Pressure measurement location selection in industrial gas headers for buffering control

Abstract

This paper uses a nonlinear model of the pressure and flow dynamics inside a gas header to obtain the optimal pressure measurement location for buffering control. The nonlinear model is derived from governing equations using the spectral element method (SEM). The pressure states at the collocation points are estimated using an extended Kalman filter (EKF) and subsequently used as potential measuring locations. A staggered regulatory pressure control scheme is used to evaluate the closed-loop performance of the objective function in a self-optimising control (SOC) framework. It is found that there are opposing goals in the optimisation scheme and that the final pressure location selection is dependent on the economic importance of these goals. The economic goals which cause opposing control objectives include minimising flaring, providing down-stream consumer stability, and limiting expensive supplier use.