Identification of Interval Discrete Models based on the Bee Swarm Optimization Algorithm with Adaptive Tuning of the Probability of Selecting Structural Elements

The problem of structural identification of interval discrete models of complex objects using the inductive approach based on the behavioral model of a bee colony (BMBC) is considered. A new computational scheme for the method is proposed, which involves introducing a quadratic objective function and relationships for adaptively changing the distribution law when randomly selecting structural elements of a difference equation. The first innovation allows the use of gradient optimization methods. The effectiveness of the proposed method in reducing its computational complexity is investigated. It is shown that the proposed new computational scheme makes it possible to reduce the computational complexity of identifying interval discrete models of systems, and the effectiveness of the method increases with the complexity of the model.