Search for: All records

We develop a comprehensive framework for storing, analyzing, forecasting, and visualizing industrial energy systems consisting of multiple devices and sensors. Our framework models complex energy systems as a dynamic knowledge graph, utilizes a novel machine learning (ML) model for energy forecasting, and visualizes continuous predictions through an interactive dashboard. At the core of this framework is A-RNN, a simple yet efficient model that uses dynamic attention mechanisms for automated feature selection. We validate the model using datasets from two manufacturers and one university testbed containing hundreds of sensors. Our results show that A-RNN forecasts energy usage within 5% of observed values. These enhanced predictions are as much as 50% more accurate than those produced by standard RNN models that rely on individual features and devices. Additionally, A-RNN identifies key features that impact forecasting accuracy, providing interpretability for model forecasts. Our analytics platform is computationally and memory efficient, making it suitable for deployment on edge devices and in manufacturing plants. 

Abstract Capillary breakup of cores is an exclusive approach to fabricating fiber-integrated optoelectronics and photonics. A physical understanding of this fluid-dynamic process is necessary for yielding the desired solid-state fiber-embedded multimaterial architectures by design rather than by exploratory search. We discover that the nonlinearly complex and, at times, even chaotic capillary breakup of multimaterial fiber cores becomes predictable when the fiber is exposed to the spatiotemporal temperature profile, imposing a viscosity modulation comparable to the breakup wavelength. The profile acts as a notch filter, allowing only a single wavelength out of the continuous spectrum to develop predictably, following Euler-Lagrange dynamics. We argue that this understanding not only enables designing the outcomes of the breakup necessary for turning it into a technology for materializing fiber-embedded functional systems but also positions a multimaterial fiber as a universal physical simulator of capillary instability in viscous threads.