基于人工蜂群算法优化支持向量机的接地网腐蚀速率预测模型

为了提高接地网腐蚀速率预测的精确度,在建立预测模型的过程中,首先对接地网进行了基于电网络理论的腐蚀诊断过程,并以经过诊断之后确定的腐蚀支路位置为采样点。考虑到仅以土壤理化性质反映接地网腐蚀速率的局限性,在接地网腐蚀诊断结果的基础上,提出接地网电阻平均增长速率作为预测模型的输入特征量之一。建立了基于人工蜂群优化支持向量机的接地网腐蚀速率预测模型,测试结果显示相对BP神经网络模型和广义回归神经网络模型,所提模型的预测结果精确度和稳定性更高,表明了对于解决接地网腐蚀速率预测问题,所提模型具有良好的适用性。

Corrosion rate prediction model of grounding grid based on support vector machine optimized by artificial bee colony algorithm

In order to improve the accuracy of corrosion rate prediction for grounding grid, firstly the corrosion diagnosis of grounding grid based on the theory of electric network is carried out, and the position of corrosion branches after diagnosis are taken as sampling points. Considering the limitation of reflecting the corrosion rate prediction of grounding grid only by soil physical and chemical properties, and based on the result of corrosion diagnosis, the ave-rage growth rate of resistance in grounding grid is proposed as one of the input characteristics of the prediction mo-del. Then the corrosion rate prediction model of grounding grid based on the support vector machine optimized by artificial bee colony algorithm is proposed. The test results show that compared with the BP neural network model and generalized regression neural network model, the proposed model has higher prediction precision and stability, and good applicability to solve the problem of corrosion rate prediction for grounding grid.