Data and code from: Deep reinforcement learning for pressure optimization in water distribution networks with multiple pumping stations

This dataset provides the complete code, model files, and tabular data
used to train and evaluate a deep reinforcement learning (DRL) agent for
pressure optimization in large-scale water distribution networks with
multiple pumping stations. It includes ten Python scripts (for environment
definition, training, testing, and evaluation), calibrated EPANET
hydraulic model files, and eight Excel workbooks containing control
parameters, diurnal demand data, synthetic and observed evaluation sets,
and energy-performance summaries. The dataset enables full reproduction of
the case-study results and supports reuse for developing alternative DRL
algorithms or benchmarking water-network optimization methods. The
repository is self-contained and can be executed using Python 3.10 with
the package versions specified herein.