« Back to Energy Insider

Renewables: High Penetration Renewables Can Work

by Bob Shively, Enerdynamics

The U.S. has in recent years experienced significant growth in renewable generation. Multiple states are now implementing electric resource plans that incorporate high levels of renewable generation. These include Hawaii at 40%, California at 33%, Colorado and Maine at 30%, as well as numerous other states at 20% or higher.

Such growth is not without apprehension. There is concern that many of these percentages have been set through a political process rather than through traditional engineering-based resource planning. Is it even possible for operators of electric grids to implement such penetrations of renewables while still delivering the reliability we have come to expect? And if it is, will the economic impacts leave consumers mourning for the days of the traditional electric system?

Modern power systems were designed for steady generation provided by a limited number of centralized, dispatchable power plants. A high penetration of renewables, provided largely by wind and solar power, presents a new paradigm – a significant amount of generation that is variable due to the intermittency of wind and sunlight. Such generation is non-dispatchable since wind and sunlight can’t be controlled, and difficult to predict since output depends on weather conditions. For the planners and operators who must ensure reliability, such variability is cause for concern. Key areas affected include long-term capacity planning, seasonal reliability planning, day-ahead and hour-ahead scheduling, and real-time operations. This article will focus on scheduling and real-time operations, the functions handled by the system operator.

From the system operator's perspective, some renewables such as geothermal and biomass operate similarly to a fossil fuel unit in that units are dispatchable. This means that output can be controlled by adjusting the input of fuel, since the availability of fuel is not variable. Conversely, wind and solar technologies are generally non-dispatchable, variable, and their unit output can be difficult to predict accurately. Examples of wind speed and solar radiation hourly variabilities are shown below:

While power systems are designed to handle significant levels of variability in loads, they are not designed to handle significant variability in generation. The impacts of variability combined with lack of dispatchability and predictability can lead to numerous difficulties for system operators. These include:

• System generation may deviate significantly from forecast amounts, thus requiring the
  use of real-time non-renewable generation to ramp-up when renewable output is low
  and ramp-down when renewable output is high
  
  
• Renewables output can change suddenly, requiring quick ramp-up or ramp-down of
  regulating units to maintain system frequency
  
  
• Extra use of system reserves may be required to cover for unexpected drops in
  renewable output
  
  
• There is the potential for overgeneration, where the system has ramped-down all
  flexible non-renewable generation and still has more generation than load

Despite these concerns, studies by various organizations including the North American Electric Reliability Corp (NERC), the National Renewable Energy Laboratory (NREL), and different Independent System Operators indicate that with changes to procedures and technology, high levels of penetration are feasible. Key tools include:

• Geographical diversity, which allows drops in supply in one area to be balanced with
  increases in other areas
  
  
• Transmission additions and reinforcements, which allow renewable power to be moved
  long distances
  
  
• Enhanced use of existing generation flexibility to ramp up or down as renewable output
  fluctuates
  
  
• Enhanced flexibility in renewable generation which allows operators to ramp down
  renewable output
  
  
• Additions of new flexible resources which allow operators to balance out renewable
  fluctuations through use of storage or demand response
  
  
• Enhanced measurement and forecasting of renewables which improves the scheduling
  process
  
  
• Upgrades of distribution circuits to allow outputs of distributed generation to flow in
  multiple directions

Implementation of these tools requires close cooperation among system operators, generators, transmission owners, distribution companies, and market participants. It also requires changes in rules and procedures. Examples of changes currently being implemented or considered in various regions include cooperative scheduling procedures across multiple control areas; new processes for approval of transmission construction in support of renewables; rewards to owners of non-renewable flexible generation for making their units available to system operators for ramp-up or ramp-down; requirements or incentives for renewable units to provide ramp-down capabilities; installation of storage technologies; enhancement of economic demand response resources; and enhancement of forecasting techniques. Development of such tools combined with market-based solutions promises to be a growing area for years to come.

This brings us to the million dollar question: can this be done at a reasonable cost? Most studies say yes. A recent U.S. Department of Energy study of 20% wind penetration indicated that integration costs are expected to be less than 10% of the wholesale cost of energy, with various cited studies showing integration costs ranging from $1.85 to $4.97/MWh3. While many would find this an acceptable cost, we won’t really know what the costs are until we get well into the process. But it appears likely that within the decade operators will handle generation variability as a routine occurrence, much like they do loads today.