How distributed should our power utilities be?

The mixed blessings of Distributed Generation (DG)

There are two probable paths that can lead to favouring distributed generation (DG):

1. When DG is seen as an ideal in itself which we should aim at for technical and/or socio-economical reasons. The fact that many DG systems make use of renewable energy is seen as just one of the overall advantages.
2. Or when the reduction of CO2 emissions is the ultimate goal. Renewable energy and cogeneration systems that reduce CO2 emissions are typically small scale, which leads automatically to a more distributed generation.

That last argument is becoming far less convincing as renewable energy farms continue to grow in size and output. The largest renewable energy plants are reaching the size of fossil fuel power plants. Texas hosts a wind farm with a capacity of 735 MW. A PV plant of 11 MW is operating in Portugal, one of 40 MW is being built in Germany, and one of 300 MW is planned in New Mexico. So to what degree generation should be distributed or centralized is a choice we have to make, even for renewables. And both concepts seem to present mixed blessings.

DG to reduce losses

Whether DG reduces or increases energy losses depends primarily on the individual case. A systems approach is required to tackle this issue. For instance:

• Cogeneration units are only highly efficient if all the heat is consumed locally. But since houses and buildings are now more properly insulated to reduce energy losses, the demand for local heat is falling. That limits the market possibilities for small or domestic cogeneration systems with a low electricity to heat ratio. Larger cogeneration systems for district heating do also require a high heat consumption density to be efficient.

• If electricity from a DG system is used locally, it requires both fewer and shorter power transmission, resulting in reduced losses. But renewable energy systems have to level out their intermittency on the regional, national, or international level. For instance, solar plants in Southern Europe can counterbalance wind farms in northern Europe. But this again leads to long distance electricity transport and its related losses. For non-grid coupled renewable energy systems, energy storage is the solution to intermittency, but this also brings about energy losses.

• DG leads to losses in the economies of scale. A small PV system for instance will have larger invertor losses than a large solar plant.

So the answer to whether DG reduces energy losses is in fact a very complex question and nearly always a case specific situation.

DG to reduce transmission capacity

Another frequently heard argument is that DG reduces the need for transmission line capacity. This is only true if those DG systems are reducing peak demand. If not, the required transmission capacity doesn’t change. Take the example of a house that is electrically heated using a PV system. Peak demand on this heating system will almost always occur during cold winter evenings, when PV production is zero. This means that the grid connection of this house needs to be just as strong as if there was no PV system installed.

Creating microgrids could be a solution

'Extended microgrids' may be a good solution to limit both grid capacity requirements and grid losses, and to ensure reliability in case of a high penetration of DG. A microgrid is a subsystem that can be separated from the main grid (see recent blog post - creating microgrids). Such a microgrid would operate in parallel with the grid (when connected) or in island mode (when disconnected). It will disconnect from the grid during significant events (faults, voltage collapses, et cetera), providing UPS services to its loads. An extended microgrid (see recent blog post - extended microgrids) consists of a group of radial feeders, each of which include not only loads and a generation unit, but also a storage device.

The capacity issue: how far can we go?

When talking about the capacity of DG units, the difficulty lies in estimating the capacity credit. This is the degree to which one can depend on a generation unit to meet the load demand. The capacity credit of wind, solar, and mini-hydroelectric resources is significantly below 100 per cent.

The capacity credit increases when aggregating the output of renewables on a regional, national, or international level. However it reaches a certain asymptotical limit and adding more units becomes useless. For wind energy, this limit certainly lies well below 100 per cent. Since photovoltaic energy is less random, its asymptotical capacity limit will be higher than that of wind energy, but still far below 100 per cent. In case demand and supply correlate, as in the use of photovoltaic cells for cooling applications, this limit value is higher.

Because of this limited capacity credit of renewables, a large energy system based 100 per cent on renewable energy will only become feasible if large, efficient storage systems become available. And that is not something likely to happen in the near future.

Balancing the hidden subsidies

When balancing centralized versus distributed generation, the hidden subsidies for both systems also have to be taken into account. Fossil fuel and nuclear power plants receive hidden subsidies via their environmental externalities. Renewable energy systems on the other hand have an extra, hidden cost for society because their intermittency must be compensated for. And if a system of feed-in tariffs is in place, small generators are not charged for the extra ancillary services they consume.

Distributed market power versus centralized know-how

Choosing either distributed or centralized generation also has an impact on the way the market and competencies are organized. An advantage of DG is that it gives easier market access to small players, avoiding the creation of semi-monopolies by large utility companies. As a general rule, more market players results in a better functioning of the free market.

Centralized generation by large utility companies has the advantage that it enables the centralization of know-how. Those companies will have the ability to create competence centres that continuously improve the operation and maintenance of their power plants, which should result in more efficient and better performing installations.