On the other hand I hear from the utilities that wind is not available at peak, that Texas only plans for 10% of the installed wind capacity to be available at peak. That the wind ramp rates force plants to be tripped to avoid having too much power in the grid– creating more CO2 than should be created for the power delivered. That wind messes up the efficient running of combined cycle gas plants and other existing infrastructure. This again leads to too much CO2 being created for the power delivered from the fuel use. Then they complain that wind forces more spinning reserve and other ancillary services. The utilities are right on all points, but it is time to cool it.
In all cases people are looking at their technology as the savior industry and the response from the other side is that they are really a demon, not an angel. It does not matter which technology the polarisation of the industry is strong. This is not good for anything but theatre and we as an industry need to get past the theatre, time is too short to continue debating the merits of either position.
Instead it is time to think from a systems stand point. There are a set of goals the stake holders have and some people see them as conflicting. The goals are:
- Utilities - Efficient use of assets, maximum power production for each unit of fuel burned, reduction of peak load and flattening of the load curve. Delivering the maximum number of kilowatt hours per year on the existing infrastructure. To handle peak demand as cheaply and efficiently as possible, and not lose money (for the investor owned utilities it is to make money);
- Environmentalists – minimum carbon production and minimum disturbance of the overall environment. Maximum use of existing assets so no new ones need to get built, decommissioning of as much fossil generation as possible as fast as possible;
- Regulators – keeping power prices within reason, keeping the reliability of the system as high as possible;
- Customers – reliable, plentiful and cheap power. This is not only a homeowner issue, but a business issue as well, energy prices have forced businesses to close or relocate over time and will continue to do so in the future. Energy arbitrage will be second only to labor arbitrage on where businesses will locate in the future;
- Wind Industry – to build as much wind infrastructure as possible as fast as possible in the best possible places to replace as much fossil generation as possible. To get paid a reasonable price for the power delivered.
The argument is that these goals conflict and that the industry is better off in one configuration or another depending on your point of view. Utilities and some customers argue that wind power is too expensive. The wind industry and environmentalists argue that the current fossil generation does not pay for the right to pollute and if they did wind would be cheaper than other sources of power. To both it is time to cool it.
Tired of being told to cool it? Starting to get angry with this article?
Well good, now you are ready to think. This is not about a technology, it is not about one segment or another it is about a systems approach. The electric grid is a system. We have tried to chop it up into pieces with regulations, but the electricity still flows from source to load. Every time you try to deal with only one aspect of a complex system you fail. Don’t believe me, read the various project management blogs on the web about the failures of large projects. In most cases one of the root causes was breaking up a complex problem into pieces and not making sure the pieces would fit together when the work was done.
Wind has its strengths – it makes power when the wind blows and it can make a lot of power when the wind speed is right. Most days wind makes power – in Europe there are only six days a year on average when wind somewhere in Europe would not make wind power in an average year. There is no fuel to transport, there are no large operations staffs, and largely once wind is installed, there are few complaints about it. Wind can even in most locations be modelled, we can know the hours on average of maximum production.
The problem with wind is in most locations its most productive time is not the peak time in the market. In a pure generation world this is a problem. In a systems approach this is an opportunity! It is time to cool it.
If we look at the average data center today they spend more than 50% of the power consumed on cooling. This cooling is required 24 hours a day 365 days a year. Austin Energy was nice enough to put all their commercial customers online so you can see the average demand for each data center in Austin – the 100 WMh number squares with Dell and the other data centers Austin Energy Serves. So we will use 100 MWh a day for the energy consumption for a data center for this example. The American Wind Energy Association (AWEA) indicates that wind generates power approximately 38% of the time. So a 3 MW wind machine would produce approximately 25 MWh a day. If the average data center uses 50 WMh of energy a day to cool the data center than two 3 MW wind machines would provide the cooling for a whole day. To be safe if there were three allocated then the data center could ride out the occasional windless day.
Storage
Largely electricity cannot be stored. Anyone in the industry knows that. batteries convert electrical energy to chemical energy and back again. Only capacitors can store electricity directly and their power density is very low. So storing wind energy does not make sense directly at the data center. But the data center does not need electricity for cooling it needs cold and cold can be stored. In a systems engineering project this is the key. Cold can be stored in many forms and data centers need cold. Wind can create electricity and electricity can create cold.
Taking 50% of the load of a data center off peak into the times when wind is efficient and the grid has lower loading means that the wind energy is used efficiently and the power delivered will be used as if it were peak power – making it more valuable to everyone. Wind producers can justify higher prices for generation off peak. Utilities can better use their existing assets. Ancillary services can drop, since the cold storage can be the way to bleed off excess power. Regulators can have more flexibility to make rules that include wind and keep overall power prices lower. Peak is reduced, increasing overall reliability. In other words everyone wins by cooling it.
As the data centers are done, commercial district cooling systems can follow suit. Finally this can filter down to malls and commercial buildings, factories and finally homes in some climate zones. In other climate zones heat can be stored in homes and smaller buildings for distribution during the peak hours.
The technology exists. It is up to the all of us to make systems engineering part of way we solve these problems.
| About the Author |
| Doug Houseman is CTO for Capgemini's global utility practice. Capgemini is based in France and develops business strategies and technologies for a range of industries. |