Grid-connected Debugging Method of Digital Power Grid Based on LR Model

As an important data analysis method, LR has been widely used in many fields. In applications to practical classification problems, LR tends to give good results. However, traditional LR has significant weaknesses in overcoming the complexity and redundancy of the solution. To this end, many solutions have been proposed, which tend to be a common approach with few results. The main purpose of this paper is to research and design the debugging method of digital grid connection (GC) based on the logistic regression model (LRM). On this basis, this paper proposes an accurate autoregressive model and provides detailed information to ensure the reliability and excellence of the model, which is verified by experiments. The experimental results show that using different classification methods, the test error and selection of data sets have different results. It can be seen that the ten-fold cross-validation error for adaptive elastic net LR is 1/38 and the test error is 1/34. Compared with other methods, the classification performance of adaptive elastic net LR is also superior, even better than other classification methods.