Our research activities are in the realm of power and energy systems, combined with elements from decision and control sciences, economics and energy policy. In the past, we have been able to work on the understanding of the impact, viewed from the ideal competitive equilibrium setting, of dynamics, constraints, and uncertainty which are inherent to power and energy systems. In a general dynamic setting, we establish many of the standard conclusions of the competitive equilibrium theory: Market equilibria are efficient, and average prices coincide with average marginal costs. However, these conclusions hold only on average and the dynamics of prices can be extreme. Price volatility, negative prices and price spikes hold even in a perfect competitive setting in which price manipulation is excluded. These findings along with other recent academic results and the empirical experience of the last decades illustrate the need for developing new models and tools for analyzing energy systems and its associated markets.
Developing such models and tools is a challenging task. Mainly, because energy systems require the coexistence of two coupled dynamical systems: a physical system driven by hard and soft engineering constraints in which reliability is one of the main objectives, and a market/financial system driven by self-interests of players in which economic efficiency is the leading metric. In the particular case of electricity markets and as a result of the many changes expected in the future’s grid — the use of information technologies, new energy policies, active demand participation and renewable energy sources among other ones — increased levels of uncertainty and exotic dynamics will naturally emerge. These features will make even more challenging the operation and planning of energy systems and the coexistence with its associated markets.
Our research has been focused in understanding the interaction between these two complex systems, developing models and tools tailored for a grid with increased levels of uncertainty and dynamics, and proposing market designs and energy policies appropriate for this new setting. Current research projects include:
Characterization of Flexibility in Power Systems: Characterization of power systems in very short time scales is a very challenging problem for the day-ahead and real-time operations, specially taking into account uncertainty and volatility of renewable sources. Efficient ways of characterizing the requirements imposed by the uncertain load on the one hand, and available resources to deploy on the other, could lead to a market which considers flexibility attributes of generators and other emerging resources in power systems.
Electricity Market Design: Proposal of an alternative architecture for electricity markets based on the notion of multi-attribute products and the definition of contracts. Investigate the product definition in electricity auctions and its key role on market outcomes. Study of the design and outcomes of electricity auctions and capacity markets. Understand how electricity markets should be organized for systems without fuel. Development of new market structures tailored to flexible products. Current research include duration-differentiated energy contracts, rate-constrained energy contracts and the development of real applications in which these products are integrated into the electricity markets.
Demand response aggregators’ participation in electricity markets: We focus in understand how to promote demand response, analyzing the usage of bilateral contracts or real-time pricing. Moreover, it is crucial to figure it out how a third-party aggregator can use the aggregated DR resource in an optimal way in electricity markets, while providing rewards to incentivize the consumers to participate in several DR programs. We also investigate how a DR aggregator participate in multi-settlement electricity markets, deciding for example how much energy purchasing in Day-Ahead markets to provide profitable balance requirements in Real-Time markets. For these purposes, we develop different optimization models that considers several aspects of the interaction between a DR aggregator and its consumer, while negotiating with the system/market operator in specific market settings.
Modeling consumer behavior in demand response schemes: Consumer behavior is key for the success of demand response applications. Standard way for modeling consumers relies on welfare maximizations formulations in which consumers optimize. An alternative ways for modeling consumers is the so-called Satisficing Theory developed by Herbert Simon. This theory brings elements from bounded rationality and the notion that consumers do not optimize but rather take decisions based on satisfaction threshold. In this project, we use satisficing theory to model the decisions of consumers who participate in DR programs.
Studying the flexibility potential of desalination plants: Desalination plants use large amounts of energy to produce fresh water from seawater. However, their energy costs may be significantly reduced by coordinating their power consumption with renewable energy generation.
Vehicle-to-grid charging strategies: Understanding how different optimization schemes impact the capacity of an electric vehicle fleet to participate in electricity markets (e.g. frequency regulation via tracking an AGC signal), while not affecting the energy requirements of the vehicle’s owner, and reducing battery cycling.
Development of a virtual power plant: The inclusion of intermittency by renewable energy imposes a need for increasing flexibility capabilities. Communication networks, that enable the coordinated management of flexible residential and industrial loads, can supply this flexibility. In this project we are constructing the hardware technology to implement this coordinated management of resources.
Development of aggregated models for DER: The research is focused on the creation of aggregated models that represent distributed energy resources including demand response resources in commercial, residential and industrial areas. These models are the scaffolding required for tackling distribution systems expansions problems.
Impact of Dynamics and Uncertainty: Study of the impact of volatility and dynamics on electricity markets. Using current operational schemes and typical assumptions, it was showed that under current electricity market designs the volatility of wind can reduce its inherent value without proper flexibility in the system. Moreover, the loss in social welfare is skewed to the supplier. The results also illustrate the need to move beyond snap-shot-based models for analyzing electricity markets and reinforce the idea of multi-attribute products.
Resource Planning Models: Development of models and tools for planning energy systems with increased levels of uncertainty and dynamics. The inherent complexity, uncertainty and dynamics was handled using techniques and methodologies from decision & control, and simulation & learning. Discussion about the need to upgrade usual power and energy reliability metrics. Study of composite reliability metrics suitable for planning studies in a market environment. Currently, our work focuses on proposing a model of expansion planning of generation and transmission that considers hydro constraints, uncertainty and operation constraints from the inclusion of an approximation of the unit commitment problem.
Resilience: Resilience of power systems is associated with the ability to anticipate, withstand, and rapidly recover from a wide range of natural and man-made disruptions. One aspect studied in this area is how to quickly restore an electric power system that has been significantly disrupted, which involves optimizing damage assessment and crew deployment activities, as well as other logistical decisions that have to be carried out taking into account power dispatch decisions throughout the restoration process. Another aspect studied is how to design power system expansion plans that maintain a proper level of resilience to disruptions and how to design strategies that strengthen the operation of the power grid through decision-support tools that can help identify liabilities, prioritize key infrastructures, and provide logistical support for emergency crews while reducing response time and cost.