电力变压器局部放电超声波检测方法研究

发布时间：2018-07-03 01:00

本文选题：局部放电 + 超声波　；参考：《沈阳工业大学》2017年硕士论文

【摘要】：伴随着我国坚强智能电网的建设与发展,电力系统规模的日益扩大,特别是特高压交直流输电的建成和运营,对电气设备的要求越来越高,因此,变压器局部放电在线检测技术引起了电力行业的高度重视。目前电力变压器绝缘状态评估的方法,主要是根据绝缘介质在劣化过程中的一些特征参数进行评定,包括介质损耗角的正切值、泄漏电流、局部放电量等。其中局部放电量最能反映变压器绝缘状态,也是变压器内绝缘劣化的重要原因。因此,开展电力变压器局部放电检测方法研究具有学术意义和实用价值。本文主要研究内容如下:(1)从气体放电理论出发,讨论了局部放电过程及超声波的产生机理,并以变压器内部气泡放电为例,分析气泡击穿时的受力过程,利用电力类比的原理,建立了基于气泡模型的局部放电产生超声波的力学等值电路模型;并以此为研究基础对超声波信号在模拟变压器油箱内部传播规律进行仿真研究和实验结果对比研究,研究主要包括超声波在变压器油、油浸绝缘纸和变压器壁中的传播速度规律,超声波信号通过油、纸绝缘介质时幅值的衰减。(2)针对在运行环境下变压器局部放电信号易受电磁干扰问题,提出了一种分层抑制干扰的联合检测法,即以带通滤波器和小波分析为核心。通过分析局部放电信号和扰动信号频率分布特点,并借助MATLAB仿真软件对变压器内部放电信号进行仿真研究。结果表明:采用分层式抑制干扰的方式可以有效去除干扰并保证原始信号能量损失较小,大大提高了局部放电的定位精度,满足了变压器局部放电在线检测的实际要求。(3)针对传统局部放电定位的最小二乘法,在迭代过程中容易陷入局部收敛且在变压器某些特殊区域无法定位问题。采用了一种具有全局搜索特性的遗传算法,并将其定位结果与最小二乘法进行比较。计算结果表明遗传算法解决了最小二乘法在边角区域定位失败问题,其定位效果客观。因此,具有实用价值。
[Abstract]:With the construction and development of strong smart grid in China, the scale of power system is expanding day by day, especially the construction and operation of UHV AC / DC transmission. Transformer partial discharge online detection technology has attracted great attention in power industry. At present, the methods of evaluating the insulation state of power transformers are mainly based on some characteristic parameters of dielectric deterioration, including tangent value of dielectric loss angle, leakage current, partial discharge, etc. Partial discharge (PD) can reflect the insulation state of transformer and is the important reason of insulation deterioration in transformer. Therefore, it is of academic significance and practical value to study the detection method of partial discharge of power transformer. The main contents of this paper are as follows: (1) based on the theory of gas discharge, the process of partial discharge and the generation mechanism of ultrasonic wave are discussed. Based on the principle of electric force analogy, a mechanical equivalent circuit model of ultrasonic wave generated by partial discharge based on bubble model is established. On the basis of this research, the simulation and experimental results of the ultrasonic signal propagation in the simulated transformer tank are carried out. The research mainly includes the ultrasonic wave in the transformer oil. The law of propagation velocity in oil-immersed insulator paper and transformer wall, and the attenuation of ultrasonic signal through oil and paper insulating medium. (2) in order to solve the problem that transformer partial discharge signal is vulnerable to electromagnetic interference in operating environment, In this paper, a hierarchical joint detection method for interference suppression is proposed, which is based on band-pass filter and wavelet analysis. The characteristics of frequency distribution of partial discharge signal and disturbance signal are analyzed, and the internal discharge signal of transformer is simulated by MATLAB simulation software. The results show that the interference can be removed effectively and the energy loss of the original signal can be reduced by using the method of delamination, which greatly improves the positioning accuracy of partial discharge. It meets the practical requirements of on-line detection of partial discharge of transformers. (3) for the traditional least square method of partial discharge location, it is easy to fall into local convergence in the iterative process and can not locate in some special areas of the transformer. In this paper, a genetic algorithm with global search property is used, and its localization results are compared with the least square method. The results show that the genetic algorithm solves the problem of the location failure of the least square method in the corner region, and its localization effect is objective. Therefore, it has practical value.
【学位授予单位】：沈阳工业大学
【学位级别】：硕士
【学位授予年份】：2017
【分类号】：TM855

【参考文献】