A Piece-Wise Linear Model-Based Algorithm for the Identification of Nonlinear Models in Real-World Applications

In this work, a data-driven approach for the identification of piece-wise linear model nitrogen oxide daily concentration simulation is presented and applied. The has been identified by using measured concentrations, meteorological variables, emission levels estimated starting from results contained in suitable databases. We propose an innovative methodology that jointly optimizes clustering parameter identification. procedure applied considering data Milan (Italy) metropolitan area. compared with two state-of-the-art approaches based on two-step, cluster-based algorithm Hammerstein–Wiener models. show how, application, devised ensures better performance respect to literature methods, both terms statistical indexes (correlation, normalized mean absolute error) problem-specific metrics (hit ratio, false alarm). For reason, can be considered used definition optimal control strategies.