【摘要】 为了简化双层结构土壤模型中地网接地电阻的计算,提出了将双层土壤模型等效为均匀土壤模型的计算方法。分析了接地网面积、接地网的长宽比、上层土壤电阻率等参数对等效均匀土壤电阻率的影响后,指出接地网面积、上层土壤的电阻率和厚度以及反射系数等参数均对等效均匀土壤电阻率有重要影响;采用CDEGS软件仿真所得数据建立求解等效均匀土壤电阻率的BP神经网络所得结果与CDEGS软件计算对比表明,该BP神经网络具有较高的准确性和可信度,可为多层土壤结构中接地网的设计提供可靠帮助。更多还原

【Abstract】 To calculate the grounding resistance of the grounding grid in the two-layer soil, a simplification method for a two-layer soil structure is proposed. The influences of grounding area, the shape of the grounding grid, top-layer soil resistivity, top-layer soil thickness and the reflective coefficient on the equivalent soil resistivity are analyzed. The results show that grounding area, top-layer soil resistivity, top-layer soil thickness and the reflective coefficient are the key parameters that influence the equivalent soil resistivity. The top-layer soil resistivity has a linear relationship with the equivalent soil resistivity, so the grounding area, top-layer soil thickness and the reflective coefficient are chosen to simplify the analysis. The grounding area, top-layer soil thickness and the reflective coefficient have nonlinear relationships with the equivalent soil resistivity. It is hard to describe the relationship by the conventional mathematic function, while the BP artificial neural network can describe the nonlinear relationship exactly. According to the data calculated by the CDEGS software, the system model of BP ANN to calculate the equivalent soil resistivity is discussed. Compared with the actual calculation results, the errors of grounding resistance is controlled within a reasonable range, this system model is reliable. This method is helpful to the design of the grounding system in the multilayer soil.更多还原