Identification of non-minimum phase processes with time delay in the presence of measurement noise

•Exact identification of non-minimum phase process dynamics.•Derivation of generalized analytical expressions.•Relay feedback testing is used to get the process information.•Estimation of process model parameters even in the presence of measurement noise.•Validation of proposed method with the help of examples from the literature.

Time domain based generalized analytical expressions are derived to estimate the exact model parameters of a class of stable non-minimum phase (NMP) processes with time delay. The novelty of the proposed method lies in generalizing a second order plus time delay (SOPDT) NMP process model in terms of a class of models like first order plus time delay (FOPDT), SOPDT underdamped, critically damped and integrating processes. The mathematical expressions are employed to estimate four accurate parameters at a time. Relay with hysteresis reduces the effect of noise and further mitigation of noise is achieved through a denoising block yielding a clean process output. Processes modeled in the form of transfer functions are considered to validate the technique.