Jupyter Notebook: Remaining Useful Life (RUL) Prediction with Variational Autoencoder

{ "title": "Jupyter Notebook: Remaining Useful Life (RUL) Prediction with Variational Autoencoder", "description": "This Jupyter Notebook demonstrates a Variational Autoencoder (VAE) model for Remaining Useful Life (RUL) prediction. It includes data loading, model creation, training, evaluation, and visualization.", "keywords": "RUL prediction, Variational Autoencoder, VAE, Jupyter Notebook, Machine Learning, Deep Learning, Predictive Maintenance, Sensor Data, Time Series", "content": "```python import model import utils import numpy as np

if name == "main": # ------------------------------ DATA ----------------------------------- dataset = input("Enter dataset (FD001, FD002, FD003, FD004): ") # sensors to work with: T30, T50, P30, PS30, phi sensors = ['s_3', 's_4', 's_7', 's_11', 's_12'] # windows length sequence_length = 30 # smoothing intensity alpha = 0.1 # max RUL threshold = 125

x_train, y_train, x_val, y_val, x_test, y_test = utils.get_data(dataset, sensors,
sequence_length, alpha, threshold)
# -----------------------------------------------------------------------

# ----------------------------- MODEL -----------------------------------
timesteps = x_train.shape[1]
input_dim = x_train.shape[2]
intermediate_dim = 300
batch_size = 128
latent_dim = 2
epochs = 10000
optimizer = 'adam'

RVE = model.create_model(timesteps, 
        input_dim, 
        intermediate_dim, 
        batch_size, 
        latent_dim, 
        epochs, 
        optimizer,
        )

# Callbacks for training
model_callbacks = utils.get_callbacks(RVE, x_train, y_train)
# -----------------------------------------------------------------------

# --------------------------- TRAINING ---------------------------------
results = RVE.fit(x_train, y_train,
        shuffle=True,
        epochs=epochs,
        batch_size=batch_size,
        validation_data= (x_val, y_val),
        callbacks=model_callbacks, verbose=2)
# -----------------------------------------------------------------------

# -------------------------- EVALUATION ---------------------------------
RVE.load_weights('./checkpoints/checkpoint')
train_mu = utils.viz_latent_space(RVE.encoder, np.concatenate((x_train, x_val)), np.concatenate((y_train, y_val)))
test_mu = utils.viz_latent_space(RVE.encoder, x_test, y_test.clip(upper=threshold))
# Evaluate
y_hat_train = RVE.regressor.predict(train_mu)
y_hat_test = RVE.regressor.predict(test_mu)

utils.evaluate(np.concatenate((y_train, y_val)), y_hat_train, 'train')
utils.evaluate(y_test, y_hat_test, 'test')
# -----------------------------------------------------------------------

**Libraries used:**

- `model`: Custom model class
- `utils`: Custom utility functions
- `numpy`: For array and matrix operations
- `keras`: For building the neural network model
- `tensorflow`: For training and evaluating the model