【作者】 王刚；

【导师】 高厚磊； 苏国强；

【作者基本信息】 山东大学 ， 电气工程（专业学位）， 2024， 硕士

【摘要】 在我国10 kV电压等级的大中城市配电网中,中性点经小电阻接地方式因其过电压水平低、故障切除速度快、对设备的绝缘要求低等优势被广泛采用,而在发生单相接地故障时其接地保护也面临一些亟待解决的痛点问题。其中,配电线路发生经树枝等高阻介质接地的可能性较大,而现有零序过电流保护无法有效处理此类高阻接地故障。此外,当分布式电源(distributed generation,DG)采用小电阻接地方式并网时,原有的接地故障特征发生改变并导致适用于传统小电阻配电网的保护失效。为解决上述问题,本文立足工程应用,主要完成了以下研究工作:(1)完成传统小电阻接地配电网(下文均简称为传统小电阻配电网)及有源小电阻接地配电网(下文均简称为有源小电阻配电网)的单相接地故障建模、故障特征分析。通过建立传统小电阻配电网的单相接地故障序网络,推导得出各条线路出口处的零序电流表达式,并从幅值和相位两个方面分析了零序电流的故障特征;讨论DG以不同的中性点接地方式并网对接地故障零序网络、故障特征及保护方法的影响;针对DG采用小电阻接地方式并网的情形,对有源小电阻配电网进行单相接地故障建模、零序电流表达式推导,并基于数学解析和软件模拟的方法分析了零序电流的故障特征。(2)提出余弦距离结合改进K均值算法的传统小电阻配电网接地保护方法。为兼顾保护的可靠性与灵敏性,同时免疫噪声等因素对保护整定值的影响,所提保护方法利用余弦距离来刻画零序电流之间的相位关系并形成表征故障特征的余弦距离矩阵,通过改进K均值算法对该矩阵进行分析,在不用设置整定值的情况下自动将线路分为健全和故障线路两类,避开了高阻接地故障时整定困难的问题。该保护方法无需电压信息,所需故障数据窗长较短,保护动作速度快。(3)提出基于电容电流补偿的传统小电阻配电网接地故障后备保护方法。根据零序电流之间的幅值倍数关系,所提保护方法利用母线电压、线路长度和零序电容信息对零序电流的测量值进行补偿并求得补偿后、前的零序电流比值,据此有效区分故障线路和健全线路,可作为现有零序过电流保护的后备保护。在低阻接地时,现有保护动作;在高阻接地时,现有保护拒动,由所提后备保护切除故障。该保护方法对线路参数的精度要求不高,无需通信手段,具有较强的耐过渡电阻能力。(4)提出基于复合零序导纳的有源小电阻配电网接地故障保护方法。该方法定义了复合零序导纳这一特征量,通过数学解析和软件模拟相结合的手段,发现故障线路、含DG的健全线路、不含DG的健全线路的复合零序导纳处于不同的范围,差异明显且与过渡电阻基本无关,据此构造判据,保证了保护的可靠性和灵敏性。该保护方法耐过渡电阻能力强,无需通信手段,同时也适应DG高比例接入的场景。(5)提出基于余弦距离和幅值比的有源小电阻配电网接地故障保护方案。根据零序电流的分布特点及其相位关系,对于相邻中性点之间的线路,采取基于线路两端零序电流之间的余弦距离的纵联保护方法;对于不含DG的线路,采取基于线路出口处与主网侧中性点处零序电流之间的余弦距离的保护方法;对于含DG线路上最末端DG并网点的下游线路,采取基于线路出口处与相邻的DG侧中性点处零序电流的幅值比的保护方法。该保护方案兼具可靠性与灵敏性,无需电压信息,耐过渡电阻能力较强,同时提高了含DG接入线路的供电可靠性。(6)以上故障特征及保护方法均经过了理论推导与仿真验证。通过在MATLAB/SIMULINK和PSCAD/EMTDC平台上建立的仿真模型,在不同过渡电阻、电弧接地故障、噪声干扰等情况下对上述保护方法进行了性能验证与分析,证明了四种保护方法均具有良好的应用前景。更多还原

