There is a lot of optimism about the potential for the Smart Grid. That optimism is justified because communication and control technologies have advanced tremendously in the last decade. Also, the carbon reduction goals will require more renewables, more distributed power, and more load management. Each of those innovations will require more intelligence in the way we manage the grid.
Unfortunately, adapting the current utility industry to these changes is an enormous challenge. The utility industry has good reasons to be extremely risk-averse. In most industries if you misjudge supply and demand your customers experience a little inconvenience, such as a busy signal for their phone. In the electric utility industry that match has to be exact every moment or a blackout can spread across multiple states at once. Yet, the advent of the Smart Grid requires experimentation on a level that is unprecedented in the industry.
Isolated utilities present an ideal test bed for these technologies. First, they are relatively small and not interconnected, so mistakes can’t cascade across regions. Secondly, they frequently burn diesel or fuel oil. Their energy costs are three times greater than larger grids, so they have a greater motivation. Finally, with just a few wind turbines they can very quickly reach levels of renewable penetration that will take large utilities decades to reach. Smart Grid technologies can be introduced and perfected on isolated utilities with much fewer complications.
The irony is that isolated utilities have never before been trend setters. Culturally, they are the least likely place to expect cutting edge innovations. The typical utility models are too difficult for small utilities to use, so the National Renewable Energy Laboratory created the HOMER model (Hybrid Optimization Model for Electric Renewables) to make it easier to see how conventional generation, renewable power, and storage can all fit together. HOMER has been used by over 43,000 energy professionals in 193 countries. Although many of its applications have been very small systems, it is now being applied more often to multi-MW island systems.