论文部分内容阅读
为了获得准确的光电跟踪伺服系统的模型,采用自适应差分进化算法对光电跟踪伺服系统进行辨识研究,该算法根据辨识误差平方和自动调整变异、交叉因子。在输入为正弦离散数字信号下辨识系统的离散模型。为了验证算法的有效性,在频域内与扫频法辨识的一、二阶模型和系统实际输出比较研究。实验结果表明:在相同正弦离散信号条件下,辨识输出与系统实际输出基本一致,与扫频法的RMSE相比减小了20.33%,二阶模型在高频段偏离系统实际输出稍大些,一阶系统输出与系统实际输出基本一致。研究结果表明,自适应差分进化算法计算量小,方法简便,辨识准确,在光电跟踪伺服控制领域具有一定的工程应用价值。
In order to get an accurate model of the electro-optical tracking servo system, the adaptive differential evolution algorithm is used to identify the photoelectric tracking servo system. The algorithm adjusts the variation and cross-factors automatically according to the square of the identification error. Discrete model of the system is identified under the input of sinusoidal discrete digital signals. In order to verify the validity of the algorithm, a comparative study on the actual output of the first and second order models and systems identified by the swept frequency method in the frequency domain is carried out. The experimental results show that under the same sinusoidal discrete signal, the output of identification is basically consistent with the actual output of the system, which is reduced by 20.33% compared with the RMSE of the swept-frequency method. The second-order model is slightly larger than the actual output of the system in high- Order system output and the actual output is basically the same. The results show that the adaptive differential evolution algorithm has the advantages of low computational complexity, simple method and accurate identification, and has certain engineering application value in the field of photoelectric tracking servo control.