Most supporters of renewable energy development are probably pretty comfortable with the way things are going. Wind and Solar generation has been increasing both in "nameplate capacity" and in actual production of electricity. There have not been any significant grid failures that can be blamed on renewables. Apart from a consolidation within the solar cell manufacturing sector there have not been any notable bankruptcies within the electricity generating sector. All visible signs are positive for a continued expansion of renewable resources.
When I talk to groups about renewable energy I start off with a Youtube video which demonstrates testing the compression strength of a concrete block. For 2 minutes and 40 seconds this is the most boring video you could imagine. The block shows absolutely no sign of stress. At 2:41 the concrete block fails and is utterly destroyed. As far as I am concerned we are at about 2 minutes and 30 seconds with respect to the electrical grid.
In order to understand what I believe to be the serious risks facing the electrical generation and distribution system it is necessary to review the structure of the system as it was before renewables began to be developed in a significant way. The chart below shows hypothetical load profiles for a peak demand day during the spring/fall, winter and summer as well as a line that represents the overall generating capacity in the system.
It can be observed that the system demand/load varies considerably throughout the day and throughout the year. It is also clear that there is a great deal of excess supply available for most hours on most days. In fact, only on the highest peak demand days of the entire year will the demand come close to the supply. That is by design as every well-managed electrical generation system in the world requires a reserve margin of 8-15% above peak demand.
This reserve is meant to provide resiliency for the grid to accommodate scheduled maintenance shut-downs at major facilities such as nuclear plants, natural gas-fired and coal-fired plants as well as unscheduled outages due to storms or switching problems or other operational issues.
(Note: I appreciate that many people will raise objections to the demand curves presented in that their local situation might be very different. That is one of the challenges facing every Independent System/grid Operator. Local demand curves can be all over the map due to the mix of commercial, residential, and industrial users. My point is not that these particular curves are the most typical in all locations. The point is that demand varies significantly over the course of the day and through different seasons.)
So before we began to develop renewable energy there was plenty of generation capacity within the system. In fact, many generation facilities were not running at anything close to capacity most of the time.
Because of a public policy decision to reduce the burning of hydro-carbons (and the associated production of CO2 emissions) wind and solar generation sources have been subsidized through a variety of financial instruments including capital grants, tax credits, and feed-in-tariffs. Renewables have also been given preferential access to the grid in most jurisdictions.
These measures have achieved the stated policy goal. Wind and solar now make up a significant percentage of generation capacity in a number of jurisdictions and at times provide a large percentage of electrical production.
For example, Germany has developed over 30 GW of solar power and over 30 GW of Wind. On a blustery spring day in Germany renewables can meet up to 40% of the total electrical demand for a few hours at mid-day. There are regular announcements of "new records" for both solar and wind generation. A similar situation exists in Texas with regards to wind and in parts of Hawaii with regards to solar.
Remembering that there was already a surplus of generation capacity in the system before the development of renewables it is obvious that when renewables hit their generation peaks most traditional thermal generation plants are unable to sell electricity. That would not be a problem if the construction of these plants had not been financed based upon assumptions regarding how often they would be used and what wholesale electricity prices would be. In fact, the economics of running these plants has deteriorated to the point where many utilities, especially in Europe, are on a "credit watch".
The rational response of companies trying to sell electricity into a market that has a great over-supply would be to decommission some of the oldest and most polluting plants to bring supply and demand into a better balance. But there is a problem. Renewable resources cannot be relied upon, particularly at peak demand times. The chart below displays the wind resource available compared to the demand curve for a week in November, 2013 in Texas (this week was not chosen on purpose to make wind look bad. It was literally the first file I found on the ERCOT site when I was starting to write this blog).
In this situation demand rose throughout the week as a strong high pressure system spread across the state bringing with it colder temperatures while at the same time shorter days required more lighting. One of the more troublesome realities of meteorology is that large, stable high pressure systems are often responsible for peak electrical demand in both winter and summer because they are associated with clear skies and temperature extremes. These systems are also commonly characterized by very low winds across a wide area.
As a result while demand continued to climb wind energy faded away to almost nothing. At this point most of the thermal generation assets available within Texas had to come on-line in order to meet demand.
So it is impossible to decommission even the oldest and least efficient thermal generation plants in the system regardless of how many wind farms have been built and solar panels deployed. German utility E.on came face-to-face with that reality in the spring of 2013 when they were instructed by the local grid operator to keep an old plant operational even though it would rarely be needed.
But a new day is dawning in the U.S. and it could be a darn cold (or hot) one.
The EPA announced regulations in December 2011 that will require coal-fired thermal generation plants to clean up or shut down. The reality is that for many of these plants it will not be feasible to clean them up. In fact, in some cases the EPA will not even allow them to be updated with modern pollution controls. As a result more than 40 GW of firm generation capacity will be decommissioned over the next several years.
Plans to replace this loss are in some cases vague and have been changing often. Increased conservation and better utilization of existing plants are frequently included in Integrated Resource Plans. In other cases greater reliance upon renewables is explicitly identified. These are not really replacements for firm capacity.
A number of new Natural Gas fired plants are also under construction. While current low gas prices make this an attractive option the threat of future significant price hikes as well as the EPA's stated goal to regulate CO2 emissions are worrisome and are impacting the ability to secure financing of these plants in some cases.
As more and more coal-fired plants are retired it is likely that total system firm generation capacity will drop resulting in smaller reserves. This, in turn, will make the system more susceptible to storms or other unplanned outages.
The degree to which grid security is compromised will vary from region to region depending upon the penetration of renewables, number of coal-fired plant retirements and the health of the local economy which has a major impact on electricity demand. Based upon those factors I believe Texas and the Mid-west are the areas most at risk.
It may be that the reduction in coal-fired generation will do nothing more than cull excess capacity out of the system with no negative impacts. But groups such as the Institution of Engineering and Technology in the UK have issued warnings about the progressive stress on a system that has taken decades to evolve and is now faced with unprecedented challenges.
Like the concrete block in the Youtube video the system is not displaying any outward signs of weakness. The question is this - will the North American electricity system encounter its own version of second 2:41?