Weighted multirate q-Markov Cover identification using PRBS – an application to engine systems

Abstract

The q-Markov COVariance Equivalent Realization (q-Markov Cover) method for identification uses either pulse, white noise or PRBS (Pseudo-Random Binary Signal) as test excitation. This paper extended the q-Markov Cover using PRBS to the weighted multirate case, that is, the sample rate of the PRBS signal is different from the system output one. Then, the multirate PRBS q-Markov Cover is applied to identify a diesel engine model from the fuel command input to the engine speed output. The identified engine model has order of two and approximates the pure fuel system time delay using a first-order transfer function with a non-minimum phase numerator. Finally, the identified engine model was successfully used for designing engine idle speed governor and obtained satisfactory performance in the first try.