基于多层降维与FOA优化的接地网腐蚀故障诊断

Corrosion Fault Diagnosis of Substation Grounding Grid based on Multiple Dimension Reduction and Fruit Fly Optimization Algorithm

针对变电站接地网腐蚀故障的精准定位与诊断问题,将果蝇优化算法运算速度快、全局收敛效果好的优点与电阻腐蚀倍数优化相结合,提出基于多层降维与果蝇优化的接地网腐蚀故障诊断方法。该方法以电压法与电阻法作为诊断机理,以果蝇优化算法作最优解全局搜索,结合多层降维解决接地网支路个数多的问题,降低优化复杂度,提高优化速度。以60条支路、36个节点的变电站接地网为例进行方法检验对比,结果表明,以果蝇优化算法为基础的电压法和电阻法均能取得较好的腐蚀故障诊断效果,电阻法的平均诊断误差略低于电压法,诊断效果较好,但时间代价更大。

Aiming at the precise positioning of corrosion failures and its diagnosis of grounding grid,the fruit fly optimization algorithm, which is of fast calculation and the overall convergence effect, is combined with the good corrosion resistance multiple optimization, and a corrosion fault diagnosis method of substation grid is proposed based on multiple dimension reduction and fruit fly optimization algorithm, which takes the voltage and resistance as the diagnosis mechanism, and makes the overall search based on the optimized result of the fly optimization algorithm, combines with multiple dimension reduction to solve the problem of great numbers of the grounding grid branches. Then, the complexity of optimization is simplified and the speed of optimization is increased. Taking 60 branches and 36 nodes of a substation grid as example, the presented method is verified and the results show that Drosophila voltage optimization algorithm-based method and the resistance method can achieve better corrosion fault diagnosis, the average diagnostic error of the resistance method is slightly lower than the voltage method, but more time costs.