摘要
为提高电机故障诊断的准确率和有效性,提出了基于智能优化算法的支持向量机电机故障诊断模型.首先采集交流电机不同位置上的振动加速度信号,使用小波包分析方法对所采集的振动加速度信号进行特征提取,将得到的能量比向量作为支持向量机故障诊断模型的输入,使用遗传算法、粒子群优化算法对支持向量机故障诊断模型进行参数优化并进行模型训练,在使用测试样本集对得到的两种故障诊断模型进行分析之后可以看出经过参数优化后的支持向量机模型提高了故障预测的准确率,并且粒子群优化方法具有比遗传算法更高的预测准确率,并极大地减小了优化时间及优化次数.
In order to improve the accuracy and efficiency of motor fault diagnosis,a new method of support vector machine was proposed based on intelligent optimization algorithms. After collecting AC motor vibration acceleration signals at different positions,the feature extraction with wavelet packet analysis method was conducted,and power ratio of the result of feature extraction were used as the input to the motor fault diagnosis model for support vector machine. The genetic algorithm and the particle swarm optimization algorithm can be used for parameter optimization and model training of the motor fault diagnosis model for the support vector machine. Fault diagnosis model analysis shows that after parameter optimization,the fault diagnosis model improves the forecasting accuracy,and the particle swarm optimization algorithm has higher prediction accuracy than the genetic algorithm with greatly reduced time and times of optimization.
出处
《大连交通大学学报》
CAS
2016年第1期92-96,共5页
Journal of Dalian Jiaotong University
基金
国家自然科学基金资助项目(51475065)
人工智能四川省重点实验室基金资助项目(2014RYJ01)
过程装备与控制工程四川省高校重点实验室开放基金资助项目(2014RYJ01)
关键词
支持向量机
粒子群优化算法
遗传算法
电机故障诊断
support vector machine
particle swarm optimization algorithm
genetic algorithm
motor fault diagnosis