MindSet-Optimizer implements advanced Computational and Swarm Intelligence algorithms to perform an efficient analysis of the problem to be optimized. For a given cost function, the Optimizer adjusts system parameters, in order to optimize model values. The input objective function may be either a mathematical model or a code-generated value.
No need of mathematical modelling is required. Optimizer's User Objective function only needs to be able to return a numeric value for given inputs. Therefore, the user can enter a simulation or data analysis procedure as input.
Optimizer features a user-friendly GUI, allowing for model development in Python, or access via API from .Net or SQL. Optimizer may be added to an existing .Net project, enabling calibration of multi-parameter problems.
Minimizes \ Maximizes an objective function, given constraints and penalty function. Problems may be non-linear or multidimensional (up to dozens of variables).
Intuitive GUI for definition of optimization problems. Problem definition (along with any solution found) can be saved to a file, allowing for quick future reference
Preformed in Python language used in GUI interface. User may also define models in .Net via API.
Use API to call for optimization from C# or other .Net language.
Particle Swarm Optimization, Gravitational Search Algorithm, and Simulated Annealing are used to solve optimization problems.
The Optimizer component may be incorporated in existing .Net projects and systems.
Optimizer algorithms simulate a virtual particle swarm, which moves through the search space defined by the optimization variables. In their movement, the particles interact with one another, sharing information regarding the various solutions found. As each particle seeks to find its best individual solution, the entire swarm advances toward finding the system's optimal solution.
The advanced features of Optimizer's implemented algorithms extend beyond those known in literature.