Efficient Design of Electricity Market Clearing Mechanisms with Increasing Levels of Renewable Generation and Carbon Price
Increased use of wind energy in electricity systems can help reduce green house gas emissions and enhance energy security. However, the traditional scheduling and dispatching processes used to ensure the cost-effective and reliable supply of electricity in wholesale energy and ancillary service markets are not designed to deal with wind production uncertainty and variability. The growing variability and uncertainty of wind resources misinforms the scheduling and dispatching processes and ultimately causes economic and environmental inefficiencies. Various approaches have been proposed to integrate the wind uncertainty and variability into the electricity market clearing processes and enhance their economic and environmental efficiency. This dissertation develops a framework that enables quantifying the inefficiencies caused by the wind uncertainty and assessing the economic and environmental efficiency that could be gained by integrating the uncertainty into the market clearing design.
To assess the potential inefficiencies posed by wind uncertainty, three objectives are addressed. (1) Elucidate the incentives that wind uncertainty might create for electricity markets’ demand-side participants to develop market manipulation strategies and determine the factors that might contribute to or mitigate such market power. (2) Estimate the economic and environmental costs of wind uncertainty and the improvements that could be achieved by various approaches for integrating the wind uncertainty into the market clearing design. (3) Investigate how CO2 pricing policies that affect the priority order of generators in the supply curve and the grid’s overall flexibility impact the uncertainty costs and the improvements that could be achieved by integrating the uncertainty into the market clearing design.
First, in order to highlight the opportunities that wind uncertainty creates for the demand-side strategic behavior, this paper explores the effects of allowing large, price-responsive consumers to provide reserves in a power system with significant penetration of wind energy when the market is cleared using stochastic market clearing (SMC). The problem is formulated as a bilevel optimization problem representing a Stackelberg game between the large consumer and the other market participants. The study highlights how a large price-responsive consumer takes advantage of the wind uncertainty and leverages its ramp reserve deployment capability to understate its demand in the day-ahead market (DAM) and reduce the overall day-ahead (DA) and real-time (RT) prices to minimize the total daily cost of purchasing electricity in the DA and RT markets. The study also reveals how wind uncertainty, reserve deployment capacity, and transmission congestion contribute to the market power of large consumers that should be limited to mitigate their market power.
Next, to estimate the economic and environmental inefficiencies of the wind uncertainty, a framework is developed that replicates the operation of wholesale energy market clearing under the traditional design and adjusted designs that indirectly or directly integrate the uncertainty into the market clearing mechanisms. The indirect integration, referred to as augmented deterministic design, maintains the deterministic nature of market clearing mechanisms, i.e., DA unit commitment (DAUC) and economic dispatch (DAED), and deals with the uncertainty through scheduling ramp capability requirements, which are quantified exogenously to the market clearing processes based on the wind uncertainty characterization. The direct integration requires transition to the stochastic market clearing design in which stochastic optimization models are used for direct integration of the wind uncertainty characterization in the DAUC and DAED processes. The stochastic design allows endogenous quantification of the ramp capability requirements and optimizes energy and ramp capability reserve schedules by accounting for the expected cost of recourse actions taken to reconcile the RT balance mismatch caused by the deviation of wind energy producers from their DA production schedules.
The proposed framework resolves the differences of adjusted market clearing designs in terms of pricing, settlement, and reliability management to ensure a fair comparison of their dispatch, economic, and environmental outcomes. The comparative analysis reveals that the augmented deterministic and the stochastic designs enhance the economic and environmental outcomes, yet the stochastic design is superior as it offers more efficient and flexible energy and reserve schedules that are well coordinated with the anticipation of RT wind energy realizations. As a result, the stochastic design’s schedules can be adjusted more conveniently and cost-effectively to reconcile the deviations leading to greater operation and startup fuel cost savings; lower cycling of slow producers, higher wind integration and finally lower air emissions. Furthermore, stochastic design offers more efficient prices that reflect the system’s operation costs and wind uncertainty more effectively, provide greater remuneration of operational flexibility by producers, and reduce the revenue sufficiency guarantee payments that collectively improve the social surplus to a higher extent with respect to the augmented deterministic design.
Lastly, the developed market simulation framework is extended to include another adjusted deterministic design, referred to as hybrid deterministic design that uses stochastic optimization for direct integration of the wind uncertainty characterization to the residual unit commitment (RUC) stage. Then the economic and environmental outcomes of alternative market clearing designs are simulated under two carbon-pricing scenarios to evaluate their sensitivity to the introduction of a carbon price that alters the merit order of generation technologies in the supply curve.
The results imply that the stochastic market clearing design is superior to all adjusted deterministic designs. With introduction of a CO2 price, augmented and hybrid deterministic designs lose their effectiveness due to the shift in merit order of producers. However, stochastic market clearing maintains its superior performance that increases its superiority with respect to adjusted deterministic designs.
Economic and Environmental Efficiency
Electricity Market Clearing Design
Power Systems Economics
Renewable Energy Integration
This work is licensed under a Creative Commons Attribution-Noncommercial-No Derivative Works 3.0 United States License.
Rights for Collection: Duke Dissertations