Introduction. Recent developments in the structure of energy systems, driven by the increasing share of renewable energy sources, the development of electric vehicles, and the widespread adoption of equipment with load management capabilities, have made conventional operation methods inadequate for meeting the requirements of complex energy systems. In multi-carrier energy networks, the interconnection among different energy carriers, such as electrical and thermal energy, provides a valuable opportunity to enhance flexibility and improve economic performance. However, the stochastic behavior of renewable energy sources, variations in energy consumption, and the technical constraints of equipment pose significant challenges to the decision-making process. Therefore, the development of intelligent energy management methods capable of simultaneously considering uncertainty, operational constraints, and multiple operational objectives has become increasingly important. The objective of this study is to propose a model predictive control (MPC)-based energy management method for the optimal operation of a multi-carrier energy hub while simultaneously considering uncertainty, operational constraints, and economic and environmental objectives.
Methods. In this study, an energy management method based on model predictive control (MPC) is proposed for the optimal operation of a multi-carrier energy hub. Unlike fixed scheduling methods, which derive operational decisions solely based on predetermined conditions, the proposed framework employs a rolling horizon strategy to evaluate the current state of the system at each time interval and update control decisions based on predicted future information. In the developed model, renewable energy generation sources, including wind and solar energy, are modeled by considering the inherent uncertainty of renewable generation. The Rayleigh probability density function is used to characterize the probabilistic behavior of wind speed, while the beta distribution is employed to represent variations in solar irradiance.
In the proposed framework, in addition to generation sources, demand-side components are considered active elements of the energy system. Electric vehicles with charging and discharging capabilities, as well as responsive loads with the ability to curtail or shift their consumption, are incorporated into the optimization model as sources of flexibility. The energy management problem is formulated as a multi-objective economic-environmental model in which the reduction of operating costs and the limitation of carbon dioxide emissions are simultaneously considered in the decision-making process. The proposed model is implemented in the MATLAB environment, and the optimization problem is solved using the YALMIP environment and the Gurobi solver.
Results and Discussion. To evaluate the effectiveness of the proposed method, three different operating scenarios were investigated. The simulation results under the base scenario showed that, without utilizing flexibility capabilities, the daily operating cost was $28.64 and CO₂ emissions were 215 kg. In the second scenario, by activating the management of electric vehicles and responsive loads, energy consumption could be shifted toward more favorable operating conditions, resulting in a 10.4% reduction in operating cost and a 24.7% reduction in CO₂ emissions compared with the base case. In the third scenario, by incorporating the environmental component into the objective function, the proposed model increased the utilization of clean energy sources and reduced CO₂ emissions to 120 kg per day. This value represents a 44.2% reduction in emissions compared with the base case. Although achieving this level of emission reduction was accompanied by a limited increase in operating cost, this result demonstrates the trade-off between the economic and environmental objectives in system operation.
Furthermore, the performance of the proposed controller was compared with those of the PID and LQR methods. The results showed that, owing to its ability to predict future behavior, directly incorporate technical constraints, and repeatedly solve the optimization problem, MPC provided more suitable performance in the simultaneous management of operating cost, emissions, and operational conditions.
Conclusions. Based on the obtained results, the proposed framework can be employed as an effective approach for managing next-generation energy systems, including multi-carrier energy hubs, microgrids, and intelligent systems equipped with renewable energy sources, electric vehicles, and flexible loads. The simultaneous use of model predictive control, renewable energy sources, and demand-side flexibility capabilities enables uncertainty, operational constraints, and economic and environmental objectives to be incorporated into the operational decision-making process.