矿用电动机转子瞬时功率分析与故障诊断方法的研究

发布时间：2018-03-31 05:02

本文选题：破碎机电动机　切入点：瞬时功率　出处：《西安科技大学》2017年硕士论文

【摘要】：在煤矿产业中,电动机作为驱动装置被广泛应用于破碎机、筛分机、刮板机、掘进机等大型设备中,因此,对矿用破碎机电动机的运行状态和故障类型进行及时检修是非常重要的,不仅关乎整个煤矿开采系统的正常安全运行,更加关乎煤炭工作人员的人身安全。本文以矿用破碎机电动机故障中最为常见的电动机转子断条和偏心故障为例。当矿用电动机转子产生故障时,在其定子两端的线电流中会产生相对应的故障特征频率,将特征频率变化反映到转子电流特征量变化中,但由于转子电流中产生的故障特征频率会因矿用电动机的转差率较小,其故障特征值被基波频率淹没。因此,求取矿用电动机其转子两端的线电流对应的线电压,进而求取矿用电动机转子的瞬时功率,对矿用电动机转子瞬时功率的率频谱图进行小波包分解、重构,从而对不同频段的频率特征量能量值进行提取,作为识别破碎机电动机转子故障的特征量。本文在所提取的不同频段矿用破碎机电动机转子故障特征量中,有些故障特征量可能存在冗余、重复、不确定情况,这些特征量不仅影响诊断速度,更会降低故障诊断的准确性。所以,在对数据样本进行故障诊断分类之前应先利用粗糙集理论(RS)对特征量样本进行预处理,从而减冗余数据,获得最终数据样本。由于矿用破碎机电动机恶劣的工作环境,使数据不能大量提取,又为了提高故障诊断率而引入的粗糙集算法对特征量样本进行预处理后,会使样本数据减少,因此,本文应用对小样本、非线性数据进行较好地诊断分类的支持向量机(SVM)算法,实现对矿用电动机转子故障的诊断与分类,使误判率降低。仿真实验结果表明,经粗糙集处理后的故障特征量的诊断结果更加准确。
[Abstract]:In coal mine industry, motor as driving device is widely used in crusher, sieve machine, scraper, roadheader and other large equipment, therefore, It is very important to examine and repair the running state and fault type of the motor of mine crusher in time, which is not only related to the normal and safe operation of the whole coal mining system, This paper takes the breakage and eccentricity faults of motor rotor, which is the most common fault of mine crusher, as an example. The corresponding fault characteristic frequency will be produced in the line current at the two ends of the stator, and the change of the characteristic frequency will be reflected in the change of the rotor current characteristic quantity. However, the fault characteristic frequency generated in the rotor current will be due to the small slip rate of the mine motor. Therefore, the linear voltage corresponding to the line current at the two ends of the rotor of the mine motor is obtained, and then the instantaneous power of the rotor of the mine motor is obtained. The frequency spectrum of instantaneous power of mine motor rotor is decomposed and reconstructed by wavelet packet, and the energy value of frequency characteristic quantity is extracted from different frequency range. As the characteristic quantity to identify the rotor fault of the crusher motor, in this paper, some fault characteristics may exist redundancy, repetition and uncertainty in the different frequency range of the rotor fault characteristic of the mine crusher motor. These features not only affect the speed of diagnosis, but also reduce the accuracy of fault diagnosis. Therefore, the rough set theory should be used to preprocess the feature samples before classifying them, so as to reduce the redundant data. The final data sample is obtained. Because of the bad working environment of the motor of the crusher, the data can not be extracted in large quantities, and the rough set algorithm introduced in order to improve the fault diagnosis rate is used to preprocess the feature sample. Therefore, this paper applies the support vector machine (SVM) algorithm to diagnose and classify the rotor faults of mine motor by using the support vector machine (SVM) algorithm for the diagnosis and classification of small sample and nonlinear data. The simulation results show that the diagnosis results of the fault feature quantity treated by rough set are more accurate.
【学位授予单位】：西安科技大学
【学位级别】：硕士
【学位授予年份】：2017
【分类号】：TD607

【参考文献】