Browsing by Subject "Grid modeling"
Now showing 1 - 1 of 1
- Results Per Page
- Sort Options
Item The benefits and challenges of renewables on the electric grid and opportunities for systems integration and demand side management(2018-03-02) Deetjen, Thomas A.; Webber, Michael E., 1971-; ; Baldick, Ross; Hebner, Robert E; Leibowicz, Benjamin DEnvironmental policies, reduced manufacturing costs, and technology improvements have all contributed to the growing installation of wind turbines and solar photovoltaic arrays in the electric grid. While these new sources of renewable electrical power provide environmental and economic benefits to the electric grid, they also complicate the balancing of supply and demand required to reliably operate the grid. The seasonal, daily, and sub-hourly fluctuations in the energy output of wind and solar generators must be compensated by operating the existing power plant fleet more flexibly or by providing more flexible sources of electricity demand. This dissertation categorizes and quantifies this compensation by studying the "flexibility requirements'' imposed by wind and solar generation, approximates the economically optimal capacities of regional wind and solar resources in the grid, and explores the ability of a central utility plant to add a flexible source of demand to the electric grid system. These topics are covered in the four chapters described below. Chapter 3 utilizes a unit commitment and dispatch (UC&D) model to simulate large solar generation assets with different geographic locations and orientations. The simulations show the sensitivity of the wholesale energy price, reserve market prices, total dispatch cost, fuel mix, emissions, and water use to changes in net load flexibility requirements. The results show that generating 22,500 GWh of solar energy in a 2011 simulation of the Electric Reliability Council of Texas (ERCOT) reduces total dispatch cost by approximately $900 Million (a 10.3% decrease) while increasing ancillary services costs by approximately $10 Million (a 3% increase). The results also show that solar PV reduces water consumption, water withdrawals, and CO₂, NO [subscript x], and SO [subscript x] emissions. Installing sufficient solar panel capacity to generate that much electricity also reduces peak load by 4% but increases net load volatility by 40--79% and ramping by 11--33%. In addition, west-located, west-oriented solar resources reduce total dispatch cost more than the other simulated solar scenarios. The west-located, west-oriented solar simulation required greater system flexibility, but utilized more low-cost generators and fewer high-cost generators for energy production than other simulated scenarios. These results suggest that the mix of energy provided by different generation technologies influences the dispatch cost more than the net load flexibility requirements. Chapter 4 develops a quantitative framework for calculating flexibility requirements and performs a statistical analysis of load, wind, and solar data from the Electric Reliability Council of Texas (ERCOT) to show how wind and solar capacity impacts these grid flexibility requirements. Growing wind capacity shows only minor correlation with increasing flexibility requirements, but shows some correlation with ramp down rates and daily volatility in the net load. Growing solar capacity shows a direct correlation with increasing flexibility requirements if load patterns do not change. While adding 15.7 GW of wind power had only minor effects on system flexibility requirements, adding 14.5 GW of solar to the ERCOT grid increases maximum 1-hr ramp rates by 135%, 3-hr ramp rates by 30%, ramp factors by 140%, 1-hr volatility by 100%, and 1-day volatility by 30%. Wind and solar impact flexibility requirements at different times of the day: wind tends to intensify demand-driven flexibility events by ramping up energy production at night when demand is decreasing and ramping down energy production in the morning when demand is increasing, while solar tends to intensify flexibility requirements due to its quick changes in energy output driven by the rising and setting sun. Adding wind to a system with large amounts of solar does not tend to increase flexibility requirements except for the daily volatility. The geographic location and orientation of solar arrays also influences flexibility requirements, with fixed, southeast-facing panels providing a significant reduction. These results can inform strategies for managing the grid flexibility requirements created by growing renewable capacity. Chapter 5 develops a model for calculating the optimal amount of transmission, wind, and solar capacity that should be built in a grid's different regions. It also presents a framework for choosing CO₂ prices by balancing increasing system cost and flexibility requirements with CO₂ emissions reductions. In a simulation of the ERCOT grid, the model suggests a 60 $/ton CO₂ price and an optimal investment of 27.0 GW of transmission capacity to five different regions. These regions install a total of 26.6 GW of wind and 11.1 GW of solar, representing a grid with about 60% thermal and 40% renewable capacity. This renewable mix produces 110 TWh of energy per year, 34% of the total electricity demand. The grid emits 82.2 million tons of CO₂ per year under this scenario, a 65% reduction from the 237 million tons produced when no renewable capacity is installed. At the optimal renewable development solution, all coal and natural gas boiler generators have capacity factors less than 20% with many of them not being dispatched at all. While these results are specific to ERCOT, the methods and model can be used by any grid considering renewable energy capacity expansion. Chapter 6 develops a mixed-integer linear program for modeling the optimal equipment capacity and dispatch of a central utility plant (CUP) in a residential neighborhood and its ability to improve rooftop solar integration. The CUP equipment includes a microturbine, battery, chiller plant, and cooling storage. The CUP model is exposed to a variety of electricity rate structures to see how they influence its operation. The model finds the optimal capacity for each piece of CUP equipment, optimizing their hourly dispatch to meet neighborhood cooling and electric demand while maximizing profit. In an Austin, TX, USA base case, the neighborhood benefits economically by including the CUP, although the CUP demonstrates limited potential to integrate high penetrations of rooftop solar resources. While peak demand and reverse power flows are reduced under all tested rate structures, the CUP worsens net demand ramp rates. A time-of-use rate with no demand charge and moderate differences between off-peak and on-peak prices balances the output parameters, reducing reverse power flows by 43%, peak demand by 51%, and annual cost by 9.1% versus the ``No CUP'' base case while limiting net demand ramp rate increase to 84% more than the base case. Building a clean, resilient, and reliable electric grid for the future is a worthwhile endeavor that will require innovative supply-side and demand-side solutions for integrating the intermittent power output of renewable generation into the electric grid. As a cohesive document, this dissertation communicates the scale and severity of the flexibility requirements that will be required to operate systems with large amounts of wind and solar generation and explores one demand-side method for providing that needed flexibility. There are many opportunities to expand these analyses and explore new sources of grid flexibility in future work.