Long-term Hydrothermal Bid-based Market Simulator

Abstract

Simulating long-term hydrothermal bid-based markets considering strategic agents is a challenging task. The representation of strategic agents considering intertemporal constraints within a stochastic framework brings additional complexity to the already difficult single-period bilevel, thus, non-convex, optimal bidding problem. Thus, we propose a simulation methodology that effectively addresses these challenges for large-scale hydrothermal power systems. We demonstrate the effectiveness of the framework through a case study with real data from the large-scale Brazilian power system. In the case studies, we show the effects of market concentration in power systems and how contracts can be used to mitigate them. In particular, we show how market power might affect the current setting in Brazil. The developed method can strongly benefit policymakers, market monitors, and market designers as simulations can be used to understand existing power systems and experiment with alternative designs.

Figures (9)

• Figure 1: Centralized cost-based markets.
• Figure 2: Bid-based market.
• Figure 3: Revenue curves, $\tilde{\Lambda}(e, \omega)$, for various values of $Q^F$, and $P^F = 0$.
• Figure 4: Algorithm for multiple agents equilibrium flowchart.
• Figure 5: Results for simulations of Brazil's Southeast under different market concentrations. All plots have the same template and contain average values of: (a) Spot Prices, (b) Normalized Revenue, (c) Spillage, and (d) Storage Level. Averages are with respect to all stages and scenarios. Spillage and Storage $\%$ are with respect to the maximum amount of water that can be stored in the system. All plots include a first bar (in green) with the result for the centralized case. In parentheses in the horizontal axis, we have the percent share of each price-maker agent, defining the database used. For each database, we have 3 bars, one for a case with no contracts (in red) and another two for cases with 75% of contracting level (in purple) and 100% of contracting level (in blue). Note that some red bars in (a), (b), and (d) are too high and do not fit the plot area, so they contain the values beside the top of the bar.