Market-based Approaches for the Coordinated Operation of Electricity and Natural Gas Systems

Abstract

Numerous unprecedented changes have taken place in the whole energy system over the last years, such as the transformation of the power generation mix and the increased interactions among different energy systems, which have introduced both opportunities and challenges. The electricity system is transitioning towards a renewable-based system mainly due to worldwide environmental concerns and natural gas is expected to play an important role in the future development of the electricity system. This is due to the fact that Gas-fired Power Plants (GFPPs) are one of the least polluting conventional technologies, as well as efficient and flexible. The co-existence of these two types of power production technologies serves as a promising combination for a smooth transition to a sustainable energy system that is flexible enough to accommodate high shares of renewables. As a consequence, the interactions between the electricity and natural gas systems will be strengthened, while the uncertainty and variability of renewables will eventually affect the operation of both systems. Moreover, electricity and natural gas are traded in markets, which eventually need to adapt to these recent and forthcoming changes. In this context, the objective of this thesis is to propose market-based coordination frameworks to support the operation of electricity and natural gas systems under the uncertainty introduced by the power production from renewable energy sources, such as wind and solar. An increasing need for flexibility has occurred due to the inherent characteristics, i.e. uncertainty and variability, of renewable energy sources that can be covered by various sources, such as electricity storage, demand response and GFPPs. On top of the ability of GFPPs to provide operational flexibility, these plants are the link between the two energy systems and thus give the opportunity to exploit the flexibility in the natural gas system. More specifically, the natural gas system can act as a storage buffer thanks to the ability of storing natural gas in the pipelines and in the seasonal storage facilities. A special focus in this thesis is placed on developing operational models for the optimal use of the different assets and components of the whole energy system, while simultaneously integrating them in the market operations. Currently, the electricity and natural gas markets are operated in a decoupled manned without taking into account each system’s complexities and limitations. In addition, a common practice is to adopt a deterministic view of uncertainties when operating the systems and markets, since these were fairly limited when the operational models were first developed. Seeking for new operational models, we analyze different levels of coordination in terms of coupling the systems, as well as consider various approaches to obtain an uncertainty-aware scheduling of the system in view of stochastic power production from renewables. We introduce a model that co-optimizes the electricity and natural gas systems under uncertainty by using stochastic programming, while also considering the flexibility of the natural gas system. Under this proposed approach, the importance of proper natural gas system modeling in short-term operations is highlighted to reveal flexibility and increase security of supply. Therefore, adopting a fully coupled view of the two systems and having a probabilistic description of uncertainty can provide ideal solutions for the system and market operators. Such stochastic dispatch models optimally use the available flexibility to provide an ideal dispatch with the minimum expected cost. However, these models are usually incompatible with the current market designs. System and market operators are highly challenged by the increasing penetration of renewables, since the traditional market designs are based on a sequential clearing of trading floors with deterministic view of uncertainties. As a result, these models become highly inefficient and experience high operational costs as the penetration of stochastic power production increases. Acknowledging the advantages of adopting a coupled view of the electricity and natural gas systems, we propose two dispatch models that exploit the physical and economic links established between the electricity and natural gas systems by GFPPs to improve the current sequential clearing of trading floors. The proposed improved dispatch models anticipate the real-time flexibility needs and optimally set flexibility volume and price signals to provide the system dispatch that approximates the ideal stochastic solution in terms of expected cost, while still clearing the day-ahead and real-time trading floors sequentially. Taking advantage of the continuously increasing availability of renewable power production data, recent developments on data-driven optimization can be employed to deal with uncertain renewables. Two data-driven models are proposed for the operation of the electricity and natural gas systems, which are able to efficiently capture the true characteristics of renewable power production directly from the available data. That way, the proposed data-driven models inherently incorporate valuable information regarding the spatial and temporal correlations. In the first data-driven model, we solve the energy and reserve dispatch model with fuel constraints for GFPPs in view of a strong interdependence between electricity and natural gas systems. The focus is placed on developing approaches to solve distributionally robust joint chance constrained programs, in which the sample data are directly utilized for the uncertainty characterization and a systematic tuning of robustness allows to attain cost-effective solutions. The second data-driven model is developed based on the estimation of the first- and second-order moments from the historical data and provides an independent, yet coordinated dispatch of electricity and natural gas systems in view of uncertain power supply. This distributed approach permits to share only a limited amount of information between the two system or market operators and is performed in a transparent manner.