Optimization of monitoring network for flushing control in a lowlying polder (Optimization of sensor placement)

Last updated: 26 June 2017 at 00:59


Saline groundwater exfiltration to surface water increases the surface water salinity levels and degrades the surface water quality in low-lying polders, making agricultural use impractical. Therefore, freshwater diverted from rivers is used for flushing (i.e. diluting) canals and ditches of the polders. Current water management strategies for flushing operations are not efficient in their use of fresh water and energy for pumping can be improved by using advanced control methods like model predictive control (MPC). However, the efficiency of the MPC depends on the system understanding. The model for the controller has to be updated with real time measurements of salinity and water level in order to compute a stable optimal feedback action based on the current status of the polder system. However, in a complex polder system that has high spatial variation of salinity concentrations due to different source of saline groundwater exfiltration, deciding on where to measure the water level and the salinity concentration is a challenging problem. In the application of an updated controller scheme for operation of the system, an optimal monitoring network strategy is required, considering cost of implementation and operational efficiency.

Research Objective:

Scope of the thesis is to develop an optimal monitoring network in a polder catchment situated in Haarlemmermeer polder.


  • improving the currently available simulation models (for water transport, salinity transport and saline groundwater exfiltration)
  • running a measurement campaign to better understand the saline groundwater exfiltration in the research area
  • developing a state of art strategy for sensor placement in the area. This strategy will be applied to a real test case.


Hydraulic modelling, Sensor Networks, Optimal Flushing for Water Quality

More information: