This is not a trivial problem: Source: Hawaii's solar power flare-up: Too much of a good thing? - latimes.com When the power grid has a large number of sources and loads, a form of stable operation is possible BUT there had been at least two, well known power outages, in the Northeast and one in Quebec when sudden load changes caused protection systems at grid interconnects to shut everything down reaching back to the power plants. Bringing the grid back took days with commitments to 'fix it.' . . . not enough although we seldom hear about successful grid stability. Certainly Sandy would qualify as a serious test. Looking at current wind and solar practices, the missing part is the need for significant power storage but installed distributed units within the grid. As significant solar sources and wind farms come online, power storage local to these interconnects are needed to provide a fail-soft mechanism and conditioning. Data links to the base power operators then allows them to respond to variations. But response to load variations is not a trivial problem. It is as if we would want to have a critical load circuit separate from the quality-of-life load. I see this in a small scale with every power outage as my 1 kW has to be wired to critical loads and everything else powered off. Critical loads: ~50-100 W - switched specific room lights: family room, hallway, bathroom ~50-100 W - communications (aka., TV, radio, cell phone, laptop) ~100 W - room-size fan Switchable critical loads: ~300-500 W - food refrigeration ~1000 W - single plate or small microwave ~800-1000 W - room-size heating and cooling How we get from here to there is not clear. But I suspect the best approach is to head towards co-generation and emergency power systems approach. Rather this is how I see the problem because if nothing else, I can do something in this area. Bob Wilson
Hawaii has some unique problems that we don't have in the continental US. Your the OP article If they had abundant natural gas and a smart grid, they could buy a fast cycling combined cycle plant, which would cycle some turbines much easier than their existing oil generation infrastructure. They likely should implement a smart grid, and invest in an energy storage system. Duke energy is building a 36MW system by a west texas wind farm with half federal money ($44M project, $22 federal grant), Bob, I'm on a smart grid, and allow the utility to turn down or off my AC - which is the critical load on the local grid. They have never done it, or if they have I wasn't home. I can ask. But this is a big grid, with lots of dispatchable peaking gas turbines, and the ability in summer 2011 to plan and buy power from out of country - they could have bought it from out of state, but mexican electricity was less expensive Microwaves and refrigerators don't need to be controlled by the smart grid, but they are encouraging AC and car charging to be.
Major wind projects in the Pacific Northwest are also contributing to this non-trivial issue. A couple generations ago, our big hydropower machines could be spun up, excited, synchronized, locked in, and loaded in under 60 seconds. But a wide palette of environmental (fish and fish egg nest survival, water temperature, oxygen content, bank erosion, etc.), navigation, and water user safety issues put an end to that. Transmission capacity between the wind/hydro facilities and the major load centers is also a factor, and hasn't kept up. One factor in the first really big event was that many of the switches that could have been tripped to shed load in New York City had been bolted down. Load shedding to protect the grid was a political hot potato -- "cut off someone else's power, not mine" -- and it was believed the grid was large and stable enough that NYC would never have to be cut. It wasn't. So crews had to go around to unbolt and open all those switches before the city could start coming back up.
