Ship data evolution sampling method based on Levy distribution

The invention discloses a ship data evolution sampling method based on Levy distribution. According to the algorithm, a transition probability function of adaptive Levy distribution is provided. And enabling each sample in the sample set to generate two samples to be selected according to parameter values of Levy distribution of 1.0 and 2.0, updating the obtained two groups of sample sets through rejection selection operation, and comparing standard deviations generated by the two groups of sample sets to determine a final sample set. The self-adaptive Levy distribution enables the selection of parameters to depend on the evolution process. Through a series of experiments performed on 15 test functions and actual ship data, the performance of four evolution sampling algorithms is compared. It is proved that an evolution sampling algorithm based on adaptive Levy distribution not only improves the convergence speed, but also increases the precision of samples. 基于Levy分布的船舶数据演化采样方法。该算法提出了一个自适应的Levy分布的转移概率函数。令样本集的