推荐星级:
  • 1
  • 2
  • 3
  • 4
  • 5

基于ALO-SVM的模拟电路故障诊断研究

更新时间:2020-10-25 18:54:49 大小:2M 上传用户:gsy幸运查看TA发布的资源 标签:模拟电路 下载积分:2分 评价赚积分 (如何评价?) 打赏 收藏 评论(0) 举报

资料介绍

针对模拟电路故障定位难的问题,采用一种新兴的群体智能优化算法--蚁狮算法(ALO),来优化支持向量机(SVM)的惩罚因子和核函数,建立一种基于蚁狮算法优化支持向量机的模拟电路故障诊断模型。ALO算法采用了自适应边界收缩机制、精英主义、随机游走和轮盘选择机制等关键技术以提高收敛速度,避免陷入局部最优解。为验证诊断模型的有效性,仿真某实际电路并用小波变换提取电路的故障特征向量,分别用蚁狮算法、蝙蝠算法、粒子群算法、遗传算法优化SVM对电路的故障进行诊断。经过比较,蚁狮算法不仅大大缩短了模型的训练时间,而且诊断准确度也有明显提升。

For the difficulty of faults location on analog circuits, a new swarm intelligence optimization algorithm, the ant lion optimize algorithm(ALO), is used to optimize the penalty factor and kernel function of support vector machine(SVM), and a fault diagnosis model of analog circuit is established based on the ant lion optimized-support vector machine(ALOSVM). The ALO can improve the convergence speed and avoid the local optimal solution, which uses of adaptive boundary contraction mechanism, elitism, random walk and wheel selection mechanism, etc. Besides, in order to verify the validity of the diagnostic model, a practical circuit is simulated and the fault feature vectors of the circuit are extracted by wavelet transform, the ALO, bat algorithm, particle swarm optimization and genetic algorithm are used to optimize the SVM to diagnose the fault of circuit. After comparison of those algorithms, the results show that the ALO has a good performance on reducing the training time of model and improving the accuracy of diagnostic.

部分文件列表

文件名 大小
基于ALO-SVM的模拟电路故障诊断研究.pdf 2M

【关注B站账户领20积分】

全部评论(0)

暂无评论

上传资源 上传优质资源有赏金

  • 打赏
  • 30日榜单

推荐下载