Information fusion infrastructure for remote-sensing and in-situ sensor data to model people dynamics

2014 ◽  
Vol 5 (1) ◽  
pp. 54-69 ◽  
Author(s):  
Florian Hillen ◽  
Bernhard Höfle ◽  
Manfred Ehlers ◽  
Peter Reinartz
Keyword(s):  
Remote Sensing ◽  
Information Fusion ◽  
Sensor Data ◽  
In Situ Sensor

scholarly journals Sensor-integrating gears: wear detection by in-situ MEMS acceleration sensors

2021 ◽  
Author(s):  
Julian Peters ◽  
Lorenz Ott ◽  
Matthias Dörr ◽  
Thomas Gwosch ◽  
Sven Matthiesen
Keyword(s):  
Regression Model ◽  
Data Quality ◽  
Gear Tooth ◽  
Measurement Data ◽  
Tooth Wear ◽  
Sensor Data ◽  
Acceleration Sensors ◽  
In Situ Sensors ◽  
In Situ Sensor

AbstractGear tooth wear is a common phenomenon leading to malfunctions in machines. To detect wear and faults, gear condition monitoring by vibration is established. The problem is that the measurement data quality for detection of wear by vibration is not good enough with currently established measurement methods, caused by long signal paths of the commonly used housing mounted sensors. In-situ sensors directly at the gear achieve better data quality, but are not yet proved in wear detection. Further it is unknown what analysis methods are suited for in-situ sensor data. Existing gear condition metrics are mainly focused on localized gear tooth faults, and do not estimate wear related values. This contribution aims to improve wear detection by investigating in-situ sensors and advance gear condition metrics. Using a gear test rig to conduct an end of life test, the wear detection ability of an in-situ sensor system and reference sensors on the bearing block are compared through standard gear condition metrics. Furthermore, a machine-learned regression model is developed that maps multiple features related to gear dynamics to the gear mass loss. The standard gear metrics used on the in-situ sensor data are able to detect wear, but not significantly better compared to the other sensors. The regression model is able to estimate the actual wear with a high accuracy. Providing a wear related output improves the wear detection by better interpretability.

scholarly journals X-ray Computed Tomography Data ofAdditive Manufacturing Metrology Testbed (AMMT) Parts: “Overhang Part X4”

2020 ◽  
Vol 125 ◽  
Author(s):  
Max Praniewicz ◽  
Brandon Lane ◽  
Felix Kim ◽  
Christopher Saldana
Keyword(s):  
Additive Manufacturing ◽  
Three Dimensional ◽  
Sensor Data ◽  
X Ray ◽  
Surface Data ◽  
Design Build ◽  
X Ray Computed ◽  
In Situ Sensor ◽  
Stereolithography Files

This document provides details on the data and files generated from post-build X-ray computedtomography (XCT) measurements of the four parts built as part of the “Overhang Part X4” dataset. The “Overhang Part X4” dataset was a three-dimensional (3D) additive manufacturing (AM) build performed on the Additive Manufacturing Metrology Testbed (AMMT) by Ho Yeung and Brandon Lane on June 28, 2019. The files discussed in this document include image sequences for each part, stereolithography files (.STL) of the surface data extracted from XCT. This data is one of a set of “AMMT Process Monitoring Datasets”, as part of the Metrology for Real-Time Monitoring of Additive Manufacturing project at the National Institute of Standards and Technology (NIST). In-situ sensor data, part design, build command and scan strategy data, materials, and associated metadata for this build are described in Ref. [1]. Readers should refer to the AMMT datasets web page for updates.

Forecasting Seasonal Streamflow Using a Stacked Recurrent Neural Network

2020 ◽  
Author(s):  
David Lambl ◽  
Dan Katz ◽  
Eliza Hale ◽  
Alden Sampson
Keyword(s):  
Remote Sensing ◽  
Hydrological Modeling ◽  
Short Term Memory ◽  
Remote Sensing Data ◽  
Sensor Data ◽  
Historical Information ◽  
Sensing Data ◽  
The Future ◽  
In Situ Sensors

<p>Providing accurate seasonal (1-6 months) forecasts of streamflow is critical for applications ranging from optimizing water management to hydropower generation. In this study we evaluate the performance of stacked Long Short Term Memory (LSTM) neural networks, which maintain an internal set of states and are therefore well-suited to modeling dynamical processes.</p><p>Existing LSTM models applied to hydrological modeling use all available historical information to forecast contemporaneous output. This modeling approach breaks down for long-term forecasts because some of the observations used as input are not available in the future (e.g., from remote sensing and in situ sensors). To solve this deficiency we train a stacked LSTM model where the first network encodes the historical information in its hidden states and cells. These states and cells are then used to initialize the second LSTM which uses meteorological forecasts to create streamflow forecasts at various horizons. This method allows the model to learn general hydrological relationships in the temporal domain across different catchment types and project them into the future up to 6 months ahead.</p><p>Using meteorological time series from NOAA’s Climate Forecast System (CFS), remote sensing data including snow cover, vegetation and surface temperature from NASA’s MODIS sensors, SNOTEL sensor data, static catchment attributes, and streamflow data from USGS we train a stacked LSTM model on 100 basins, and evaluate predictions on out-of-sample periods from these same basins. We perform sensitivity analysis on the effects of remote sensing data, in-situ sensors, and static catchment attributes to understand the informational content of these various inputs under various model architectures. Finally, we benchmark our model to forecasts derived from simple climatological averages and to forecasts created by a single LSTM that excludes all inputs without forecasts.</p><p> </p>

Quantification of shallow water quality parameters by means of remote sensing

2003 ◽  
Vol 27 (1) ◽  
pp. 24-43 ◽  
Author(s):  
Yansui Liu ◽  
Md Anisul Islam ◽  
Jay Gao
Keyword(s):  
Remote Sensing ◽  
Water Quality ◽  
Suspended Sediments ◽  
Remotely Sensed ◽  
Quality Parameters ◽  
Water Quality Parameters ◽  
Sensor Data ◽  
Remotely Sensed Data ◽  
Ancillary Data

Quantification of quality parameters of inland and near shore waters by means of remote sensing has encountered varying degrees of success in spite of the high variability of the parameters under consideration and limitations of remote sensors themselves. This paper comprehensively evaluates the quantification of four types of water quality parameters: inorganic sediment particles, phytoplankton pigments, coloured dissolved organic material and Secchi disk depth. It concentrates on quantification requirements, as well as the options in selecting the most appropriate sensor data for the purpose. Relevant factors, such as quantification implementation and validation of the quantified results are also extensively discussed. This review reveals that the relationship between in situ samples and their corresponding remotely sensed data can be linear or nonlinear, but are nearly always site-specific. The quantification has been attempted from terrestrial satellite data largely for suspended sediments and chlorophyll concentrations. The quantification has been implemented through integration of remotely sensed imagery data, in situ water samples and ancillary data in a geographic information system (GIS). The introduction of GIS makes the quantification feasible for more variables at an increasingly higher accuracy. Affected by the number and quality of in situ samples, accuracy of quantification has been reported in different ways and varies widely.

Sign in / Sign up

Export Citation Format

Share Document