In the next years energy grids are expected to become increasingly complex, due to the integration between traditional generators (operating with fossil fuels, especially natural gas), renewable energy production systems and storage devices. Furthermore, the increase of installed distributed generation systems is posing new issues for the existing grids. The integration involves both electric grids and thermal networks, such as district heating networks. In this scenario, it is fundamental to optimize the production mix and the operation of each system, in order to maximize the renewable energies exploitation, minimize the economic costs (in particular the fossil fuel consumption) and the environmental impact.
The aim of this paper is the analysis of different solutions in terms of energy generation mix, in order to define the optimal configuration for a given network.
With this purpose, in this study a real district heating network served by a combined heat and power unit and four boilers has been considered. The current mode of operation of the selected network has been simulated, in order to individuate eventual criticism and/or improvement possibility. On the basis of the obtained results, several scenarios have been developed by considering the addition of thermal or electric energy production systems from renewable energy sources and/or heat pumps.
For a given scenario, a whole year of operation has been simulated with an in-house developed software, called EGO (Energy Grid Optimizer), based on genetic algorithms and able to define the load distribution of a number of energy systems operating into an energy grid, with the aim to minimize the total cost of the energy production. Further considered constraints have been the avoiding of thermal dissipations and the minimization of the electric energy sale to the national grid (in order to increase the grid stability).
The carried out analysis has allowed to evaluate the yearly fuel consumption, the yearly electric energy sold to the network and the yearly electric energy purchased from the network, for each of the developed configurations. In this study the obtained results have been discussed in order to compare the proposed scenarios and to define an optimal solution, which enables to reduce the yearly operation costs of the production plant.