Smart management tools for integrated Power-to-Gas systems

Abstract

The growing penetration of renewable energies, which have a fluctuating nature, requires the enhancement of energy system flexibility. This can be achieved through sector integration, which encompasses the conversion of energy into the most convenient vectors. Within this context, the utilization of highly integrated energy systems, known as Multi-Energy Systems (MES), where different energy vectors optimally interact with each other, becomes imperative. Power-to-Gas (PtG) technologies, i.e. the production of gaseous fuels from electricity, emerge as a promising solution when considering sector integration, by allowing surplus renewable electricity to be directly transformed into green hydrogen or methane, which can be utilized or stored. Furthermore, these fuels exhibit excellent long-term storage capabilities, making them a promising option for seasonal energy storage. However, the full potential of PtG systems can be unlocked only if the waste heat released by electrolysis and methanation processes is recovered and fed, for instance, into a district heating network to be supplied to an end-user. A smart energy system, however, which includes these innovative solutions and allows for full integration of the fuel, electrical, and heating sectors, requires advanced management and control tools for optimizing its operation. The scope of this thesis is to investigate the operational strategies of energy systems integrated with PtG solutions by developing novel planning and control tools that enable optimal system management. The core of this research involves three main analyzed problems. First, an innovative optimization model for the system is introduced, formulated as a Mixed-Integer Linear Programming (MILP) problem. The model tackles the uncertain nature of future disturbances, such as energy needs, generation, and price through a two-stage stochastic programming approach. The algorithm is tested on grid-connected and positive energy districts case studies, allowing for more robust optimization compared to a deterministic approach. Furthermore, the integration of PtG solutions ensures the energy security of the systems and acts as a buffer to forestall unpredictable behavior of the disturbances. Second, a control strategy based on Model Predictive Control (MPC) is presented. The controller aims to operate the production of methane from a PtG system and the supply of waste heat to a district heating network with minimal cost. The MPC makes use of a detailed MILP algorithm, able to optimize the system over a future time horizon. The feasibility of the controller is demonstrated through a Model-in-the-Loop simulation platform, and its performances are compared to those obtained with a conventional controller. The novel controller enables a 54 % increase in operating margin and more than halves carbon dioxide emissions. A better exploitation of renewable energy is also obtained (+4.6 %), as well as an increase in the share of heat recovered from the PtG plant. Lastly, a novel control architecture is proposed that combines the first two developed tools into a more comprehensive and exhaustive tool for coordinating multiple MES integrated by means of a shared natural gas seasonal storage. Each individual MES has its own short-term control logic based on MPC, managing the day-ahead energy exchanges, while a long-term MPC controller, employing a two-stage stochastic programming MILP algorithm, takes into account yearly dynamics and the interactions among the different energy systems, managing the seasonal storage. It provides additional constraints to the short-term controllers, ensuring that yearly goals are met. With the developed control architecture, a multi-temporal and multi-spatial control is obtained. The proposed management is validated in a Model-in-the-Loop configuration, and the benefits of the novel control strategy are quantified. Notably, a smart management for the system is achieved, and the controller is able to optimally control the system by making use of the seasonal storage to balance the seasonal mismatch between production and demand. Indeed, the surplus renewable generation is stored when available, and used in periods of shortage, resulting in a higher utilization of renewable energy and lower emissions and costs. Overall, the tools proposed in this thesis offer innovative solutions for the effective integration of PtG systems and their optimal management. They are versatile tools, and their utilization for different case studies is straightforward as they were developed in a general way. By optimizing energy management, enhancing efficiency, and ensuring sustainability of energy systems, the novel proposed tools allow taking some steps forward in the realization of smarter and more resilient energy systems for a sustainable future.