Multi-Objective Optimisation with Focus on Environmental Models
State-of-the-art Multi-Objective Particle Swarm Optimiser (MOPSO), based on the algorithm developed by Lin et al. (2018) with improvements described by Marinao-Rivas & Zambrano-Bigiarini (2020) . This package is inspired by and closely follows the philosophy of the single objective 'hydroPSO' R package ((Zambrano-Bigiarini & Rojas, 2013) ), and can be used for global optimisation of non-smooth and non-linear R functions and R-base models (e.g., 'TUWmodel', 'GR4J', 'GR6J'). However, the main focus of 'hydroMOPSO' is optimising environmental and other real-world models that need to be run from the system console (e.g., 'SWAT+'). 'hydroMOPSO' communicates with the model to be optimised through its input and output files, without requiring modifying its source code. Thanks to its flexible design and the availability of several fine-tuning options, 'hydroMOPSO' can tackle a wide range of multi-objective optimisation problems (e.g., multi-objective functions, multiple model variables, multiple periods). Finally, 'hydroMOPSO' is designed to run on multi-core machines or network clusters, to alleviate the computational burden of complex models with long execution time.