Continuous-time Modelling by Multirate Processing of Sampled Data

Abstract In most control design methodologies a model of the plant to be controlled is assumed to be known. This model is commonly obtained by processing data obtained by sampling the input and output of the plant and then using data identification techniques. However, a severe restriction in this type of model is that it can only be used with the original sampling rate. In many cases of practical interest the estimation and control sampling periods may be different. This precludes direct digital controller design based on the discrete identified model. The purpose of this paper is to present a new algorithm which allows one to extract the continuous time model of a plant, from sampled I/O data. This continuous time model can be used for sampled data control design where the sampling rate can differ from that one applied during the identification experiment. The simplicity of the algorithm allows one to use it for on-line modelling