Optimal sensor placement for agro‐hydrological systems

The estimation of soil moisture is essential for developing advanced closed‐loop irrigation schemes. One associated problem is how to place the sensors appropriately in the soil to provide good measurements for state estimation. In this work, we address the problem of optimal sensor placement for state estimation of agro‐hydrological systems. A systematic approach is proposed to find the minimum number of sensors that ensures the observability of the entire system and then to find the best locations of the sensors in terms of degree of observability. The Richards equation that is used to describe the dynamics of the agro‐hydrological system is discretized into a large‐scale nonlinear state‐space model. In the proposed procedure, the key steps include order reduction of the large‐scale system model, exploration of the minimum number of sensors needed for state estimation and optimal placement of the sensors in the soil. Three different scenarios are considered and optimal sensor placement is addressed for all the scenarios using the proposed procedure. Simulation results show the effectiveness of the proposed procedure and methods.