DL for predicting RLU

As we have discussed, we are trying to calculate the amount of time before an engine needs maintenance. What makes this dataset special is that the engines run all the way until failure, giving us the precise RLU information for every engine at every point in time.

Calculating cut-off times

Let's consider the FD004 dataset, that contains as much as 249 engines (engine_no) monitored over time (time_in_cycles). Each engine has operational_settings and sensor_measurements recorded for each cycle:

data_path = 'train_FD004.txt'
data = utils.load_data(data_path)

To train a model that will predict RUL, we can simulate real predictions by choosing a random point in the life of the engine and only...