(On Screen): There's a famous quote from H. L. Mencken to the effect that for any problem there's a solution which is straightforward, easy to understand, and useless. [DWL!]
TM Lutas suggests that the way to keep the power system from collapsing is to make all the meters bidirectional, so that anyone can either consume electricity or generate it, being charged for consumption and credited for generation.
But what if smart meters existed? What if people could participate in the electrical markets as a matter of course, buying the way that the major users buy based on the current market value of their juice at that time. What if you could just plug in a generator and automatically have your power flow into the grid and get compensated for it at the spot market price. What if every electrical device had a smart switch that could selectively cut the wall current to the backup battery units and the overhead lights in the offices but keep the electricity running for the elevators to get out of the building. What if all these devices could talk to each other and sort out what needed doing when there was a service reduction or outright cut.
All of a sudden the entire grid system looks different, much more local, more complex, and vastly more resilient. The incentives for local power generation increase as it becomes easier to do and you not only get stable power but also have profit potential. Pulling cheap juice in the evening and feeding it back at peak demand becomes an economic opportunity. And when a neighborhood loses its external juice, local power sources can be prioritized so that the more important uses get priority and bad effects are minimized. Whether these local sources are a bank of batteries, fuel cells, microturbine generators, or standard gas or diesel models, with a universal communications method and pre-made rules created by their owners, you end up with a very important adjunct that increases the ability of people to make fine power adjustments.
Alas, it's true that the grid would look different, more local and more complex, but not true that it would be more resilient. Nor would a system like this obviate the need for the majority of our power to be created by large centralized plants.
There are non-trivial practical problems involved with his idea. For one thing, the meter isn't the only reason that consumers can't produce. Another problem is that the local power distribution system isn't designed to let power flow the other direction, or to actually control power distribution at the level of blocks and individual houses. The transformers aren't designed properly, and most of the wiring isn't heavy enough. To do what he's talking about you'd pretty much have to redesign the entire local power distribution system from scratch and reinstall it, with a truly immense capital cost. Of course, that's a minor detail. There are much worse problems with his idea, much less easily overcome. The biggest difficulty is that his solution doesn't actually solve the problem even if we could figure out a way to deploy it.
His idea is based on something of a misconception about the power grid. He's thinking of it as being not dissimilar to other kinds of product distribution systems (e.g. food, fuel). But there's a critical difference between electricity and any other product we distribute widely: there's no practical way to store electrical energy in adequate quantities.
I fear I must be pedantic for a moment, because I want to make sure that everyone understands the critical difference between energy and power. Energy is a physical quantity which permits work to be done. (Work is a term of art in mechanics and has a precise meaning, but it isn't really relevant here.) Energy is measured in joules.
Power refers to the rate at which energy is produced or consumed, and both of those are measured in watts. If one joule is being produced or consumed per second, then one watt of power is involved. (So a watt-second, i.e. one watt of power for a period of one second, is one joule, a watt-hour is 3600 joules, and a kilowatt-hour is 3.6 megajoules.)
Power and energy are not coupled. High energy does not imply high power, or vice versa. I can consume one megajoule in one second, at one million watts, or over a million seconds, at one watt, or over a trillion seconds at one microwatt. But it's the same amount of energy. (I can consume one joule at a megawatt, in a microsecond, or a microwatt in a million seconds.)
There are physical laws involved in electrical power systems which can't be ignored. The most important is this: in any system involving electric power, the amount of power being generated will exactly match the amount of power being consumed at any given instant. It will always be exactly the same; it's physically impossible for it to not be. [DWL!] If it isn't consumed any other way, excess powe