文章着重研究了含分数阶微分算子的van der Pol方程的数值解法。首先,基于Adams离散提出了一种针对Caputo分数阶导数的离散格式;然后,进一步基于Newmark-β法构造了完整的逐步迭代格式;最后,通过Newton-Raphson迭代求得了非线性系统的...文章着重研究了含分数阶微分算子的van der Pol方程的数值解法。首先,基于Adams离散提出了一种针对Caputo分数阶导数的离散格式;然后,进一步基于Newmark-β法构造了完整的逐步迭代格式;最后,通过Newton-Raphson迭代求得了非线性系统的响应。在算例分析部分,讨论了分数阶次为0<α<1和1<α<2的van der Pol系统的数值响应。当α→1和α→2时,将所提算法和四阶Runge-Kutta法进行了对比。结果表明,所提数值方法对整数阶微分系统也同样适用。展开更多
Artificial bee colony(ABC) algorithm is motivated by the intelligent behavior of honey bees when seeking a high quality food source. It has a relatively simple structure but good global optimization ability. In order ...Artificial bee colony(ABC) algorithm is motivated by the intelligent behavior of honey bees when seeking a high quality food source. It has a relatively simple structure but good global optimization ability. In order to balance its global search and local search abilities further, some improvements for the standard ABC algorithm are made in this study. Firstly, the local search mechanism of cuckoo search optimization(CS) is introduced into the onlooker bee phase to enhance its dedicated search; secondly, the scout bee phase is also modified by the chaotic search mechanism. The improved ABC algorithm is used to identify the parameters of chaotic systems, the identified results from the present algorithm are compared with those from other algorithms. Numerical simulations, including Lorenz system and a hyper chaotic system, illustrate the present algorithm is a powerful tool for parameter estimation with high accuracy and low deviations. It is not sensitive to artificial measurement noise even using limited input data.展开更多
文摘文章着重研究了含分数阶微分算子的van der Pol方程的数值解法。首先,基于Adams离散提出了一种针对Caputo分数阶导数的离散格式;然后,进一步基于Newmark-β法构造了完整的逐步迭代格式;最后,通过Newton-Raphson迭代求得了非线性系统的响应。在算例分析部分,讨论了分数阶次为0<α<1和1<α<2的van der Pol系统的数值响应。当α→1和α→2时,将所提算法和四阶Runge-Kutta法进行了对比。结果表明,所提数值方法对整数阶微分系统也同样适用。
基金supported by the National Natural Science Foundation of China(Grant Nos.11172333&11272361)the Guangdong Province Natural Science Foundation(Grant No.2015A030313126)the Guangdong Province Science and Technology Program(Grant Nos.2014A020218004&2016A020223006)
文摘Artificial bee colony(ABC) algorithm is motivated by the intelligent behavior of honey bees when seeking a high quality food source. It has a relatively simple structure but good global optimization ability. In order to balance its global search and local search abilities further, some improvements for the standard ABC algorithm are made in this study. Firstly, the local search mechanism of cuckoo search optimization(CS) is introduced into the onlooker bee phase to enhance its dedicated search; secondly, the scout bee phase is also modified by the chaotic search mechanism. The improved ABC algorithm is used to identify the parameters of chaotic systems, the identified results from the present algorithm are compared with those from other algorithms. Numerical simulations, including Lorenz system and a hyper chaotic system, illustrate the present algorithm is a powerful tool for parameter estimation with high accuracy and low deviations. It is not sensitive to artificial measurement noise even using limited input data.
