气体绝缘变电站接地网的冲击特性和地表电位计算分析

发布时间：2018-06-22 04:42

本文选题：接地网 + 暂态模型　；参考：《沈阳工业大学》2017年硕士论文

【摘要】：变电站的接地装置是由地网导体、接地引下线以及避雷器等部分组成的,它是发电厂、变电站的重要组成部分。其中地网是由水平导体和垂直导体组成的具有泄流和均压作用的网状接地装置。在系统发生故障时,能够起到迅速泄放电流、改善地网导体和变电站场区地表电位分布的作用,是电力系统安全运行的重要举措。近年来,我国电力系统快速发展,GIS变电站接地网的面积也在逐渐增大,当变电站发生故障时,通过接地体流入地网的冲击电流也在逐渐增加,造成地表电位分布不均或出现异常,给站内工作人员的安全带来威胁,使系统二次设备绝缘遭到破坏。因此,分析GIS变电站接地网的冲击特性以及地表电位分布情况是一个重要的课题。本文考虑冲击电流作用下地网导体的火花效应和电感效应,考虑波在地网导体上的传播过程,建立地网导体的暂态数学模型。首先,分析雷电流注入位置对地网导体各节点电压波形以及幅值的影响,分析雷电流注入位置对地网导体中电流流散的影响,分析雷电流波形对接地网各节点电压波峰值的影响。然后将仿真结果与国外参考文献里面的试验结果对比,来验证本文地网暂态模型的有效性。其次,以某800kV GIS变电站接地网实际工程为例,根据变电站和接地网的实际布置图,运用ANSYS建立接地网的三维有限元模型。计算变电站接地网在雷电流作用下的地表电位分布情况,然后分析了地网的埋设深度和土壤的电阻率对地表电位的影响规律。最后,计算了地网故障情况下的地表电位分布,结合BP神经网络的故障检测方法,实现根据地表电位分布进行地网故障的有效识别。结果表明:地网在内部注入雷电流时,注入点的电压波峰值相比于在边角注入时明显减小。地网导体上各节点的电压波峰值与它的波头时间成反比。雷电流沿导体向土壤中流散过程中,导体内的电流值逐渐减小,并且电流流散情况与注入点的位置有关。各节点的电压波峰值与注入的冲击电流的波头时间成反比。地表电位最大值出现在雷电流注入点,地网的埋设深度和土壤电阻率的改变会影响地表电位值的大小。地网故障点区域附近的地表电位值出现明显的陡降,并且地网上各点的地表电位值均会发生改变,结合BP神经网络故障识别方法能够实现对地网断裂故障的有效识别,有一定的工程实用价值。
[Abstract]:The grounding device of substation is composed of ground grid conductor, grounding lead line and arrester. It is an important part of power plant and substation. The earth net is a grid earthing device which consists of horizontal conductor and vertical conductor. It is an important measure for the safe operation of the power system to discharge the current rapidly and improve the surface potential distribution in the ground network conductor and substation field area when the system fails. In recent years, with the rapid development of power system in China, the area of grounding grid of GIS substation is increasing gradually. The surface potential distribution is uneven or abnormal, which threatens the safety of the staff in the station and destroys the insulation of the secondary equipment of the system. Therefore, it is an important task to analyze the impact characteristics and surface potential distribution of grounding grid in GIS substation. In this paper, considering the spark effect and inductance effect of the earth net conductor under the action of impulse current, and considering the propagation process of the wave on the earth net conductor, the transient mathematical model of the earth net conductor is established. Firstly, the influence of the lightning current injection position on the voltage waveform and amplitude of each node of the earth net conductor is analyzed, and the influence of the lightning current injection position on the current dispersion in the ground net conductor is analyzed. The effect of the peak voltage wave of each node in ground network is analyzed. Then the simulation results are compared with the experimental results in foreign references to verify the effectiveness of the proposed transient model. Secondly, taking the actual engineering of a 800 kV GIS substation grounding grid as an example, according to the actual layout of substation and grounding grid, the three-dimensional finite element model of grounding grid is established by using ANSYS. The distribution of ground potential of substation grounding grid under the action of lightning current is calculated, and then the influence of the depth of grounding grid and the resistivity of soil on the surface potential is analyzed. Finally, the surface potential distribution in the case of ground network fault is calculated, and the fault detection method of BP neural network is combined to realize the effective identification of the ground network fault according to the surface potential distribution. The results show that the peak value of the voltage wave at the injection point is obviously smaller than that at the edge angle when the lightning current is injected inside the ground net. The peak value of voltage wave at each node on the earth net conductor is inversely proportional to its wave head time. In the course of the lightning current flowing along the conductor to the soil, the current value in the conductor decreases gradually, and the current dispersion is related to the position of the injection point. The peak voltage wave of each node is inversely proportional to the wave head time of the injected impulse current. The maximum surface potential appears at the lightning current injection point. The change of the buried depth and the soil resistivity will affect the magnitude of the surface potential. The surface potential value in the area near the fault point of the ground network has a sharp drop, and the surface potential value of each point on the ground network will change. Combining with the BP neural network fault identification method, it can realize the effective recognition of the fault of the ground network fracture. It has certain engineering 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】：TM862;TM63

【相似文献】