Corrosion Rate Prediction of Grounding Network Based on Improved Least Square Support Vector Machine

Prediction of the corrosion rate of the grounding grid is an important measure to ensure the safe operation of the power system. Aiming at the characteristics of ground network detection data with few samples and non-linearity, which can not use a single prediction algorithm to achieve accurate prediction, a gray wolf algorithm optimized least square support vector machine combined with error correction for ground network corrosion rate prediction model was proposed. This method first uses the gray wolf algorithm to optimize the least squares support vector machine. At the same time, an error correction model is introduced to further correct the predicted data, thereby improving the prediction accuracy and ensuring the accuracy of the prediction data. The results show that the combined prediction model of the least squares support vector machine improved by the gray wolf algorithm and the error correction is used to predict the corrosion rate of the ground network, which is better than the least squares support vector improved by the least squares support vector machine model or the gray wolf algorithm. The prediction error of the machine model is smaller, the accuracy is higher, and the prediction of the corrosion rate of the grounding grid is more effective.