Sheng Zheng, Wang Jun, Zhou Shudao, Zhou Bihua
College of Meteorology and Oceanography, PLA University of Science and Technology, Nanjing 211101, China.
National Defense Key Laboratory on Lightning Protection and Electromagnetic Camouflage, PLA University of Science and Technology, Nanjing 210007, China.
Chaos. 2014 Mar;24(1):013133. doi: 10.1063/1.4867989.
This paper introduces a novel hybrid optimization algorithm to establish the parameters of chaotic systems. In order to deal with the weaknesses of the traditional cuckoo search algorithm, the proposed adaptive cuckoo search with simulated annealing algorithm is presented, which incorporates the adaptive parameters adjusting operation and the simulated annealing operation in the cuckoo search algorithm. Normally, the parameters of the cuckoo search algorithm are kept constant that may result in decreasing the efficiency of the algorithm. For the purpose of balancing and enhancing the accuracy and convergence rate of the cuckoo search algorithm, the adaptive operation is presented to tune the parameters properly. Besides, the local search capability of cuckoo search algorithm is relatively weak that may decrease the quality of optimization. So the simulated annealing operation is merged into the cuckoo search algorithm to enhance the local search ability and improve the accuracy and reliability of the results. The functionality of the proposed hybrid algorithm is investigated through the Lorenz chaotic system under the noiseless and noise condition, respectively. The numerical results demonstrate that the method can estimate parameters efficiently and accurately in the noiseless and noise condition. Finally, the results are compared with the traditional cuckoo search algorithm, genetic algorithm, and particle swarm optimization algorithm. Simulation results demonstrate the effectiveness and superior performance of the proposed algorithm.
本文介绍了一种用于确定混沌系统参数的新型混合优化算法。为了克服传统布谷鸟搜索算法的缺点，提出了一种自适应布谷鸟搜索与模拟退火算法相结合的算法，该算法在布谷鸟搜索算法中融入了自适应参数调整操作和模拟退火操作。通常，布谷鸟搜索算法的参数保持不变，这可能会导致算法效率降低。为了平衡和提高布谷鸟搜索算法的精度和收敛速度，提出了自适应操作来适当地调整参数。此外，布谷鸟搜索算法的局部搜索能力相对较弱，这可能会降低优化质量。因此，将模拟退火操作融入布谷鸟搜索算法中，以增强局部搜索能力，提高结果的准确性和可靠性。分别通过无噪声和有噪声条件下的洛伦兹混沌系统对所提出的混合算法的功能进行了研究。数值结果表明，该方法在无噪声和有噪声条件下都能高效、准确地估计参数。最后，将结果与传统布谷鸟搜索算法、遗传算法和粒子群优化算法进行了比较。仿真结果证明了所提算法的有效性和优越性能。
