Multivariate multiplicative composite modelling techniques

Measurements perform the determination of physical quantities from experiment. Herewith an experimental system model is the key to extracting information from the system. The quality of obtained information depends directly on the quality of the model. With this concern novel techniques for model quality improvement have been fashioned. Experimental system comprehensive modelling was accomplished by theoretical and empirical data composition. These modelling techniques combine prior knowledge based approach in mechanistic model elaboration, and for attainment the enhanced level of adequacy, accuracy and precision carry out the approximation of the exact unknown system model, simultaneously by available mechanistic and polynomial empirical functions. Two approaches for multiplicative composite model solving were elaborated. From models’ validation it was found that multiplicative modelling approach in comparison with pure statistical modelling permits to attain less discrepancy to experimental evidence for the whole region of interest for models’ predictor variables.