Prepared datasets and models for iterative modeling of heat plumes in groundwater. Models were trained with Iterative modeling.

File explanation:

1HP.zip

This zip file contains all prepared datapoints with a single heat pump. The input data fields are pressure, permeability, position of the heat pump, normalized distance to the heat pump, and temperature.

2HP.zip

This zip file contains all prepared datapoints with two heat pumps. The input data fields are pressure, permeability, position of the heat pump, normalized distance to the heat pump, and temperature.

3HP_raw.zip

This zip file contains all raw datapoints with three heat pumps.

5HP_raw.zip

This zip file contains all raw datapoints with five heat pumps.

unet_stand_f64_d5_k4_2500dp

This folder contains the model for the standard architecture. F stands for the number of initial features, d for the depth of the network, and k for the kernel size. The amount of datapoints the model was trained on is indicated by dp.

unet_para_f64_d5_k4_pcs_2500dp

This folder contains the model for the parallel architecture. F stands for the number of initial features, d for the depth of the network, and k for the kernel size. The amount of datapoints the model was trained on is indicated by dp.

unet_quad_f64_d5_k4_2500dp

This folder contains the model for the quadratic architecture. F stands for the number of initial features, d for the depth of the network, and k for the kernel size. The amount of datapoints the model was trained on is indicated by dp.

For usage instructions see Iterative modeling.