solar, smart energy systems, and LED lighting retrofits.    

Voltage Sags
The waveform in Figure 1 shows a voltage sag where about 30% of the voltage was lost for about six electrical cycles, which is only 0.1 second, literally the blink of an eye. However, facility lighting, emergency stop relays, motor contactors, HVAC drives, or sensitive test equipment may all be impacted by such a short event.

Figure 1 – Voltage sag event waveform

A voltage sag may be caused by equipment inrush: a motor starting or transformer energizing. These events are also commonly caused by a fault on a utility system, when there is a short circuit on a line. The fault may occur on an overhead line, exposed to animal contacts, lightning, accidents, or even metallic balloons. An underground line is not immune either, because the insulation may fail or there may be an accidental digging in to the line. Faults may be self-clearing, where the contamination burns away and the system heals itself. Or the fault may be sustained when a fuse or circuit breaker is required to clear the fault. To customers who are downstream of the fault clearing device, they will sustain an outage until the fault cause is cleared and the circuit is restored. But what of customers upstream of the event? They are also affected; they will experience a voltage sag instead of an outage.

Figure 2 – Customer on a parallel distribution feeder experiencing a voltage sag event

Figure 2 is a simple one-line diagram that depicts a typical utility distribution system, where a 12 kV distribution feeder experiences a fault on one of the feeders, customers fed from other non-faulted feeders will also experience a voltage sag during the clearing time (typically 4–20 cycles). The voltage sag magnitude will depend on how close in the fault is, the relative strength of the power system, and the location of the customer. In this diagram, for a three-phase fault, it is a fairly simple voltage divider calculation:

Seamless Transition?
What happens during a nearby fault if this customer has a microgrid? A common misperception is that if the microgrid has energy storage, then it should be able to island seamlessly and operate like an uninterruptible power supply (UPS) as shown in Figure 3. However, if the microgrid is connected in parallel to utility grid, then the power sources (generators, UPS, solar inverters, etc.) will all feed fault current to the event while the two systems are interconnected. An electrical device with a large fault current contribution (5–10 per unit), such as a diesel generator, may restore some of the missing voltage. But solar and energy storage system inverters are usually fault current limited, and so these microgrid components will have very little effect to restore the voltage. As a consequence, the customer microgrid system will experience the voltage sag event until the system is able to island. The islanding operation can be accelerated to less than a cycle if the system is connected via a static (power electronic switch), but this is a very expensive component, not commonly found on microgrid systems.

Response to Inrush and Fault Events
An uninterruptible supply system (UPS) at a data center, due to the inverter characteristics, is a relatively weak source. Faults on the UPS protected load panels may therefore be difficult to detect or clear. Also, the energizing of a transformer leads to 10–20 times normal currents, which is another challenge for these systems. So for UPS systems common in data centers, these events are often supported by a “UPS bypass assist” which provides the fault and inrush current boost from the grid or diesel generators to support faults and inrush current needs of the protected load system. Traditional non-inverter based (diesel) generator backup systems can rely upon a high fault current capability of a synchronous generator. But these techniques are very limited on islanded microgrid systems with a high inverter content—systems with a high content of solar power, battery energy system, and fuel cells. These inverter dominated systems will need special consideration of these events. Adaptive protection for short circuits may be necessary for the main distribution panels of a microgrid.

Figure 3 – Block diagram of an online uninterruptible power system (UPS)

Frequency Regulation
On the large interconnected grids of North America, frequency deviations of 0.1–0.2 Hz are caused by major disturbance events. On an islanded microgrid, the frequency may commonly vary in a much larger scale of 1–2 Hz or more, depending upon the inertia of the dispatched generation resources. Figure 4 shows the simulated response of one microgrid project to the largest motor start. Now on many electronic loads, frequency variation may not have much impact. But for other rotating loads and transformers, there can be important heating and control impacts. Work needs to be done to better understand the effect of frequency variations on different types of load equipment, so that practical frequency regulation limits can be developed.

Figure 4 – Simulated frequency response of a microgrid

Flicker and Harmonics
Grid strength is an important determinant in the response of the power system to fluctuating loads, motor starts, and harmonic currents from electronic loads. Inverter-based microgrid systems will be challenged to regulate voltage, avoid rapid voltage variations, and limit harmonic voltage distortion to acceptable levels. Inverter controls that regulate voltage will need appropriate droop setting and delays to avoid “hunting” instability phenomena. Inverter control systems must be hardened and dampened so that they will not be involved in harmonic or interharmonic grid resonance interactions.

Microgrids, especially systems with high inverter-based energy sources, will have many challenges to simply maintain the power quality levels available on an electric utility grid. These systems will need simulations to plan for load variations, harmonics, and inrush conditions. The controls of inverters will become more sophisticated in dealing with faults and other considerations like grid resonance interactions to meet these challenges. New work is also needed to better establish “emergency” limits for frequency, voltage, and harmonic variations for loads to safely endure without failure or degradation to equipment. BE_bug_web

Related Posts


  1. A rotary UPS with a bi-directional flywheel energy storage system can manage all of the power quality issues you identified for a Microgrid. The flywheel energy storage device MUST be capable of taking energy in as fast as it can supply it to manage power surges on the Microgrid related to block load changes. You mention that some “expensive” devices are not typically found on a Microgrid, but Microgrids for mission critical power such as data centers, hospitals, financial institutions, etc. already use “expensive” devices such as diesel generators and battery-backed UPS systems. If a rotary UPS with a bi-directional flywheel is used as the common point of connection between the normal utility power and the Microgrid, power quality will be maintained under normal and islanding modes. A rotary UPS and bi-directional flywheel system is not cost-effective for residential or commercial Microgrids, but Microgrids for critical power systems over 1MW that will require UPS are definitely possible candidates, especially when the microgrid controller is integrated with the UPS controls.

    1. Thank you for you comment. I would agree that rotary UPS is an interesting option for larger microgrid systems. However, it by itself is not a “silver bullet” that manages all of the power quality issues. Some rotary systems are combined with an isolating inductance to help support the voltage to the critical loads during parallel operation while nearby faults are cleared – but these conditions must be carefully analyzed for specific fault conditions and voltage regulation issues. Frequency response and harmonics also need consideration during islanding mode.

      For sure there are solutions to these power quality issues, and rotary UPS is an interesting option for larger systems. My hope for this article and these discussions is to increase awareness of these problems and solutions. Thanks again for your contribution to the discussion.

Leave a Reply

Your email address will not be published. Required fields are marked *