Prediction of the Thermal Rating of Overhead Conductor Based on Empirical Mode Decomposition-Sample Entropy and Quantile Regression Forest

Prediction of the thermal rating of overhead conductor can help operators foresee and make full use of the current-carrying capacity of overhead lines, which is of great significance to the safe and economic operation of the power grid. In view of the characteristics of high volatility and randomness of the thermal rating of overhead conductors, based on the existing probability prediction methods, this paper proposes a probability prediction model based on empirical mode decomposition - sample entropy and quantile regression forest. This method first uses empirical mode decomposition technology to decompose the original thermal rating into a series of components with different characteristics and then uses sample entropy theory to stack the approximate components to reduce the task load. After that, the corresponding quantile regression forest model is established for each reconstructed subsequence to predict. Finally, the predicted results of each subsequence are superimposed to obtain the final thermal rating predicted results. The prediction simulation results show that by reducing the complexity of the original sequence and modeling each new sequence separately, the prediction results can be significantly improved, which verifies the effectiveness of the proposed model.