Abstract: There is relatively less research on cross-field general algorithms,and the effectiveness needs to be further improved.Aiming at this problem,an eagle eye optimization algorithm for cross-field problems that combined Depth of Field Control (DFC) and Lens Imaging Learning (LIL) was proposed.The DFC and LIL strategies were introduced into the eagle eye of the Golden Eagle Optimization (GEO) algorithm.The distance between the golden eagle and prey was converted into the distance between the shooting focus and camera focus plane.Combined with other shooting parameters,the front and back depth of field,the focal point of the mirror and the actual position of the object were calculated,and the position of new golden eagle individual was updated.In addition,to test the generality of the algorithm,the performances of different algorithms on 56 open-source test functions and 5 cross-field datasets were compared.The experimental results indicated that the proposed algorithm had some competitiveness.

Key words: eagle eye optimization, depth of field control, swarm intelligence algorithms, cross-field optimization, engineering applications

摘要： 鉴于跨领域通用算法的研究相对较少,且效果有待进一步提升,提出一种解决交叉领域问题的结合景深控制和透镜成像学习的鹰眼优化算法。在金鹰优化算法的鹰眼中引入景深控制和透镜成像学习策略,将金鹰与猎物的距离转换为拍摄焦点与相机焦点平面之间的距离,结合其他拍摄参数计算前后景深、镜像的对焦点和物体的实际位置,并更新金鹰新个体的位置。为了测试算法的通用性,分别在56个开源测试函数和5个交叉领域数据集上对比了不同算法的效果,结果表明所提算法具有一定竞争力。

关键词: 鹰眼优化算法, 景深控制, 群智能算法, 跨领域优化, 工程优化