基于复合智能算法的电机故障检测研究

Research on motor fault detection based on complex intelligent algorithm

传统智能故障检测模型中神经网络存在泛化能力弱,易陷入局部极小值、缺乏自学习、自组织能力、算法单一等缺点。组合应用智能检测算法可整合不同算法优势,避免单一算法缺点,为此,文中提出支持向量机算法与粒子群算法相结合的电机故障检测模型:以电机故障特征频率特征数据为基础,首先应用启发性较好的粒子群算法求解影响支持向量机分类检测性能的最佳参数,然后把最佳参数应用于的擅长模式识别的支持向量机算法,进行样本数据的训练,构建故障检测模型;最后,使用故障检测模型对电机的状态进行预测。实验结果表明,采用该方法进行故障检测的准确率,比传统的神经网络方法提高17%,比纯支持向量机算法提高3.33%。

Traditional intelligent fault detection model such as neural network has some faults. For example: generalization ability is weak, easy to fall into local minimum value, the lack of self-learning and self-organization, the algorithm is single. Combination application of intelligent detection algorithm can integrate different algorithms advantages and avoid the disadvantage of single algorithm. Therefore, this paper propose a mine based motor fault detection model based on combination of vector machine(SVM)algorithm and particle swarm optimization(PSO) algorithm. On the basis of motor fault characteristic frequency characteristic data,firstly, application of inspiring good influence particle swarm algorithm support vector machine(SVM) Particle swarm algorithm is used to solve the optimal parameters for SVM, and then apply the optimal parameters to SVM algorithm to train sample data, at last,a fault diagnosis model has built up; The experimental results show that the method adopted to improve the accuracy of fault detection by 3.33%-17%.