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. 

The spread of the COVID-19 pandemic and consequent lockdowns all over the world have had various impacts on atmospheric quality. This study aimed to investigate the impact of the lockdown on the air quality of Nanjing, China. The off-axis measurements from state-of-the-art remote-sensing Multi-Axis Differential Optical Absorption Spectroscope (MAX-DOAS) were used to observe the trace gases, i.e., Formaldehyde (HCHO), Nitrogen Dioxide (NO2), and Sulfur Dioxide (SO2), along with the in-situ time series of NO2, SO2 and Ozone (O3). The total dataset covers the span of five months, from 1 December 2019, to 10 May 2020, which comprises of four phases, i.e., the pre lockdown phase (1 December 2019, to 23 January 2020), Phase-1 lockdown (24 January 2020, to 26 February 2020), Phase-2 lockdown (27 February 2020, to 31 March 2020), and post lockdown (1 April 2020, to 10 May 2020). The observed results clearly showed that the concentrations of selected pollutants were lower along with improved air quality during the lockdown periods (Phase-1 and Phase-2) with only the exception of O3, which showed an increasing trend during lockdown. The study concluded that limited anthropogenic activities during the spring festival and lockdown phases improved air quality with a significant reduction of selected trace gases, i.e., NO2 59%, HCHO 38%, and SO2 33%. We also compared our results with 2019 data for available gases. Our results imply that the air pollutants concentration reduction in 2019 during Phase-2 was insignificant, which was due to the business as usual conditions after the Spring Festival (Phase-1) in 2019. In contrast, a significant contamination reduction was observed during Phase-2 in 2020 with the enforcement of a Level-II response in lockdown conditions i.e., the easing of the lockdown situation in some sectors during a specific interval of time. The observed ratio of HCHO to NO2 showed that tropospheric ozone production involved Volatile Organic Compounds (VOC) limited scenarios. 

Recently, the occurrence of fog and haze over China has increased. The retrieval of trace gases from the multi-axis differential optical absorption spectroscopy (MAX-DOAS) is challenging under these conditions. In this study, various reported retrieval settings for formaldehyde (HCHO) and sulfur dioxide (SO2) are compared to evaluate the performance of these settings under different meteorological conditions (clear day, haze, and fog). The dataset from 1st December 2019 to 31st March 2020 over Nanjing, China, is used in this study. The results indicated that for HCHO, the optimal settings were in the 324.5–359 nm wavelength window with a polynomial order of five. At these settings, the fitting and root mean squared (RMS) errors for column density were considerably improved for haze and fog conditions, and the differential slant column densities (DSCDs) showed more accurate values compared to the DSCDs between 336.5 and 359 nm. For SO2, the optimal settings for retrieval were found to be at 307–328 nm with a polynomial order of five. Here, root mean square (RMS) and fitting errors were significantly lower under all conditions. The observed HCHO and SO2 vertical column densities were significantly lower on fog days compared to clear days, reflecting a decreased chemical production of HCHO and aqueous phase oxidation of SO2 in fog droplets. 

Abstract. The IceCube Neutrino Observatory instruments about 1 km3 of deep, glacial ice at the geographic South Pole. It uses 5160 photomultipliers to detect Cherenkov light emitted by charged relativistic particles. An unexpected light propagation effect observed by the experiment is an anisotropic attenuation, which is aligned with the local flow direction of the ice. We examine birefringent light propagation through the polycrystalline ice microstructure as a possible explanation for this effect. The predictions of a first-principles model developed for this purpose, in particular curved light trajectories resulting from asymmetric diffusion, provide a qualitatively good match to the main features of the data. This in turn allows us to deduce ice crystal properties. Since the wavelength of the detected light is short compared to the crystal size, these crystal properties include not only the crystal orientation fabric, but also the average crystal size and shape, as a function of depth. By adding small empirical corrections to this first-principles model, a quantitatively accurate description of the optical properties of the IceCube glacial ice is obtained. In this paper, we present the experimental signature of ice optical anisotropy observed in IceCube light-emitting diode (LED) calibration data, the theory and parameterization of the birefringence effect, the fitting procedures of these parameterizations to experimental data, and the inferred crystal properties.