A new dynamic bacterial foraging optimization and its application on model reduction

Inspired by the foraging behavior of E. coli bacteria, bacterial foraging optimization (BFO) has emerged as a powerful technique for solving optimization problems. However, BFO shows poor performance on complex and high-dimensional optimization problems. In order to improve the performance of BFO, a new dynamic bacterial foraging optimization based on clonal selection (DBFO-CS) is proposed. Instead of fixed step size in the chemotaxis operator, a new piecewise strategy adjusts the step size dynamically by regulatory factor in order to balance between exploration and exploitation during optimization process, which can improve convergence speed. Furthermore, reproduction operator based on clonal selection can add excellent genes to bacterial populations in order to improve bacterial natural selection and help good individuals to be protected, which can enhance convergence precision. Then, a set of benchmark functions have been used to test the proposed algorithm. The results show that DBFO-CS offers significant improvements than BFO on convergence, accuracy and robustness. A complex optimization problem of model reduction on stable and unstable linear systems based on DBFO-CS is presented. Results show that the proposed algorithm can efficiently approximate the systems.