On-line, closed-loop identification of multivariable systems

New approach to constrained predictive control with simultaneous model identification

“…Closed-loop identification has been aGdressed extensively in a linear stochastic control setting (Astrom and Wittenmark, 1989). Good discussions from a stochastic control viewpoint on how to circumvent the difficulty of nonidentifiability and further obtain a good model for a process under feedback control are given in Box (19761, Gustavsson et al (1977), Caines (1984), and Melo and Friedly (1992). Van den Hof and Schrama (1994) and Gevers (1993) review recent research on new criteria for closed-loop identification of state space or input-output models for control purposes.…”

Decentralized identification for multivariable integrating processes with time delays from closed-loop step tests

“…. , n (as referred to in [9,17,18]) and the system error of the kth loop during the step change on the ith set point with zero initial value is sampled as e i k (t), k = 1, . .…”

Identification from step response – The integral equation approach

“…[35] Identification from closed loop step response has been another area of extensive research; [36][37][38][39][40][41][42][43] there are some closed loop methods applicable for unstable processes [44][45][46] and for multiple input multiple output models. [47][48][49][50][51] Closed loop identification methods using optimization approaches have also been proposed. [46,52,53] MIMO open loop identification methods have been proposed in the literature.…”