In the paper a new approach to fuzzy confidence interval identification is presented. The method combines a fuzzy identification methodology with some ideas from statistics. The idea is to find, on a finite set of measured data, the confidence interval defined by the lower and upper fuzzy bound that define the band that contains all the output measurements. The fuzzy confidence interval model can be used in process monitoring, fault detection or in the case of robust control design.
COBISS.SI-ID: 6931284
The objective of the present work was to study the influence of particle size (and indirectly, the influence of dry granulation process) and the settings of the tableting parameters on the tablet capping tendency. Artificial neural network and fuzzy models were used for modelling the effect of the particle size and the tableting machine settings on the capping coefficient. The suitability of routinely measured quantities of tablet quality were tested. Results showed that model-based expert systems can significantly improve the trial-and-error procedures.
COBISS.SI-ID: 7095124
This paper deals with a novel formulation of continuous-time model-predictive control for nonlinear systems. A nonlinear-mapping approximation, employing a PWL approximation, is also an integral part of the control scheme, and thus removes the need for output-function invertibility. The analytical formulation of the control law makes it possible to use the method in practice, especially in the chemical industry. An illustrative experiment is conducted to compare the proposed approach with the method of nonlinear H1 control of a pH-neutralization process.
COBISS.SI-ID: 7440468
The paper deals with the problem of localizing a mobile robot in a line-based environment description by using the extended Kalman filter (EKF). Uncertainties of the line parameters, which comprise the output noise covariance matrix of the EKF, arise from the LRF’s distance- and angle-measurement error. In this paper a method for estimating the covariances of the line parameters, which arise from the classic least squares, is proposed. The method is compared with the method resulting from the orthogonal LSQ in terms of computational complexity and statistical accuracy.
COBISS.SI-ID: 7315284
In neurology and pharmacokinetics, the systemic paradigm was already well accepted, while other areas of bio-medicine were more or less accepting the reductionism paradigm where systems were studied through the analysis of its isolated sub-systems. Mathematical modelling, as one of the principal tools of control engineering, is now becoming necessary tool also in the analysis of biological systems.
COBISS.SI-ID: 7378260