基于振动分析的高压断路器机械故障诊断研究

发布时间：2018-11-18 08:26

【摘要】：高压断路器在电力系统中起着极为重要的作用,它承担着系统的控制和保护工作。在正常运行情况下,高压断路器控制线路和设备的投切操作;当发生故障时,则快速切断故障线路,以防止故障的进一步扩大。因此一旦高压断路器发生故障,将直接危害系统的运行可靠性,并可能引起重大的停电损失。国内外相关调查表明,断路器大部分故障都是由机械原因造成的,因此开展高压断路器机械故障诊断的研究具有重要的现实意义。高压断路器分合闸动作期间产生的振动信号包含了重要的状态信息,本文通过对断路器振动信号进行处理和分析,来获取高压断路器的机械状态情况。其主要研究内容包括信号采集、特征提取和状态识别三个部分。首先,选择合理的振动传感器和信号采集设备,设计并搭建断路器振动信号采集系统平台,采集高压断路器在不同机械状态下振动信号。其次,针对高压断路器振动信号的非平稳、非线性特点,采用变分模态分解对断路器振动信号进行时-频分解处理,得到相应的本征模态函数,然后将得到的模态函数矩阵划分为若干子矩阵以计算矩阵局部奇异值,并选择各个子矩阵的最大奇异值作为故障诊断的特征向量。最后,构建单类分类器和多类分类器联合的多层分类器,采用两层独立的单类支持向量机分别用于准确区分高压断路器正常与故障状态、已知故障类型和未知故障类型,在此基础上,进一步采用支持向量机识别已知故障的具体故障类型。对SF6高压断路器在正常和三种典型故障下开展实例诊断测试,实验结果表明,采用变分模态分解和局部奇异值分解方法能够准确提取断路器故障特征,而将单类分类器应用到高压断路器机械故障诊断中能够有效提高故障识别的准确率,从而提高了故障诊断的可靠性,具有较高的工程应用价值。
[Abstract]:High-voltage circuit breaker plays an important role in the power system, it is responsible for the control and protection of the system. Under the normal operation condition, the high voltage circuit breaker controls the switching operation of the circuit and equipment; when the fault occurs, the fault line is cut off quickly to prevent the further expansion of the fault. Therefore, once the high voltage circuit breaker breaks down, it will directly endanger the reliability of the system, and may cause significant loss of power. The investigation at home and abroad shows that most of the faults of circuit breakers are caused by mechanical reasons, so it is of great practical significance to study the mechanical fault diagnosis of high voltage circuit breakers. The vibration signal generated during the switching operation of high voltage circuit breakers contains important state information. In this paper, the mechanical state of high voltage circuit breakers is obtained by processing and analyzing the vibration signals of circuit breakers. The main research contents include three parts: signal acquisition, feature extraction and state recognition. First of all, select reasonable vibration sensor and signal acquisition equipment, design and build the circuit breaker vibration signal acquisition system platform, collect high-voltage circuit breaker vibration signals in different mechanical states. Secondly, in view of the non-stationary and nonlinear characteristics of the vibration signals of high voltage circuit breakers, the variational mode decomposition is used to process the vibration signals of the circuit breakers by time-frequency decomposition, and the corresponding eigenmode functions are obtained. Then the modal function matrix is divided into several submatrices to calculate the local singular value of the matrix and the maximum singular value of each submatrix is selected as the eigenvector of fault diagnosis. Finally, a multi-layer classifier combined with single-class classifier and multi-class classifier is constructed. Two independent single-class support vector machines are used to accurately distinguish the normal and fault states of HV circuit breakers, known fault types and unknown fault types, respectively. On this basis, support vector machine (SVM) is used to identify the specific fault types of known faults. An example diagnosis test of SF6 high voltage circuit breaker under normal and three typical faults is carried out. The experimental results show that variational mode decomposition and local singular value decomposition can accurately extract the fault characteristics of the circuit breaker. The application of single-class classifier to mechanical fault diagnosis of high voltage circuit breakers can effectively improve the accuracy of fault identification, thus improving the reliability of fault diagnosis, and has a higher engineering application value.
【学位授予单位】：东北电力大学
【学位级别】：硕士
【学位授予年份】：2017
【分类号】：TM561

【参考文献】