【Abstract】 In the distribution network with a 10 kV voltage level of large and medium-sized cities in China,the neutral point grounding method through low resistance is widely adopted due to its advantages such as low overvoltage level,fast fault removal speed,and low insulation requirements for equipments.In the event of a single-phase grounding fault,its grounding protection also faces some urgent pain points that need to be solved.Among them,there is a high possibility of distribution lines being grounded through high resistance media such as tree branches,and the existing zero-sequence overcurrent protection cannot effectively handle such high resistance grounding faults.In addition,when distributed generation(hereinafter referred to as DG)which adopts low resistance grounding method is connected to the grid,the original grounding fault characteristics change,leading to the failure of protection applicable to traditional low resistance distribution networks.To solve the above problems,this paper focuses on engineering applications and mainly completes the following research work:(1)Complete single-phase grounding fault modeling and fault characteristic analysis for the traditional low resistance grounded distribution network(hereinafter referred to as TLRGDN)and active low resistance grounded distribution network(hereinafter referred to as ALRGDN).By establishing a single-phase grounding fault sequence network for TLRGDN,the expression of zero-sequence current at the outlet of each line is derived.The fault characteristics of zero-sequence current are analyzed from the perspectives of amplitude and phase.Then discuss the influence of different neutral grounding methods of DG on the zero-sequence network.fault characteristics,and protection methods when connected to the grid.In response to the situation where DG adopts low resistance grounding method for grid connection,single-phase grounding fault modeling as well as zero-sequence current expression derivation are carried out for ALRGDN,and based on mathematical analysis and software simulation methods,fault characteristics are analyzed.(2)Propose a grounding protection method for TLRGDN using cosine distance combined with improved K-means algorithm.In order to balance the reliability and sensitivity of protection while being immune to the influence of noise and other factors on the protection setting value,cosine distance is used to characterize the phase relationship between zero-sequence currents and form a cosine distance matrix that characterizes the fault characteristics.Then analyze the matrix by the improved K-means algorithm,and the line is automatically divided into two categories of non-fault and fault lines without the process to set protection setting value,avoiding the difficulty in setting value during high resistance grounding faults.This protection method does not require voltage information,requires a short fault data window,and has a fast protection action speed.(3)Propose a backup protection method based on capacity current compensation for TLRGDN.Based on the amplitude multiple relationship between zero-sequence currents,the measured values of zero-sequence currents are compensated using bus voltage,line length,and zero-sequence capacitance information,and the ratio of zero-sequence current before and after compensation is obtained.Based on this,fault lines and non-fault lines are effectively distinguished,serving as the backup protection for existing zero-sequence overcurrent protection.When grounded through low resistance,the existing protection operates;When grounded through high resistance,the existing protection refuses to operate and the fault is removed by the proposed backup protection.This protection method does not require high precision in line parameter accuracy,does not require communication means,and has strong ability to endure the fault resistance.(4)Propose a grounding fault protection method for ALRGDN based on composite zero-sequence admittance.This method defines the characteristic quantity of composite zero sequence admittance.Through a combination of mathematical analysis and software simulation,it is found that the composite zero-sequence admittance of fault lines,non-fault lines with DG,and non-fault lines without DG are in different ranges,with significant differences and are basically independent of fault resistance.The criteria is constructed based on this and ensures the reliability and sensitivity of protection.This protection method has strong ability to endure the fault resistance,does not require communication means,and is also suitable for scenarios with high proportion of DG access.(5)Propose an ALRGDN grounding protection scheme based on cosine distance and amplitude ratio.Based on the distribution characteristics of zero-sequence current and its phase relationship,for lines between adjacent neutral point,a pilot protection method based on the cosine distance between the zero-sequence currents at both ends is adopted.For lines without DG,a protection method based on the cosine distance between the zero-sequence current at the line outlet and the neutral point on the main grid side is adopted.For the downstream line of the final DG grid connection point,a protection method based on the amplitude ratio of zero sequence current at the line outlet and adjacent DG side neutral point is adopted.This protection scheme combines reliability and sensitivity,does not require voltage information,has strong ability to endure the fault resistance,and improves the power supply reliability of lines with DG access.(6)The above fault characteristics and protection methods have been theoretically derived and verified through simulation.Through the simulation models established on MATLAB/SIMULINK and PSCAD/EMTDC platforms,performance verification and analysis of the above protection methods were conducted under different fault resistances,arc grounding faults,noise interference,and other conditions,proving that all four protection methods have well application prospects.更多还原