Random forest grid fault prediction based on genetic algorithm optimization

The operation of the power grid is closely related to meteorological disasters. Changes in meteorological conditions may have an impact on the operation and stability of the power system, leading to economic losses. This paper proposes a Random Forest grid fault prediction model based on Genetic Algorithm optimization (GA-RF) to classify the grid fault types, which improves the distribution network fault prediction accuracy by constructing an optimized random forest model. Specifically, the model’s performance is initially enhanced by calculating the Gini index for each feature. The weather attributes with higher Gini indices are subsequently selected as pivotal features to alleviate the detrimental impact of unnecessary attributes on the model. In addition, a genetic algorithm is used to optimize the parameters of the random forest model for early warning of grid fault occurrence. The experimental results demonstrate that the proposed GA-RF in this paper achieves significantly higher accuracy compared to Random Forest (RF), Support Vector Machine (SVM), and Linear Regression (LR). Specifically, it outperforms them by 14.77%, 23.22%, and 13.77% respectively. This method effectively supports the safe and stable operation of the power system.