An iterative learning control method based on an equilibrium single evolution cuckoo algorithm

Abstract

The invention relates to an iterative learning control method based on a balanced single evolution cuckoo algorithm. a novel equilibrium single evolution evaluation strategy is given; each generationof evolution only randomly updates the single dimension of the target function; the dimensionality of random updating obeys integer uniform distribution; combining with other dimensions to form a newcandidate solution; evaluating the candidate solution, if the candidate solution is superior to the fitness value of the previous generation of function, keeping the updated candidate solution and continuing to evolve until an algorithm stop condition is met, and only receiving an update value capable of improving the current candidate solution by adopting a greedy rule, so that the targeted adjustment of a search direction in an optimization process is ensured, and the efficiency is not influenced. According to the method, the global search capability and the local optimization capability ofthe cuckoo algorithm can be effectively balanced, the hysteresis phenomenon appearing at the end of execution of the optimization algorithm is avoided, and therefore the global search speed and the convergence precision of the algorithm are improved.

Description

technical field [0001] The invention relates to the technical field of iterative learning control, in particular to an iterative learning control method based on a balanced single-evolution cuckoo algorithm. Background technique [0002] Iterative learning control is a control method that seeks a learning control law that enables the system output of the controlled object to track the ideal expected trajectory in a finite time period to achieve zero-error trajectory tracking. Iterative learning control was first proposed by Japanese scholar Uchiyama in 1978, and has been widely studied by the industry since the pioneering contribution made by Japanese scholar Arimoto et al. in 1984, and has achieved rapid development. Iterative learning control is suitable for controlled objects with repetitive motion properties, and the ideal value of control input is obtained through iterative correction. It is worth noting that the iterative learning control method does not depend on the...

AI Technical Summary

Problems solved by technology

[0006] The purpose of the present invention is to improve on the existing method, to provide a kind of iterative learning control method based on balanced single evolutionary cuckoo algorithm, aiming at overcoming the defect of traditional norm optimal iterative learning control

Images