SHIPP: Sizing optimization for HybrId Power Plants

Description

SHIPP is used for studying the design and operation of hybrid power plants, i.e. power plants combining one or more renewable energy production with energy storage systems.

Documentation is available at https://jennaiori.github.io/shipp/

Installation

The package can be installed using pip.

pip install shipp

Usage

Examples are given in the folder examples/.

The folder experiments/ contains scripts to reproduce the results presented in the following publications:

• hyb24_bl_hpp/: Iori, J., Zaaijer, M., von Terzi, D., & Watson, S. (2024). Design drivers for the storage system of baseload hybrid power plants. In 8th International Hybrid Power Plants and Systems Workshop, HYB 2024 (2 ed., Vol. 2024, pp. 245-250) https://doi.org/10.1049/icp.2024.1844
• we25_robust_dispatch/: Iori, J., Zaaijer, M., Kreeft, J., von Terzi, D., Watson, S. (2025) Reliable operation of wind-storage systems for baseload power production, WindEurope Annual Event 2025 Copenhagen, Denmark. https://doi.org/10.1088/1742-6596/3025/1/012023

Dependencies

A valid access or license to a solver compatible with pyomo (MOSEK, CPLEX, Gurobi, etc.) is recommended to solve large problems (see more information here: https://www.pyomo.org/).

[1.2.2]

• Changed class Storage: the depth-of-charge parameter dod is replaced by a minimum and maximum state-of-charge soc_min and soc_max.
• Changed functions solve_lp_pyomo, solve_lp_sparse and solve_dispatch_pyomo: the input p_min is now optional and default to 0.
• Changed function solve_dispatch_pyomo to take a single dictionary argument options for optional arguments like fixed_cap and the penalty factors (similar to solve_lp_pyomo and solve_lp_sparse).
• Changed optimization problem implemented in solve_dispatch_pyomo
  • The ramp-limit is now characterized by separate up- and down- bounds dp_min and dp_max, with dp_min $\leq 0$.
  • Addition and renaming of tuning parameters beta_obj and gamma_obj in alignment with the optimization problems implemented in solve_lp_sparse and solve_lp_pyomo.
  • Addition of a slack variable to reduce deviations to the baseload constraint, similar to what was implemented for the ramp constraint.
  • Reformulation of the ramp-limit and baseload constraint at the first time step to combine a binary variable and a slack variable.
  • Addition of tuning parameters mu1_obj, mu2_obj and mu3_obj for the total reliability and the slack variables of the baseload and ramp penalties, respectively.

Authors and acknowledgment

This project is developed by Jenna Iori at Delft University of Technology and was initially part of the Hollandse Kust Noord wind farm innovation program, with funding from CrossWind C.V.

The code is release under the Apache 2.0 License (see License.md).

Copyright notice:

Technische Universiteit Delft hereby disclaims all copyright interest in the program “SHIPP” (a design optimization software for hybrid power plants) written by the Author(s).

Henri Werij, Faculty of Aerospace Engineering, Technische Universiteit Delft.

© 2024-2026, Jenna Iori