Table 5 Analysis of model hyperparameter settings for fault detection in NEVs.
From: Enhancing fault detection in new energy vehicles via novel ensemble approach
Technique | Hyperparameters |
|---|---|
Logistic Regression | random_state: 0, max_iter: 1000 |
Passive Aggressive Classifier | C: 0.5, max_iter: 1000, early_stopping: True, n_iter_no_change: 5, random_state: 0 |
Ridge Classifier | alpha: 1.0, solver: ’auto’, tol: 1e-3 |
Perceptron | penalty: ’l2’, alpha: 0.0001, max_iter: 1000, tol: 1e-3, random_state: 0 |
GRU Model | GRU Layer 1: units=128, activation=’relu’, dropout=0.2, GRU Layer 2: units=64, activation=’relu’, dropout=0.2, Dense Layer: units=32, Output Layer: units= 4, activation=’softmax’ |
CNN Model | Conv1D Layer 1: filters=256, kernel_size=2, activation=’relu’, MaxPooling1D: pool_size=2, Conv1D Layer 2: filters=128, kernel_size=2, Dense Layer: units=32, Output Layer: units=4, activation=’softmax’ |
ANN Model | Dense Layer 1: units=512, activation=’relu’, dropout=0.3, Dense Layer 2: units=256, activation=’relu’, dropout=0.3, Dense Layer 3: units=128, Dense Layer 4: units=64, Dense Layer 5: units=32, Output Layer: units=4, activation=’softmax’ |
