Continuous plastics compounding processes are highly complex from a physicochemical point of view and correspondingly difficult to optimize. Model-based digital twins suitable for this purpose require complex simulation approaches, trained personnel and input variables that are difficult to determine. However, data-based digital twins, which are in principle suitable for this purpose, often fail because of the enormous trial effort which is required to generate a sufficiently large database. To overcome this problem this paper describes a hybrid approach for generating the necessary database for a data-based digital twin. By intelligent combination of real experiments, adaptation of the physical-chemical process model to these few experimental data and subsequent data cloud generation with the adapted process model results in a sufficient data base for the digital twin.
We describe a nonlinear regression problem, where the regression functions have an additive structure and the dependent variable is a one-dimensional time series. Multivariate time series with unknown time delay operators are used as independent variables. By fitting a feedforward neural network with block structure to the data, we estimated the additive regression function and, parallel to this, the time lags. We present the consistency proof of neural network weights estimator and the time lag estimator independently from each other. In the practical part of the article, we present the useful feature of blocked neural networks to estimate the relevance measures of each input variable in a simple way. Furthermore, we propose an approach to solve the well-known variable selection problem for the class of nonlinear multivariate beta-mixing time series models considered here. Finally, we apply the methodology to an artificial example.
We present a simulation framework for spunbond processes and use a design of experiments to investigate the causeand-effect-relations of process and material parameters on the fiber laydown on a conveyor belt. The analyzed parameters encompass the inlet air speed and suction pressure, as well as the E modulus, density and line density (titer) of the filaments. The fiber laydown produced by the virtual experiments is statistically quantified and the results are analyzed by a blocked neural network. This forms the basis for the prediction of the fiber laydown characteristics and enables a quick ranking of the significance of the influencing effects. We conclude our research by an analysis of the nonlinear cause-and-effect relations. Compared to the material parameters, suction pressure and inlet air speed have a negligible effect on the fiber mass distribution in (cross) machine direction. Changes in the line density of the filament have a 10 times stronger effect than changes in E modulus or density. The effect of the E modulus on the throwing range in machine direction is of particular note, as it reverses from increasing to decreasing in the examined parameter regime. K E Y W O R D S spunbond process; machine learning; blocked neural networks 1 arXiv:1911.06213v2 [stat.ML]
In the face of climate change and rising energy prices, lowering energy usage of industrial machines is gaining widespread attention. Αpropriate machine settings could lead to reduced production costs and lower environmental impact, while simultaneously maintaining products' quality. However, defining the complex, nonlinear dependencies between these settings and energy usage or quality in manufacturing is often a challenging task. In the presented work, a method for optimized machine settings recommendation is proposed using inverse classification via autoencoders. The algorithm can suggest operation parameters, based on predefined intervals of energy consumption and product properties. The performance is evaluated on data generated by a digital twin of an extrusion process.
scite is a Brooklyn-based organization that helps researchers better discover and understand research articles through Smart Citations–citations that display the context of the citation and describe whether the article provides supporting or contrasting evidence. scite is used by students and researchers from around the world and is funded in part by the National Science Foundation and the National Institute on Drug Abuse of the National Institutes of Health.