Dynamic programming based methodology for the optimization of the sizing and operation of hybrid energy plants

The dynamic programming method is mainly used to deal with energy management and optimal control problems of hybrid energy plants. This paper extends the application of this method and documents the development of a dynamic programming based methodology for the optimization of both the sizing and operation of hybrid energy plants. The optimization problem is carried out with the aim of minimizing primary energy consumption over the simulation period. Moreover, in order to demonstrate the validity and usefulness of the optimization methodology presented in this paper, a comparison with an optimization methodology developed by the same authors is made. The optimization methodology used as a benchmark is based on the genetic algorithm and is commonly used in the literature. A case study consisting of a building located in the north of Italy is considered to demonstrate the developed methodology. The hybrid energy plant used to fulfil the energy demands of the building comprises a photovoltaic panel, solar thermal collector, combined heat and power, ground and air source heat pumps, hot water storage and auxiliary boiler. Compared to the genetic algorithm based methodology, the proposed methodology allows a primary energy saving and computation time saving of about 5.4% and 41%, respectively. In addition, compared to a traditional plant composed of a boiler and the grid, the developed methodology allows a primary energy saving of about 24%. The proposed methodology is fast, easy to implement and also addresses the non-linearity associated with the optimization problem of hybrid energy plants.