Abstract:We consider the process of bidding by electricity suppliers in a dayahead market context where each supplier bids a linear non-decreasing function of her generating capacity with the goal of maximizing her individual profit given other competing suppliers' bids. Based on the submitted bids, the market operator schedules suppliers to meet demand during each hour and determines hourly market clearing prices. Eventually, this game-theoretic process reaches a Nash equilibrium when no supplier is motivated to modif… Show more
“…Similarly, a bidder may seek to impute several unknown parameters which can be used in the process of deciding her bid. These parameters include cost coefficients of rivals' models (Chen et al, 2019); rival bids, which are objective function coefficients in the facilitator's problem (Ruiz et al, 2013); and parameters that describe the routing of energy through the network and the capacity of transmission lines, which are left-hand-side constraint parameters in the facilitator's problem (Birge et al, 2017).…”
Most inverse optimization models impute unspecified parameters of an objective function to make an observed solution optimal for a given optimization problem with a fixed feasible set. We propose two approaches to impute unspecified left-hand-side constraint coefficients in addition to a cost vector for a given linear optimization problem. The first approach identifies parameters minimizing the duality gap, while the second minimally perturbs prior estimates of the unspecified parameters to satisfy strong duality, if it is possible to satisfy the optimality conditions exactly. We apply these two approaches to the general linear optimization problem. We also use them to impute unspecified parameters of the uncertainty set for robust linear optimization problems under interval and cardinality constrained uncertainty. Each inverse optimization model we propose is nonconvex, but we show that a globally optimal solution can be obtained either in closed form or by solving a linear number of linear or convex optimization problems.
“…Similarly, a bidder may seek to impute several unknown parameters which can be used in the process of deciding her bid. These parameters include cost coefficients of rivals' models (Chen et al, 2019); rival bids, which are objective function coefficients in the facilitator's problem (Ruiz et al, 2013); and parameters that describe the routing of energy through the network and the capacity of transmission lines, which are left-hand-side constraint parameters in the facilitator's problem (Birge et al, 2017).…”
Most inverse optimization models impute unspecified parameters of an objective function to make an observed solution optimal for a given optimization problem with a fixed feasible set. We propose two approaches to impute unspecified left-hand-side constraint coefficients in addition to a cost vector for a given linear optimization problem. The first approach identifies parameters minimizing the duality gap, while the second minimally perturbs prior estimates of the unspecified parameters to satisfy strong duality, if it is possible to satisfy the optimality conditions exactly. We apply these two approaches to the general linear optimization problem. We also use them to impute unspecified parameters of the uncertainty set for robust linear optimization problems under interval and cardinality constrained uncertainty. Each inverse optimization model we propose is nonconvex, but we show that a globally optimal solution can be obtained either in closed form or by solving a linear number of linear or convex optimization problems.
“…If an inverse model is a good fit to the data, we can insert the fitted parameters in the original problem to obtain good predictive power [3]. Although this is a nice feature, we limit this paper to only consider formulating and solving inverse equilibrium problems, and refer the interested reader to [10] and [20] that use inverse optimization for prediction.…”
Section: Relationship To Econometrics and Machine Learningmentioning
confidence: 99%
“…Applications include the investigation of price response of consumers [5], [6], estimation of offer prices from rival producers [7], and investigation of the parameters of transmission constraints in electricity markets based on locational marginal prices [8]. Relevant work on inverse equilibrium models include [9] and [10], which use the variational inequality approach of [3] to estimate bid curves of competing firms that employ strategic bidding.…”
Section: Introductionmentioning
confidence: 99%
“…The inverse demand function (9) sets the price λ i at a particular bus i ∈ N , where N is the set of nodes, with respect to total quantity q i , slope a i and intercept b i . Equation (10) denotes the linear marginal cost for a producer p, where x p is its generation.…”
Inverse equilibrium modeling fits parameters of an equilibrium model to observations. This allows investigation of whether market structures fit observed outcomes and it has predictive power. We introduce a methodology that leverages relaxed stationarity conditions from Karush-Kuhn-Tucker conditions to set up inverse equilibrium problems. This facilitates reframing of existing equilibrium approaches on power systems into inverse equilibrium programs. We illustrate the methodology on network-constrained and unconstrained Nash-Cournot games between price-making power generators. The inverse equilibrium problems in this paper reformulate into linear programming problems that are flexible and interpretable. Still, inverse equilibrium modeling provides generally inconsistent estimation and econometric approaches are better for this purpose.
“…In addition, we use a scarcity factor similar to [5] in order to account for strategic bidding. Complementary approaches to infer aggregated supply curves [6] and electricity suppliers' cost functions [7] have also been proposed recently.…”
Anticipating electricity prices on the day-ahead market has become a key issue for both risk assessment and revenue optimization. In this paper, we propose to generate time series of prices with an hourly resolution using a structural model that simulates a simplified market clearing process. The aggregated supply curves in this model are composed of orders based on the available capacity of generation units. The ask prices are parametrized, and the calibration is performed by applying statistical learning to historical market and power system data. To reflect the strategic behavior of market participants, these prices depend on the scarcity of power at the national level. The model's performance is assessed based on the case of France with a one-year horizon and data from 2013-2015. This approach illustrates how open data on the electric power system enable links to be drawn between technical constraints and price formation. Index Terms-day-ahead markets, electricity prices, statistical learning, structural model The authors wish to thank the French Environment and Energy Management Agency (ADEME), the Association pour la Recherche et le Développement des Méthodes et Processus Industriels (ARMINES) and Coruscant SA for financially supporting this research.
scite is a Brooklyn-based organization that helps researchers better discover and understand research articles through Smart Citations–citations that display the context of the citation and describe whether the article provides supporting or contrasting evidence. scite is used by students and researchers from around the world and is funded in part by the National Science Foundation and the National Institute on Drug Abuse of the National Institutes of Health.