混沌变步长萤火虫优化的随机共振微弱信号检测
|
摘要
针对传统随机共振只能完成单参数寻优及寻优能力差的问题,提出了混沌变步长萤火虫优化的随机共振微弱信号检测方法。该方法选用二阶Duffing振子随机共振系统作为研究对象,将随机共振问题转化为系统的多参数同步寻优问题,利用追尾行为的混沌变步长萤火虫优化算法寻找系统的最优参数,实现随机共振,检测出强噪声背景下的微弱周期信号。仿真结果表明,随着输入信号的信噪比越低,混沌变步长萤火虫优化算法寻优结果越好;在寻优结果上,混沌变步长萤火虫优化算法的随机共振明显优于量子粒子群优化算法,主要表现为输出信噪比提高了5.70 dB,相对于原始信号,信噪比提高了28.76 dB。
The traditional stochastic resonance can only realize one-parameter optimization, and the optimization ability is poor. A weak signal detection method for chaotic variable step glowworm swarm optimization stochastic resonance was proposed to solve the problem. The method selected two order Duffing stochastic resonance system as research subject, the problem of two order Duffing stochastic resonance was converted into the problem of multi-parameter optimization, then, by using the rear-end behavior chaotic variable step glowworm swarm optimization of the follow behavior, the optimal parameters of the system were found, the weak period signal in noisy background could be detected by achieving stochastic resonance. Simulation of input signals with different amplitude and noise intensity showed that the lower the SNR of input signal was better by using chaotic variable step glowworm swarm optimization of the follow behavior. And the method based on chaotic variable step glowworm swarm optimization of the follow behavior was obviously superior to the method based on quantum particle swarm optimization.
The traditional stochastic resonance can only realize one-parameter optimization, and the optimization ability is poor. A weak signal detection method for chaotic variable step glowworm swarm optimization stochastic resonance was proposed to solve the problem. The method selected two order Duffing stochastic resonance system as research subject, the problem of two order Duffing stochastic resonance was converted into the problem of multi-parameter optimization, then, by using the rear-end behavior chaotic variable step glowworm swarm optimization of the follow behavior, the optimal parameters of the system were found, the weak period signal in noisy background could be detected by achieving stochastic resonance. Simulation of input signals with different amplitude and noise intensity showed that the lower the SNR of input signal was better by using chaotic variable step glowworm swarm optimization of the follow behavior. And the method based on chaotic variable step glowworm swarm optimization of the follow behavior was obviously superior to the method based on quantum particle swarm optimization.
引文
[1]冷永刚,赖志慧.基于Kramers逃逸速率的Duffng振子广义调参随机共振研究[J].物理学报,2014,63(02):38-40.
[2]Jung P.Periodically driver stochastic systems [J].Physics Reports, 1993,234(4):175-295.
[3]Gomes I, Mirasso C R, Toral R, et al.Experimental study of high frequency stochastic resonance in Chua circuits [J].Physica A: Statistical Mechanics and its Applications, 2003,327(1):115-119.
[4]行鸿彦,卢春霞,张强.自适应随机共振微弱信号检测[J].系统仿真学报,2018,30(02):587-592.
[5]王志霞,郭利.改进PSO算法调参的随机共振微弱信号检测[J].计算机测量与控制,2018,26(01):42-45.
[6]张勇亮,李国林,尹洪伟.基于tsPSO算法的阵列自适应随机共振方法研究[J].机械强度,2017,39(06):1288.
[7]孔德阳,彭华,马金全.基于人工鱼群算法的自适应随机共振方法研究[J].电子学报,2017, 45(8):1864-1865.
[8]崔伟成,李伟,孟凡磊,等.基于果蝇优化算法的自适应随机共振轴承故障信号检测方法[J].振动与冲击,2016,35(10):96-99.
[9]黄咏梅,林敏.基于外差式随机共振的涡街频率检测方法[J].机械工程学报,2008, 44(4):138-141.
[2]Jung P.Periodically driver stochastic systems [J].Physics Reports, 1993,234(4):175-295.
[3]Gomes I, Mirasso C R, Toral R, et al.Experimental study of high frequency stochastic resonance in Chua circuits [J].Physica A: Statistical Mechanics and its Applications, 2003,327(1):115-119.
[4]行鸿彦,卢春霞,张强.自适应随机共振微弱信号检测[J].系统仿真学报,2018,30(02):587-592.
[5]王志霞,郭利.改进PSO算法调参的随机共振微弱信号检测[J].计算机测量与控制,2018,26(01):42-45.
[6]张勇亮,李国林,尹洪伟.基于tsPSO算法的阵列自适应随机共振方法研究[J].机械强度,2017,39(06):1288.
[7]孔德阳,彭华,马金全.基于人工鱼群算法的自适应随机共振方法研究[J].电子学报,2017, 45(8):1864-1865.
[8]崔伟成,李伟,孟凡磊,等.基于果蝇优化算法的自适应随机共振轴承故障信号检测方法[J].振动与冲击,2016,35(10):96-99.
[9]黄咏梅,林敏.基于外差式随机共振的涡街频率检测方法[J].机械工程学报,2008, 44(4):138-141.