This is a class paper I wrote this week and thought it might be of interest to readers here. I can provide more information if desired. The point to the paper was to write concisely for a policy audience about a decision support planning method in a subject that interests me. Note that this is only from one journal paper among many that I read every week between class and research. I will let readers know how I did after I get feedback. As always, comments are welcome.
40% of the United States’ total carbon dioxide emissions come from electricity generation. The electric power sector portfolio can shift toward generation technologies that emit less, but their variability poses integration challenges. Variable renewables can displace carbon-based generation and reduce associated carbon emissions. Two Stanford University researchers demonstrated this by developing a generator portfolio planning method to assess California variable renewable energy penetration and carbon emissions (Hart and Jacobson 2011). Other organizations should adopt this approach to determine renewable deployment feasibility in different markets.
The researchers utilized historical and modeled meteorological and load data from 2005 in Monte Carlo system simulations to determine the least-cost generating mix, required reserve capacity, and hourly system-wide carbon emissions. 2050 projected cost functions and load data comprised a future scenario, which assumed a $100 per ton of CO2 carbon cost. They integrated the simulations with a deterministic renewable portfolio planning optimization module in least-cost and least-carbon (produced by minimizing the estimated annual carbon emissions) cases. In simulations, carbon-free generation met 2005 (99.8 ± 0.2%) and 2050 (95.9 ± 0.4%) demand loads in their respective low-carbon portfolios.
System inputs for the 2005 portfolio included hourly forecasted and actual load data, wind speed data generated by the Weather Research and Forecasting model, National Climatic Data Center solar irradiance data, estimated solar thermal generation, hourly calculated state-wide aggregated solar photovoltaic values, hourly temperature and geothermal data, and approximated daily hydroelectric generation and imported generation. They authors calculated 2050 load data using an assumed annual growth rate of 1.12% in peak demand and 0.82% growth in annual generation.
The Monte Carlo simulations addressed the uncertainty estimation of different system states. As an example, the authors presented renewables’ percent generation share and capacity factor standard deviations across all Monte Carlo representations. The portfolio mix (e.g., solar, wind, natural gas, geothermal, and hydroelectric), installed capacities & capacity factors of renewable and conventional energy sources, annual CO2 emissions, expected levelized cost of generation, and electric load constituted this method’s outputs.
A range of results for different goals (i.e., low-cost vs. low-carbon), the capability to run sensitivity studies, and identification of system vulnerabilities comprise this method’s advantages. Conversely, this method’s cons include low model transparency, subjective definition and threshold of risk, and a requirement for modeling and interpretation expertise.
This method demonstrates that renewable technologies can significantly displace carbon-based generation and reduce associated carbon emissions in large-scale energy grids. This capability faces financial, technological, and political impediments however. Absent effective pricing mechanisms, carbon-based generation will remain cheaper than low-carbon sources. The $100 per ton of CO2 assumption made in the study’s 2050 portfolio is important, considering California’s current carbon market limits, its initial credit auction price of $10.09 per metric tonne (Carroll 2012), and its a $50/ton price ceiling. In order to meet the projected 2050 load with renewable sources while reducing emissions, technological innovation deserves prioritization. More efficient and reliable renewable generators will deliver faster investment returns and replace more carbon-based generators. Improved interaction with all stakeholders during the planning phase of this endeavor will likely reduce political opposition.
Carroll, Rory. 2012. “California Carbon Market Launches, Permits Priced Below Expectations.” Reuters, November 19. http://www.reuters.com/article/2012/11/19/us-california-carbonmarket-idUSBRE8AI13X20121119.
Hart, E. K., and M. Z. Jacobson. 2011. “A Monte Carlo Approach to Generator Portfolio Planning and Carbon Emissions Assessments of Systems with Large Penetrations of Variable Renewables.” Renewable Energy 36 (8): 2278–2286.