A model-free method based on Laguerre series expansion for dynamic measurement improvement

In this paper, a model-free method for the improvement of time-varying dynamic measurements, based on an orthogonal Laguerre model is proposed. It is appropriate in the case of well-damped measurement process. The main difference with the standard model-free method is that for modelling of measurement process instead of a specific type of FIR model an orthogonal Laguerre model is used. In this manner, the proposed method combines the advantages of model-free method and orthogonal models. As a result, the obtained measurement model is of a lower order than the estimated one with the conventional model-free method. The online estimation of time-varying measurement model is performed by modified linear regression model and a recursive least squares algorithm with a constant trace of the covariation matrix. The method proposed in the paper is validated on temperature measurement examples.