Leakage localization, demand forecasting and pump scheduling optimization

Increase the efficiency of water resources management in urban water networks

Reduce waste of energy and water

Demand forecast driven optimization

Online learning and optimization (reinforcement learning)

Leakage localization:

  • Simulation of several leakage scenarios for the computation of induced flow and pressure variations
  • Machine Learning for inverting the relation: inferring the set of (simulated) scenarios associated to the actual flow and pressure data (from sensors)

Demand forecasting:

  • Time series clustering for the identification of typical patterns
  • Learning a forecasting model for each identified pattern

Pump scheduling optimization:

  • Global Optimization using hydraulic simulation and demand forecasts
  • Reinforcement Learning for online control/optimization 

Accurate (water) demand forecast(MAPE lower than 2-3%) and anomaly detection (on smart metering data)

Leakage localization (60-80% of correct localization)

Pump scheduling optimization (5-10% energy costs reduction)