1143 Islanded Operation

In its broadest sense, islanding can be defined as the operation of the distributed resource and the intended user's loads in complete isolation from the utility grid — the feeder line. This definition does not discriminate between intentional (safe) islanding, where there is positive disconnection from the feeder, and unintentional or accidental islanding, where power flows not only to the direct user's load but also to the feeder (and possibly to all or some of the other loads on that feeder). Therefore, this definition must be refined when referring to protection issues.

With respect to protection (for the most part, the protection of the utility's power grid), islanding is defined as the unintended supply of power from one or more distributed power plants to a portion of the utility network (for example, a feeder line) following the separation of the feeder from the distribution network. Islanding is possible if the distributed resource controller misinterprets or does not detect the opening of the utility's feeder breaker so that the distributed power unit continues to feed power to the intended customer and to the dead feeder line. It is possible that some or all of the other loads may remain connected to the feeder. They will be fed power with possibly poor voltage and frequency regulation (that could damage some loads) until the distributed generators and/or loads are disconnected (trip) as a result of the detection of over- or undervoltage or over- or underfrequency by conventional, passive protection systems.

The quality of the voltage and frequency delivered depends on the rating of the DG that remains connected and the size of the loads being supported when the feeder breaker opened. For example, it is possible that the voltage will collapse almost immediately if the connected load demand far exceeds the rating of the connected DG. In this case, undervoltage will be detected. However, if the connected DG rating is such that it can meet the demand of the loads, fairly stable operation within the over- or underfrequency and over- or undervoltage acceptable operating limits is possible. Utility personnel dispatched to service the apparently disconnected feeder may be placed in great danger. Therefore, the distributed generator system controller must include a means to continuously and reliably test the status of the feeder line and instantly disconnect the distributed generator in the event of a tripped feeder.

The basic protection philosophy followed by utilities until now has been to limit the total capacity of interconnected capacity on a given feeder to less than about 10% of the minimum expected load on the feeder. Therefore, for any fault on the feeder it is expected that the feeder protective relays will operate, isolating the feeder and leaving the loads connected to a much smaller amount of DG. The voltage is therefore expected to collapse, resulting in the automatic shutdown of the distributed generators due to under-voltage. This basic protection approach has been successfully implemented and practiced by small, interconnected wind and PV power generators. In particular, wind turbines driving induction generators have demonstrated that basic over- or undervoltage and over- or underfrequency sensors, relays, and control logic are quite adequate to prevent islanding.

The islanding hazard can be exacerbated when more than one DG source is connected to the feeder and the ratio of the total distributed generator capacity to the instantaneous demand load is relatively large. The results of tests conducted by Sandia National Laboratories on PV inverters in 1997 showed that when several inverters (with different anti-islanding techniques) were operating on a single 120 V circuit, the inverters frequently continued to feed power from the PV arrays to the circuit loads for more than two seconds (times greater than 30 seconds were observed) following the disconnection of the circuit from the utility network (Sandia National Laboratories, 1998). In these cases, the ratio of power being generated to power being used by the loads on the circuit was in the range of 0.8 to 1.2. Interestingly, it was observed that the presence of a transformer in the circuit resulted in much shorter disconnect times, less than 0.5 seconds. This was a result of the fact that most inverters cannot supply the nonlinear magnetizing current required by the transformers. Sandia has proposed a method for designing an anti-islanding inverter that includes both active and passive techniques. The active methods are termed SFS (Sandia frequency shift) and SVS (Sandia voltage shift). The passive methods are over- or underfrequency and over-or undervoltage.

Solar Stirling Engine Basics Explained

Solar Stirling Engine Basics Explained

The solar Stirling engine is progressively becoming a viable alternative to solar panels for its higher efficiency. Stirling engines might be the best way to harvest the power provided by the sun. This is an easy-to-understand explanation of how Stirling engines work, the different types, and why they are more efficient than steam engines.

Get My Free Ebook


Post a comment