为了便于接地网导体腐蚀故障诊断，建议在设计接地网时，在一些网络节点增设监测线，并研究了一种基于遗传算法的接地网监测线优化布置方案生成方法。该方法由对子优化过程的多次迭代组成，能够达到以最少的监测线使所有支路都明晰的目的。论述了染色体编码方法和能避免不可行解的初始种群生成方法、交叉方法和变异方法。得到至少一组或多组接地网腐蚀故障监测线优化布置方案。 考虑到干扰的影响，采用了一种基于蒙特卡罗的接地网监测线布置方案评估方法，计算每组优化布置方案下，接地网的平均相对累积误差均值和标准差。利用F检验判断出抗干扰性最好的一组方案。此外，为了解决监测线的断线对接地网可测性的影响问题，采用了M-X+Y系列健壮性原则进行方案遴选。 基于MATLAB平台研制出接地网监测线优化布置通用应用软件，并详细论述了软件流程、数据结构和关键技术的实现方法。以一些接地网为例，对研制出接地网监测线优化布置通用应用软件进行了测试，结果表明该软件达到了预期目标。 将上述技术推广应用到已投运接地网故障诊断中，利用遗传算法确定需要补充的可及节点的位置，以便使该接地网的全部支路都能得到准确诊断，应用MATLAB平台编写了通用程序。以一个60支路的实验接地网为例说明了这种方法的可行性。

To benefit grounding grids corrosion diagnosis, monitoring lines are suggested to implant on some nodes of the grounding grid in the design process. A genetic algorithm based method of layout optimization of monitoring lines is used, which consists of some iteration of a sub-process. The optimum layout of making all of the branches clear with the minimum monitoring lines can be worked out. The coding of chromosomes is described. The ways to avoid of the unfeasible solutions in the processes of the initial population forming, crossover and mutation operation are discussed. Consequently, a set of optimal layouts of monitoring lines are obtained. Considering the effect of disturbance, an approach based on Monte-Carlo method is presented. The averaged value and its standard deviation of relative accumulated error of grounding grids based on each optimization layout scheme are calculated. The scheme with the best anti-disturbance performance is selected by using F test. In addition, the M-X+Y series indexes are used to find out the most robust scheme. The application software of the layout optimization of monitoring lines is programmed on the platform of MATLAB. The block diagrams, the data structures and the critical technologies are discussed. Some grounding grids are used as examples to test the application software. The results show that the software is feasible and correct. The technology has been extended and applied to diagnose the existing grounding grids. A genetic algorithm based method to determine the positions of digging and adding touchable nodes to make all branches clear is used. The application software is developed on the platform of MATLAB. A grounding grid with sixty branches is used as an example to show the feasibility of the proposed approach.