The Challenges of Microgrid Deployment
Back to Top
A successful microgrid implementation starts with effective front-end engineering design, which determines the distributed-generation parameters and any appropriate modifications to the existing distribution system. This becomes even more clear when we start to consider how much of our existing design standards assume that a large utility source is powering the feeder and how removing the presence of the utility drastically changes the new microgrid power system.
To illustrate this, let’s take a look at some of the most common design elements on a distribution system, starting with protection coordination.
A successfully coordinated system is intentionally designed to minimize outages by isolating a fault as close as possible to the fault’s location. To do this, each overcurrent-protection device on a feeder responds a little faster and more sensitively the farther it is from the substation. The fault current on such a system is typically in the thousands of amperes, and in many systems the concern is that the fault current grows over time to exceed the physical capability of the distribution equipment.
What happens in a microgrid environment, specifically, when the feeder from the utility system is disconnected?
Because the fault-current contribution of distributed energy resources is inherently much lower than the utility contribution (which is connected to an entire transmission system’s worth of generation), the fault current on an islanded feeder is going to be orders of magnitude less than the load current. In many cases, a fault on an islanded microgrid could be of comparable magnitude to load current, challenging protection schemes that use typical overcurrent-protection devices. A fault occurring on such a system may never be cleared by a traditional fuse, and most microprocessor-based overcurrent-protection devices are not set sensitively enough to detect such a fault in the first place.
If we were to just implement a microgrid on a feeder without considering the protection-coordination implications, our protection would no longer coordinate. A fault anywhere on this system would cause all the distributed generation to trip, thereby de-energizing the entire feeder.
To combat this, the existing protective devices on the system should be evaluated for their suitability to this application, for which they were not originally intended. This is a complex engineering process, and there is no single solution that can be implemented across all microgrids, given the unique mixes of loads, generation sources, and existing infrastructure. The process to overcome this challenge starts with expertly evaluating the utility’s system, the current protective equipment on site, and a thorough understanding of how the microgrid is expected to operate.
Another commonly overlooked problem when applying microgrids to the distribution system is what happens during start-up when in island mode.
Normally, the utility source allows most distribution feeders to be energized with relative ease because the inrush current from the transformers and loads is easily distinguishable from the utility fault current. Most energization procedures on a tradition feeder are to ensure the system is safe to energize (e.g. faults have been repaired as needed), then closing the open breaker. But in a microgrid application, the system designer must understand the relationship between the generation-equipment capabilities and the loads on the system to determine whether more complicated procedures must be established.
When starting a microgrid in island mode (operating independently from the grid), the system inrush current may cause significant deviations in system frequency and voltage, and it could cause the generator protection to trip offline during startup. Mitigation techniques involve specialized analysis of the types of generation selected to energize the microgrid during an islanding situation and reengineering their controls for microgrid applications. Furthermore, the protection must be capable of distinguishing energization scenarios from fault scenarios to ride through the former and trip quickly in the latter scenario. This only happens through thoughtful design and with a thorough understanding of the system.
Balance between generation and loads also has to be continuously maintained throughout the operation of the islanded microgrid. Changing loads, especially large block loading, can have a more dramatic effect on overall stability on the islanded system than when grid connected. This is caused by the load changing typically being a much larger percentage of the available full output power of the generation (by orders of magnitude). Subsequently, generation must be intentionally selected, and the microgrid control system must be fully capable of monitoring system stability and dispatching/curtailing generator production as needed to maintain system stability.
Designing the feeder to operate without a strong utility source being present is one of the largest challenges of implementing a successful microgrid, especially because the industry has decades of established design philosophy assuming a strong utility source. As microgrids become more commonplace, this is a rising problem that often goes overlooked until it is too late, resulting in projects that drastically exceed their budgets and schedules to fix the problem or, even worse, fail to be brought online.
I'd be interested in learning your thoughts on the challenges you face in implementing a successful microgrid.
This blog originally appeared in the IEEE February E-Newsletter.