MOUSE

MOUSE (Model Uncertainty and Sensitivity Estimation) is a software package for the management of uncertainties during the development of numerical models.

MOUSE provides tools for estimates of uncertainties, sensitivity analysis and parameter optimization.

The main features of MOUSE code are sensitivity analysis and parameter optimization tools.

The sensitivity analysis module of MOUSE software includes a wide variety of global approaches such as:

Variance-based methods (Sobol’ and FAST).
Moment-independent measures (PAWN).
Correlation and regression coefficients and Morris screening method.

Parameter optimization methods that MOUSE provides for model calibration include different heuristic methods such as:

Particle Swarm Optimization (PSO).
Genetic algorithm (GA).
Harmony search (HS).
Cuckoo search (CS).
Grey wolf optimizer (GWO).
Firefly algorithm (FA).

MOUSE’s parameter optimization module supports the use of user-specified objective function, simultaneous fitting of different experimental curves, parameter transformations, and other convenient features

MOUSE is model-independent software and could be applied to any model with known input and output structure. At the same time, MOUSE is one of the core parts of the software framework for numerical safety assessment of radioactive waste disposal and nuclear decommissioning projects that is developed by IBRAE. Due to this role MOUSE is constantly updated with specific features that are demanded by model developers in this particular field. Particular modules are being developed for interactions with GeRa, FENIA, CORIDA.

Developer: The Nuclear Safety Institute of the Russian Academy of Sciences (IBRAE).

COPYRIGHT HOLDER: Rosatom State Atomiс Energy Corporation ROSATOM on behalf of the Russian Federation.

Savel`eva E.A., Svitelman V.S. et. al.

For any additional information please contact us: mouse@ibrae.ac.ru

Site map | Feedback