A MINLP optimization of the configuration and the design of a district heating network: Academic study cases

- A combinatory non-linear optimization in steady state is performed.
- Optimization of the topology and the sizing at the same stage.
- Potential to reduce the total cost when allowing the cascade of consumers.
- Isolated consumer: centralized, decentralized or isolated heat supply.
- Academic study case to illustrate the ability of this optimization formulation.

The aim of this work is to propose a tool for the design assistance of District Heating Network (DHN). Two goals of DHN optimization are handled simultaneously: the optimization of the configuration and its design. The optimization objective is to minimize the global cost of the DHN over 30 years. It includes both operating costs (heating and pumping cost, including thermal losses and pressure drop) and investment costs (line, trench, heating plant, heat exchanger). The formulation leads to a mixed integer non-linear programming (MINLP) problem in steady state. The model is solved with DICOPT within GAMS (around 5s for this study cases).One of the outputs of these academic study cases is the layout of the DHN, supplied in parallel or in cascade: a consumer with hot temperature requirement can supply another consumer with lower temperature requirement. Even a looped network in cascade is optimal (−4.6% total cost reduction) when the cost of the trench is lower than 500 €/m. Furthermore, different structures are optimal (between −4 and −8% of total cost reduction) depending on whether the heat production(s) are decentralized, centralized, isolated collective, renewable or not. Finally the balance between heat loss and pressure drop is detailed.