The proliferation of inverter-interfaced renewable generation is changing the underlying assumptions of power systems protection design, particularly for distribution systems. Due to current-limiting controls of inverters, the short circuit capacity of distribution systems is decreasing as the penetration of renewable resources is increasing. As a result, traditional overcurrent protection will not be adequate to protect future distribution systems with higher penetration of renewable energy. Additionally, higher renewable penetration levels could greatly complicate achieving proper overcurrent protection coordination.
In that regard, the case of microgrid protection design is interesting for two main reasons; first, emerging interconnection standards evolve toward intentionally islanding distributed generation to increase system’s reliability and resilience. But also, from a protection design perspective, the case of an islanded microgrid resembles, to some extent, the case of distribution systems with higher renewable penetration. Therefore, efficient microgrid protection schemes could be used in protecting distribution systems with high penetration of inverter-interfaced renewable generation.
Since the introduction of the microgrid concept, it was realized that designing efficient protection schemes for microgrids would be challenging and would require advancing the state-of-the-art of protective relaying. The main challenge facing the development of standardized microgrid protection originates from the fact that microgrids differ in their topology, generation mix, feeder sizes and types and locations of fault interruption devices. Microgrid fault current levels could be quite sensitive to generation dispatch especially in islanded mode of operation. Additionally, fault levels change drastically between grid-connected and islanded modes of operation which makes it very difficult to maintain overcurrent protection coordination for both cases. As discussed earlier, microgrids with significant renewable generation could have very limited fault current levels as a result of inverter current-limiting protection functions which typically limit fault contribution to as low as 1.1 per unit. As a result, overcurrent protection could fail completely to detect faults.
Although several non-overcurrent protection schemes were introduced in academic literature for renewable-dominated low-fault microgrids, the main protection schemes currently available for microgrids are overcurrent-based protection and differential protection. While overcurrent protection is relatively inexpensive and is widely available in the distribution system, it is not particularly suitable for low-fault renewable-rich microgrids. On the other hand, differential protection promises a robust protection solution, but at a very high cost which could be prohibitively expensive for many microgrid applications. Therefore, there is a need to develop protection solution that fill the gap between low-cost low-reliability schemes like overcurrent protection and high-cost high-reliability schemes like differential protection as depicted in Fig. 1.
Figure 1: Gap in available Microgrid Protection Schemes
The current trend to increase communication and data acquisition infrastructures in microgrids, and distribution systems in general, could provide a great opportunity to develop advanced microgrid protection schemes for renewable-dominated microgrids. Some of the advanced protection schemes which Sandia has investigated recently are impedance-based communication-assisted protection and signal processing-based protection. Impedance-based schemes rely on monitoring impedance changes at different feeder relays to detect the occurrence of faults. Different pilot-schemes were developed to locate the fault utilizing impedance and directional protection elements. Signal processing-based protection schemes rely on analyzing transient signatures generated during faults. Detection of single phase to ground faults and sensitivity to capacitor switching are some of the main challenges facing the implementation of these schemes. A combination of transient-based, voltage-based and zero sequence protection could potentially provide effective protection for low-fault microgrids.
Given the criticality of protection systems, there is a need to thoroughly study all implementation issues related to new non-overcurrent protection schemes. Hardware-in-the-loop testing represents a great tool to test advanced protection scheme and build confidence in their performance before transitioning those schemes to the filed. Additionally, integration of protection functions with existing microgrid controllers could provide advanced protection for the microgrid in a cost-effective way utilizing microgrid communication and data acquisition infrastructure.
Mohamed Elkhatib – Sandia National Laboratories
Abraham Ellis – Sandia National Laboratories
Sukumar Brahma – New Mexico State University
Leave a Reply