Though being presented and tested in silico, this work assumes a strong practical relevance in view of the results achieved. Moreover, the models identified in the in situ conditions are exploited to synthesise linear regulators and assess the closed-loop performances of the overall control methodology. The novel approach can cope with exogenous, non-measured disturbances acting during the identification experiments, and considers possible constraints limiting the experimental design. This work moves a step forward towards the application of service pressure regulation in situ, by proposing an effective model identification approach for the linear models, based on spectral analysis. These models can be straightforwardly derived in a simulated environment, but their in situ identification may be challenging due to the presence of non-measurable, exogenous disturbances.

Specifically, the tuning should be based on high order, linear models, which describe the system dynamics around a nominal working point. numerically, that linear control systems can be effectively adopted for this task, provided that a careful tuning is performed. Recent works demonstrated in silico, i.e. In the context of Water Distribution Networks, service pressure regulation is an important technique that allows reductions in leakage, risk of pipe bursts and mechanical stress to the infrastructure.

Rf toolbox malab l section