Aiming at the problem that basic Krill Herd (KH) algorithm has poor local search ability and insufficient exploitation capacity on complex function optimization problems, a Krill Herd algorithm based on Dynamic Pressure Control operator (DPCKH) was proposed. A new dynamic pressure control operator was added to the basic krill herd algorithm, which made it more effective on complex function optimization problems. The dynamic pressure control operator quantified the induction effects of several different outstanding individuals on the target individual through Euclidean distance, accelerating the production of new krill individuals near the excellent individuals and improving the local exploration ability of krill individuals. Compared to ACO (Ant Colony Optimization) algorithm, DE algorithm, KH algorithm, KHLD (Krill Herd with Linear Decreasing step) algorithm and PSO (Particle Swarm Optimization) algorithm on 7 benchmark functions, DPCKH algorithm has stronger local exporatioin and exploitation ability.

Fuzzy rule-based models are widely used in many fields. The existing performance indexes for the models are mainly numeric, which ignore the characteristic of fuzzy sets in the models. Aiming at the problem, a new method of evaluating the performance of fuzzy rule-based models was proposed, to effectively evaluate the non-numeric (granular) nature of results formed by the fuzzy models. In this method, different from the commonly used numeric performance indexes (such as Mean Squared Error (MSE)), the characteristics of information granules were used to represent the quality of granular results output by the model and this proposed index was applied for the performance optimization of the fuzzy model. The performance of information granule was quantified by two basic indexes, coverage rate (of data) and specificity (of information granule itself), and the maximization of the output quality of granularity (expressed as the product of coverage rate and specificity) was realized with the use of particle swarm optimization. Moreover, the distribution of information granules formed through fuzzy clustering was optimized. The experimental results show the effectiveness of the proposed method on the performance evaluation of fuzzy rule-based models