Signal estimation adaptive algorithm with latch-mechanism for real-time water quality monitoring

The protection of water reservoirs requires precise monitoring of quality parameters, which is a crucial aspect of water resource management. This article proposes an algorithm for real-time reconstruction of unmeasured signals using only easily accessible measurement data. The proposed method relies on a specially designed latch mechanism and a mathematical river model. The river model under consideration is described by nonlinear ordinary differential equations obtained via a transformation of partial differential equations. The task was accomplished through the application of an adaptive signal estimation algorithm specifically developed for this purpose, with additional tuning of dynamic properties achieved through precise placement of eigenvalues. The results of simulation studies confirmed improved accuracy in reproducing dynamic processes, particularly for signals that are challenging to measure directly, compared to other analysed methods. As a practical application, the proposed algorithm is implemented as a soft sensor for monitoring a biochemically polluted river. • Mathematical modelling and estimation of difficult-to-measure signals. • Algorithms to monitor river pollution in real time. • Improved control performance by enforcing faster system dynamics. • PALM algorithm with latching mechanism to select the required system dynamics.