The impact of renewable energy on the grid


By Stephen Asbury and Doug Houseman

Renewable energy is an important part of the future for the South African energy industry. With political requirements to manage the cost of energy, reduce carbon, deal with global warming, be greener and allow local participation in the power grid, South African regulators have had no choice but to change the requirements for what is allowed on the grid.

Renewable generation is the buzz word today. The Department of Minerals and Energy in South Africa has made a decision that 10 000 GWh of electricity should be generated from renewable energy by 2013, although the policy governing how this should be done had not yet been finalised. This has accelerated the need for electricity generators to explore and implement the use of renewable energy sources. In the US, more than 30 states now have renewable portfolio standards (RPS))—mandatory percentages of power that have to be produced and delivered from renewable sources by specific dates.

In the US, the state requirements will require that an additional 1% of the total electric power consumed in the US be produced by renewable sources each year. In Europe, triple 20 will force a similar requirement.

This installation of renewables will not be without problems. None are impossible to deal with, but some are very expensive to deal with and others just take time and energy. Renewables will have an impact on the whole utility value chain and how it operates. No utility is exempt, and it does not matter if the market is fully regulated or fully de-regulated, there will be an impact. In some cases it is an engineering issue, in others it is a people issue.

In South Africa in order to ease the cost pressures of generating electricity from renewable energy sources the National Energy Regulator of South Africa (NERSA) has announced that it will implement a Renewable energy feed-in tariff (Refit) which it explicitly detailed on 28 February 2009. The Refit would aim to stimulate investment into this sector and although it would not lower the cost of electricity for the customer (as this was averaged out), it would go some way to subsidise renewable energy generators. Initial indications are that the proposed tariff schedule would include a contribution of between 50c/kWh and 80c/kWh for all renewable energy generated, and this could be from wind, small hydro, landfill gas, or concentrating solar power.

To really look at the impact of renewable sources on the electric grid, you need to separate the renewables into several categories.

We have created the following categories:

  1. Schedulable central station generation (S-Cent) – This includes biomass and other alternative fuels that can be used in place of fossil fuels. For instance, wood chips replacing coal, ethanol replacing oil, or biogas replacing natural gas.
  2. Variable Central Station 2. Generation (V-Cent) – Wind farms are the best known examples of this class of asset and are the most widely deployed renewable generation sources today producing electricity. Large solar and wave power installations also fit into this category. The assumption is that V-Cent would be connected to the transmission (high voltage) network.
  3. Schedulable Distributed Generation (S-Dist) – This is small scale generation that uses renewable fuels; normally, the generation facilities would be found on farms, in businesses and in homes. One good example would be wood-fired combined heat and power. The expectation would be that the generation can be turned on and off as needed.
  4. Variable Distributed Generation (V-Dist) – This is the category that most environmentalists mean when they discuss the next generation of the electric network. Solar Cells on a homeowner’s roof, or a small windmill in the backyard.

These four categories of renewables bring different issues to the grid and have different level of grid friendliness. The location of the connection and the ability to produce power on demand will determine how friendly they are.

S-Cent is so similar to conventional generation that it offers few or no issues in integration into the grid. Since it is so simple to integrate, where the fuels are available and economically viable, it should be considered as the first choice of deployment. Although there are very few challenges from a grid management point of view, they do offer other challenges for site approval and operations. In some areas of the US incinerators that were installed only a few years ago are being shut down because of protests by people who live near the facility.

Like the S-Cent facilities they are connected to the high voltage networks and produce large amounts of power. Also like the S-Cent facilities they are highly visible on the horizon and cannot be hidden from view easily. Unlike the S-Cent facilities they run when the environment is right and not when they are needed. For example, wind has to blow fast enough but not too fast in order to generate power. These systems require some level of other generation to provide support, since the power they produce comes and goes depending on the weather.

What the existing systems cannot do is to predict with any level of confidence what the output of wind power will be 24 to 48 hours in advance. This less-than-perfect understanding of the weather means that the utility has to be ready to react to changes in power output on a very short time line.

In most cases the locations that have the best wind are not the places that have high densities of people or power consumption. For the grid operator, that means building extensions to the transmission grid and high voltage transmission lines are not cheap.

No one likes large steel towers in their backyard, so you can expect that even if people embrace wind mills, they will pan the transmission corridors. With these new large transmission links, there are issues with power quality and voltage management on the transmission network that did not exist before.

Schedulable distributed generation differs from V-Dist in that it is done on a more ‘human scale” typically in less than 1 megawatt size. Changing the grid to support S-Dist means changing the relaying and the protective devices to allow power to flow backwards on the grid. The grid is designed to allow power to flow one way, from the high voltage network to the customer; but power has to be able to flow in both directions.

As S-Dist is integrated you would need to create a measurement system and change the tariff to support the payment of customers who generate power for you.

S-Dist has a further problem. Because of the small size of the generator they generate a single phase of power. This can lead to very large phase imbalances and big differences in voltage on a phase to phase basis. Phase imbalance normally leads to wasted power. Further, most of the generators use electronic inverters that create harmonics that can be harmful to the operation of the grid as well as shortening the life of appliances and computers that are installed on the grid.

Because the S-Dist is installed in the distribution network the transmission SCADA system does not offer the ability to operate or monitor these generation sources.

Loss of power from a system failure would cause most S-Dist system to automatically disconnect from the grid since they are seldom large enough to support the entire load in the remaining section of the grid.

This means that customers that own S-Dist would have a higher level of reliability than the rest of the customers. Finally these generation sources have site approval problems, they are often noisy and smelly.

Finally there is V-Dist, the kind of renewables that most people think of when you mention renewable generation—human-scale renewables that should blend into the neighborhood. These solar and wind powered devices are owned by business owners and homeowners. They are deployed in the low voltage network and suffer from all the same problems that S-Dist does. This includes the issues with harmonics and with phase imbalance. They have even had issues with smell and noise. In addition, these devices pose another problem for the grid operator—they only run when the environment is right. They use the grid like a large battery, putting excess power into the grid and drawing power from the grid when they do not make enough. In similar cases, going from a net generator to a net consumer can take seconds. This works well when the percentage of power produced from V-Dist is small, but as the percentage increases there will need to be real batteries installed on the grid, and that is a whole different story with its own issues.

Most of the best places to put wind mills and solar cells are out in the suburban and rural areas where the people density is less, and that means the power consumption is also less. In many cases, this means the size of the wires in the distribution grid are also smaller. To put larger numbers of V-Dist units in the rural areas and move the power to people who will use it may involve re-conductoring—replacing the physical wire in the distribution grid so that it will carry more power than it does today. For an overhead system, where the lines are up on poles, reconductoring can cost as little as R1 million (US$100,000) a mile. For underground systems, that number can run into more than R100 million a mile.

Unlike the V-Cent, these systems are maintained by the homeowner and many of them do not get the power output they were promised, do not make the money they expected and hence do not maintain the generator. This leads to a collapse in the capability of the system to make power. The ability to monitor the health of the power electronics is critical to the safety of the people working on the system and simple fine for failure to maintain the system is not enough to remove the human hazard that exists.

The good news is that nothing new needs to be invented — only improved and deployed (with the exception of storage). Renewables are a key part of fixing the carbon issue and providing electricity where the cost of power is not tied to the cost of fossil fuels. Fuel price swings and the move to carbon markets will tend to accelerate the move to renewables. Regulators, customers, utilities, manufacturers and independent power producers will all have to work together to make this work.