Smart districts can provide flexibility from emerging distributed multi-energy technologies, thus bringing benefits to the district and the wider energy system. However, due to nonlinearity and modelling complexity, constraints associated with the internal energy network (e.g., electricity, heat and gas) and operational uncertainties (for example in energy demand) are often overlooked. For this purpose, a robust operational optimisation framework for smart districts with multi-energy devices and integrated energy networks is proposed. The framework is based on two-stage iterative modelling that involves Mixed Integer Linear Programming (MILP) and linear approximations of the nonlinear network equations. In the MILP optimisation stage, the time-ahead set points of all controllable devices (e.g., electrical and thermal storage) are optimised considering uncertainty and a linear approximation of the integrated electricity, heat and gas networks. The accuracy of the linear model is then improved at a second stage by using a detailed nonlinear integrated network model, and through iterations between the models in the two stages. To efficiently model uncertainty and improve computational efficiency, multi-dimensional linked lists are also used. The proposed approach is illustrated with a real UK district; the results demonstrate the model’s ability to capture network limits and uncertainty, which is critical to assess flexibility under stressed conditions.

A key feature of smart grids is the use of demand side resources to provide flexibility to the energy system and thus increase its efficiency. Multienergy systems where different energy vectors such as gas, electricity, and heat are optimized simultaneously prove to be a valuable source of demand side flexibility. However, planning of such systems may be extremely challenging, particularly in the presence of long-term price uncertainty in the underlying energy vectors. In this light, this paper proposes a unified operation and planning optimization methodology for distributed multienergy generation (DMG) systems with the aim of assessing flexibility embedded in both operation and investment stages subject to long-term uncertainties. The proposed approach reflects real options thinking borrowed from finance, and is cast as a stochastic mixed integer linear program. The methodology is illustrated through a realistic U.K.-based DMG case study for district energy systems, with combined heat and power plant, electric heat pumps, and thermal energy storage. The results show that the proposed approach allows reduction in both expected cost and risk relative to other less flexible planning methods, thus potentially enhancing the business case of flexible DMG systems.

Multienergy systems (MES) can optimally deploy their internal operational flexibility to use combinations of different energy vectors to meet the needs of end-users and potentially support the wider system. Key relevant applications of MES are multienergy districts (MEDs) with, for example, integrated electricity and gas distribution and district heating networks. Simulation and optimization of MEDs is a grand challenge requiring sophisticated techno-economic tools that are capable of modeling buildings and distributed energy resources (DERs) across multienergy networks. This article provides a tutorial-like overview of the state-of-the-art concepts for techno-economic modeling and optimization of integrated electricity-heat-gas systems in flexible MEDs, also considering operational uncertainty and multiple grid support services. Relevant mixed integer linear programming (MILP) formulations for two-stage stochastic scheduling of buildings and DER, iteratively soft-coupled to nonlinear network models, are then presented as the basis of a practical network-constrained MED energy management tool developed in several projects. The concepts presented are demonstrated through real-world applications based on The University of Manchester MED case study, the details of which are also provided as a testbed for future research.