Abstract: An on-line nuclear magnetic resonance (NMR) system, and related methods, are useful for predicting one or more properties of interest of a polymer. In one embodiment, a neural network is used to develop a model which correlates process variables in addition to manipulated NMR output to predict a polymer property of interest. In another embodiment, a partial least square regression technique is used to develop a model of enhanced accuracy. Either the neural network technique or the partial least square regression technique may be used in conjunction with a described multi-model or best-model-selection scheme according to the invention. The polymer can be a plastic such as polyethylene, polypropylene, or polystyrene, or a rubber such as ethylene propylene rubber.

Abstract: A real-time, on-line nuclear magnetic resonance (NMR) system, and related method, can predict various properties of interest of a sample of polymer material. A regression or neural network technique is used to develop a model based upon manipulated NMR output and a resin age factor which compensates for time dependent aging phenomena and enhance predictive accuracy of the model. In a preferred embodiment, the resin age factor is a function of elapsed cycle time prior to sample measurement, sample temperature at time of measurement, and/or sample form. The polymer can be a plastic such as polyethylene, polypropylene, or polystyrene, or a rubber such as ethylene propylene rubber.

Abstract: A nuclear magnetic resonance (NMR) system, and related method, develops two or more regression equations or models for a particular polymer property of interest (e.g., melt index or MI) during a calibration procedure using known samples of a material. The polymer material can be, for example, a plastic (e.g., polyethylene, polypropylene, or polystyrene) or a rubber (e.g., ethylene propylene rubber). Regression models for one or more discrete (i.e., two-valued) variables also are developed during calibration, and these models allow a prediction to be made about which of the two or more property (e.g., MI) models should be used for any particular sample of unknown material. The prediction obtained from the discrete variable model indicates which of the two or more models will produce the most accurate estimation of the property of interest for the unknown sample. The best model is thus selected, and then it is used to estimate the property of interest.

Abstract: A pulsed nmr analysis system for polymers materials extracted from industrial processes at a mobility enhancing temperature (at or above glass transition temperature for amorphous polymers, at the crystalline transition temperature for crystalline and semi-crystalline polymers). The sample is measured via nmr techniques and results correlated to viscosity and melt index or melt flow (which are related to average molecular weights). The nmr system (in or out of resonance) includes: sample throughput system (P, LI, V1, V2) and user system controls (104) to establish digitized free induction decay curves (C), from which components functions are determined using linear or non-linear regression techniques to correlate the curve components to the target nuclei and to flow rates in plastics.