Plentiful, Affordable, Reliable, but especially Reliable

Reliability is one of those things we don’t notice until it’s not there.

There was an interesting article published last week in The New York Times by James Glanz entitled, “The Cloud Factories: Power, Pollution and the Internet.” While the story line revolves around the vast amount of energy that modern data centers consume and the pollution associated with it, the story also highlights an issue most folks outside the industry do not  usually think about.

We turn on a light, switch on the stove burner, or jump on the internet and expect it to work. Service is there, 24/7, without fail — usually.

Our energy delivery systems, especially the electric grid, are designed and operated to be highly reliable. The electric grid is designed and operated to ensure system integrity. Ancillary services spread throughout the generation, transmission, and distribution systems keep it up and running nearly without fail. But, when it does fail, it’s a newsworthy event.

One of the tools used to provide this flawless delivery is reserve capability. The power grid is built with excess capacity over and above what the operators expect the maximum demand to be. (called “peak demand”) This backup capacity is used when a primary power source fails, or if demand exceeds the planners estimates of peak demand. It’s an effective “belt and suspenders” approach.

In the U.S., the electric transmission grid and wholesale sale of electricity in interstate commerce (sales that cross state borders) is regulated the Federal Energy Regulatory Commission (“FERC”). FERC, in turn, certified the North American Electric Reliability Corporation (“NERC”) to ensure the electric network’s reliability by developing and enforcing reliability standards and assessing the grid’s adequacy.

NERC divides the U.S power grid into 14 regions and makes reliability assessments three times each year. With the exception of Texas, every region has a positive reserve margin. That is, there is excess capacity in 13 of the 14 regions. Due to its operating and commercial structure, the Electric Reliability Council of Texas, (“ERCOT”) faces some unique challenges. As a result, ERCOT has experienced rolling brownouts and blackouts over parts of Texas during recent peak summer and winter periods.

The point of all this, however, is that the data centers examined in the NY Times article are not as unique as the author seemingly implies. Maintaining a high degree of reliability demands that redundant systems that sit idle a good deal of time be in place. Whether its Google, your bank, a hospital, or the power grid, standby capacity is essential if the system is going to operate without a hitch.

As our legislators and regulators develop new energy policies and make choices about how best to generate, transmit, and deliver plentiful and affordable electricity, reliability issues will be front and center. Studies by both industry and academic institutions have shown that if intermittent renewables (particularly wind and solar) become a significant portion of our electricity generation, it will increase the need for backup generation (especially natural gas) to provide the reserve margin.

Of course, reserve capacity that sits idle most of the time is expensive. The trade off between spending more money on reserve capacity and risking reliability is a judgement call. How much is enough? Just as important is the question of how to create incentives for investors to build such expensive, seldom used, but essential facilities. Certainly, there may be room for efficiency offsets, but as the NY Times article points out, infrastructure is needed to support these operations.

Whether it’s using the cloud for computing or smart grid technology to deliver our electricity, large scale infrastructure remains a necessity.  Just as with reliability, most of us will never notice it except when it’s not there.

Tags: ERCOT, FERC, internet, NERC, reliability, renewables, reserve capacity