Subscribe
Share
Share
Embed
Mays, J. (2023). Sequential Pricing of Electricity. Working Paper, School of Civil and Environmental Engineering, Cornell University.
“The goal of this paper is to promote a shift in the discussion of price formation in wholesale electricity markets from a static to a dynamic modeling framework. While the design and analysis of systems with significant reservoir hydropower have long relied on dynamic models, most other systems have come to rely on simpler static models that have nevertheless been useful in contexts with limited variability, uncertainty, and intertemporal constraints. The entry of large quantities of renewable and battery storage has increased the salience of all these factors, necessitating a richer modeling framework.” p. 40“This paper argues for the centrality of real-time markets, which are cleared sequentially with a single binding interval.” p. 3.
“This paper develops a framework to assess the effect of such proposals on market outcomes, investigating how choices made by wholesale market operators regarding algorithms for commitment, dispatch, and market clearing can affect incentives for operation and investment.” p. 2“Step 1: specifying the model for operations” Section 3.2.1, p. 15“Step 2: specifying a parameterization” Section 3.2.2, p. 15“Step 3: specifying a pricing policy” Section 3.2.3, pp. 15-16
“The paper’s conceptual goal is a shift from the static picture of the merit-order curve in thermal-dominant markets to a dynamic understanding of price formation. Electricity prices are often colloquially described as the cost to serve an additional unit of load for a given period. In a static, convex economic dispatch model, prices that maximize efficiency both in short-run operations and long-run investment can be calculated as the dual variables corresponding to power balance constraints equating supply and demand. With no intertemporal operating constraints, dual values are typically determined by the fuel cost of thermal resources. In a dynamic model, an additional unit of load in a given period not only entails a direct cost in the present period, but also places the system in a slightly different state entering the subsequent operating period (e.g., with more or less energy stored in batteries). Dynamic models have been long been understood as necessary to the design and analysis of markets in regions with significant reservoir hydropower (see, e.g., Pereira and Pinto (1991) as well as more recent reviews in Steeger et al. (2014) and Aasg ̊ard et al. (2019)). In other regions, including the U.S. systems that serve as the primary motivation of this paper, markets have evolved in a context where storage was negligible. This state of affairs is set to change rapidly over the coming decade, as models typically find that decarbonized electricity systems will feature substantial quantities of storage (Jenkins et al., 2018; Williams et al., 2021; Frazier et al., 2021).” p. 2
“The broader point of the examples is to establish the need for the proposed framework in evaluating price formation: even with the limited scope of tests included here, the models produce a wide range of price outcomes.” p. 37
“We note several differences between the (DLAC − NLB − θ) and (DLAC − RT − θ) policies. First, while we assumed in Section 4.3 that net load biasing would not result in any binding forward financial positions, the (DLAC −RT −θ) policy directly affects the quantity of reserves procured in the binding interval.” p. 30
“The choice of operating policy can have a significant impact on the total revenue across all resources. [...] In general, resource-specific revenues track the total charges.” however “storage is a notable exception with revenues deviating significantly from that seen by the market as a whole.”“They indicate the potential for substantial misallocation of investment due to inefficiencies in spot price formation.”
“Accordingly, a second policy recommendation is to advocate for high-fidelity simulation tools enabling system operators to compare operational performance and pricing outcomes with alternative algorithmic choices and reserve product specifications, putting them in position to credibly demonstrate the value of new reserve products and parameterize them efficiently.” p. 5.
Public Power Underground is more than a discussion about public ownership of electric infrastructure, the infotaining episodes cover the energy enthusiast trifecta of electrification, markets, and people. The hosts interview industry experts on a broad range of energy industry and energy-industry-adjacent topics at the nexus of electric utilities and the energy transition. The podcast doesn’t take itself too seriously and frequently plays energy inspired games like “energy enthusiasm distilled,” “draw an analogy,” and “say something nice about electric utilities."