PyDOLPHYN: Dynamic simulations and optimization of multi-energy assets

The optimal design and operation of (future) multi-energy assets crucially depend on a dynamic description of the system components. Assets will face increasing levels of dynamics associated with variability of renewable energy and feedstocks, or volatile market conditions. In this study, a flexible Python framework (PyDOLPHYN) that integrates techno-economics parameters with a physics-based description of energy components is introduced. The framework is structured in three levels: a core layer containing a library of multi-fidelity component models, an intermediate layer defining asset configurations and operations, and an upper layer to perform optimization and sensitivities. Advantages and limitations of the framework are described and results for green hydrogen production assets are showcased. Focus is on optimally sizing asset components. Comparison with static approaches revealed that dynamic simulations are critical to avoid unfeasible or oversized designs. Furthermore, non-trivial trends in component sizes are found as a function of demand variability and strictness of the security of supply requirements. An additional example quantifying the impact of different operational strategies on the asset profit when coupled with (fictitious) hydrogen and electricity markets is also provided. More design and operational challenges can be addressed by straightforward adaptation of PyDOLPHYN.