Open pagand opened 1 year ago
Build a LSTM model as the base model.
This model contains the following as input: water depth, engine flow rate temperature, pitch, power, rate of turn, speed, thrust, torque, mode, season, weekday, current, direction, weather info, wind info, resistance ratio, previous longitude, lattitude, heading, SOG, STW
and the following as output: fuel consumption, current heading, longitude, latitude, SOG, STW
Something to mention is that, as the base model, this model does not include the clustered feature, so we can compare how does the clustered feature impact the model performance in the future
Goal: training a model with high accuracy.
Working on the new model with revised inputs and outputs
@yim-fan put the github link for the LSTM model.
Some correction: I am now using time series transformer model, and the link can be found here: https://github.com/pagand/model_optimze_vessel/blob/bc58d55b0b7dfd5bb4beb9232de2290f9b5be86e/Modeling/fc_transformer.ipynb
Build a working version of the fuel consumption prediction model with high accuracy.