摘要
针对传统的自适应随机共振以单个参数为优化对象忽略参数间交互作用的不足及采用遗传算法优化参数在种群数量增加时算法收敛速度明显减缓的缺陷,提出基于人工鱼群算法的自适应随机共振新方法。该方法利用人工鱼群算法对初值、参数设定容许范围较大、具备并行处理能力及人工鱼个体数目增加时鱼群算法收敛速度能提高的特性,自适应实现与输入信号最佳匹配的随机共振系统。仿真数据与轴承滚动体故障数据分析表明,基于该算法的自适应随机共振方法可有效实现微弱特征检测与早期故障诊断。
Based on analyzing the disadvantages of traditional detection method of adaptive stochastic resonance, for example,optimizing only one parameter while ignoring the interaction between parameters and the convergence speed of genetic algorithm slowing down with the increase of population,a new adaptive stochastic resonance method was proposed.With a wide range of setting initial values and parameters,the proposed method adaptively realizes the optimal stochastic resonance system to match input signals,by virtue of its ability of parallel processing and its characteristic that the algorithm has faster convergence speed with the increase of artificial fish number.The analysis of the simulation data and the bearing fault data shows that the new adaptive stochastic resonance method can effectively realize the weak signal detection and early fault diagnosis.
出处
《振动与冲击》
EI
CSCD
北大核心
2014年第6期143-147,共5页
Journal of Vibration and Shock
基金
国家自然科学基金(10972207)
浙江省自然基金(Y7080111)
关键词
自适应随机共振
人工鱼群算法
参数优化
轴承故障诊断
adaptive stochastic resonance
artificial fish swarm algorithm
parameter optimization
bearing fault di-agnosis