改进型免疫克隆布谷鸟算法求解软后坐火炮多参数辨识  

Multi-parameter identification of soft recoil artillery launch process using IICCA

作　　者：赵伟 侯保林 鲍丹 ZHAO Wei;HOU Baolin;BAO Dan(College of Mechanical Engineering,Nanjing University of Science and Technology,Nanjing 210094,China;National Key Laboratory of Transient Physics,Nanjing University of Science and Technology,Nanjing 210094,China)

机构地区：[1]南京理工大学机械工程学院,南京210094 [2]南京理工大学瞬态物理国家重点实验室,南京210094

出　　处：《振动与冲击》2023年第21期43-51,62,共10页Journal of Vibration and Shock

基　　金：武器装备预先研究项目(30107070201);国防重点实验室基金(2022-JCJQ-LB-061-02)。

摘　　要：软后坐火炮利用前冲动能抵消部分后坐能量,从而减小火炮发射过程中的冲击与振动。为进一步研究软后坐火炮发射过程中的动力学特性以及解决软后坐火炮发射过程动力学模型多参数辨识困难的问题,首先依据伯努利方程建立软后坐火炮发射过程的动力学模型,其次提出了一种改进型免疫克隆布谷鸟算法(improved immune clone cuckoo algorithm,IICCA)。该算法在更新过程中引入随机交叉与高频变异提高了算法的局部搜索能力,引入自适应算子来克服算法不易取得最佳概率值的问题,引入精英抗体提取算子来提高算法的计算效率。为进一步提高算法的收敛速度,引入动态疫苗接种策略,对疫苗接种后的抗体种群采用莱维飞行和巢寄生行为进行二次搜索。通过5个多峰测试函数对改进算法进行验证,计算结果表明该改进算法相比免疫克隆选择算法、免疫遗传算法、自适应粒子群算法和改进型布谷鸟算法均具有更高的计算精度和更快的收敛速度。最终,辨识结果与试验结果的曲线相似度高达97.02%,表明IICCA在解决软后坐火炮发射过程多参数辨识问题上的有效性和准确性。Soft recoil artillery utilizes forward kinetic energy to offset some recoil energy,and reduce impact and vibration in artillery launch process.Here,to further study dynamic characteristics of soft recoil artillery launch process and solve the problem of difficult identification of multi-parameter in dynamic model of soft recoil artillery launch process,firstly,a dynamic model of soft recoil artillery launch process was established based on Bernoulli equation,and secondly,an improved immune clone cuckoo algorithm(IICCA) was proposed.The algorithm could introduce random crossover and high-frequency mutation in update process to improve its local search ability,introduce an adaptive operator to overcome the difficulty of algorithm not easily obtaining the optimal probability value,and introduce an elite antibody extraction operator to improve the algorithm's computational efficiency.In order to further improve the algorithm's convergence rate,dynamic vaccination strategy was introduced,and antibody population after vaccination was searched twice using Lévy flight and nest parasitism behavior.The improved algorithm was verified with 5 multi-peak test functions.The calculation results showed that IICCA has higher calculation accuracy and faster convergence rate than the immune clonal selection algorithm,immune genetic algorithm,adaptive particle swarm optimization algorithm and improved cuckoo algorithm do;the curve similarity between identification calculation results and experimental results is as high as 97.02% to reveal the effectiveness and correctness of IICCA in solving the problem of multi-parameter identification of soft recoil artillery launch process.

关 键 词：软后坐技术 伯努利方程 多参数辨识 免疫克隆选择 高频变异 疫苗接种 莱维飞行 巢寄生行为 

分 类 号：TJ303.4[兵器科学与技术—火炮、自动武器与弹药工程]