Artificial neural network approaches for prediction of backwater through arched bridge constrictions

This paper presents the findings of laboratory model testing of arched bridge constrictions in a rectangular open channel flume whose bed slope was fixed at zero. Four different types of arched bridge models, namely single opening semi-circular arch (SOSC), multiple opening semi-circular arch (MOSC), single opening elliptic arch (SOE), and multiple opening elliptic arch (MOE), were used in the testing program. The normal crossing (ϕ = 0), and five different skew angles (ϕ = 10°, 20°, 30°, 40°, and 50°) were tested for each type of arched bridge model. The main aim of this study is to develop a suitable model for estimating backwater through arched bridge constrictions with normal and skewed crossings. Therefore, different artificial neural network approaches, namely multi-layer perceptron (MLP), radial basis neural network (RBNN), generalized regression neural network (GRNN), and multi-linear and multi-nonlinear regression models, MLR and MNLR, respectively were used. Results of these experimental studies were compared with those obtained by the MLP, RBNN, GRNN, MLR, and MNLR approaches. The MLP produced more accurate predictions than those of the others.     

Obtaining maps and data from the U.S. Geological Survey1

The U.S. Geological Survey produces a variety of resource information for the United States. This includes many data bases of particular interest to planners such as land use and terrain information prepared by the National Mapping Division, water quantity and quality data collected by Water Resources Division, and coal resource information gathered by the Geologic Division. These data are stored in various forms, and information on their availability can be obtained from appropriate offices in the U.S. Geological Survey as well as from USGS Circular 777. These data have been used for the management, development, and monitoring of our Nation's resources by Federal, State, and local agencies.     

Model Predictive Control Coupled with Artificial Intelligence for Eddy Current Dynamometers

The recent studies on Artificial Intelligence (AI) accompanied by enhanced computing capabilities supports increasing attention into traditional control methods coupled with AI learning methods in an attempt to bringing adaptiveness and fast responding features. The Model Predictive Control (MPC) technique is a widely used, safe and reliable control method based on constraints. On the other hand, the Eddy Current dynamometers are highly nonlinear braking systems whose performance parameters are related to many processes related variables. This study is based on an adaptive model predictive control that utilizes selected AI methods. The presented approach presents an updated the mathematical model of an Eddy Current Dynamometer based on experimentally obtained system operational data. Finally, the comparison of AI methods and related learning performances based on the assessment technique of mean absolute percentage error (MAPE) issues are discussed. The results indicate that Single Hidden Layer Neural Network (SHLNN), General Regression Neural Network (GRNN), Radial Basis Network (RBNN), Neuro Fuzzy Network (ANFIS) coupled MPC have quite satisfying performances. The presented results indicate that, amongst them, GRNN appears to provide the best performance.     

Obtaining maps and data from the U.S. Geological Survey*

The U.S. Geological Survey produces a variety of resource information for the United States. This includes many data bases of particular interest to planners such as land use and terrain information prepared by the National Mapping Division, water quantity and quality data collected by Water Resources Division, and coal resource information gathered by the Geologic Division. These data are stored in various forms, and information on their availability can be obtained from appropriate offices in the U.S. Geological Survey as well as from USGS Circular 777. These data have been used for the management, development, and monitoring of our Nation's resources by Federal, State, and local agencies.     

Application of ANN techniques for estimating modal damping of impact-damped flexible beams

A comprehensive laboratory work relating impact damping phenomenon of a flexible beam was carried out by Butt and Akl [Butt AS, Akl FA. Experimental analysis of impact-damped flexible beams. J Eng Mech ASCE 1997;123(4):376–83] to investigate the relation between system’s modal damping ratio $(\zeta )$ and system parameters, namely gap (c, mm), mass (m, kg), modal amplitude $({\Phi }_d)$, frequency (f, Hz), and peak value of the imaginary part of the frequency response functions $(F_I)$. Using a multiple nonlinear regression technique (MNLR), they established a relation between these system parameters and the resulting damping ratio, based on 60 steady-state vibration tests of a flexible beam. In current work, three different artificial neural network approaches (ANNs), namely FFBP (Feed-Forward Back Propagation), RBNN (Radial Basis Function Based Neural Network), and GRNN (Generalized Regression Neural Networks), for estimating modal damping ratio $(\zeta )$ were developed using the data collected by Butt and Akl (1997) and compared with MNLR. The results showed that the RBNN produced slightly better estimations than those of the FFBP and was significantly superior to the MNLR and GRNN.     

On the use of ocean color remote sensing to measure the transport of dissolved organic carbon by the Mississippi River Plume

We investigated the use of ocean color remote sensing to measure the transport of dissolved organic carbon (DOC) by the Mississippi River to the Gulf of Mexico. From 2000 to 2005 we recorded surface measurements of DOC, colored dissolved organic matter (CDOM), salinity, and water-leaving radiances during five cruises to the Mississippi River Plume. These measurements were used to develop empirical relationships to derive DOC, CDOM, and salinity from monthly composites of SeaWiFS imagery collected from 1998 through 2005. We compared our remote sensing estimates of river flow and DOC transport with data collected by the United States Geological Survey (USGS) from 1998 through 2005. Our remote sensing estimates of river flow and DOC transport correlated well (r² ∼ 0.70) with the USGS data. Our remote sensing estimates and USGS field data showed low variability in DOC concentrations in the river end-member (7-11%), and high seasonal variability in river flow (∼ 50%). Therefore, changes in river flow control the variability in DOC transport, indicating that the remote sensing estimate of river flow is the most critical element of our DOC transport measurement. We concluded that it is possible to use this method to estimate DOC transport by other large rivers if there are data on the relationship between CDOM, DOC, and salinity in the river plume.     

An approach to regional wetland digital elevation model development using a differential global positioning system and a custom-built helicopter-based surveying system

Accurate topographic data are critical to restoration science and planning for the Everglades region of South Florida, USA. They are needed to monitor and simulate water level, water depth and hydroperiod and are used in scientific research on hydrologic and biologic processes. Because large wetland environments and data acquisition challenge conventional ground-based and remotely sensed data collection methods, the United States Geological Survey (USGS) adapted a classical data collection instrument to global positioning system (GPS) and geographic information system (GIS) technologies. Data acquired with this instrument were processed using geostatistics to yield sub-water level elevation values with centimetre accuracy (±15 cm). The developed database framework, modelling philosophy and metadata protocol allow for continued, collaborative model revision and expansion, given additional elevation or other ancillary data.     

The ASTER spectral library version 2.0

The Advanced Spaceborne Thermal Emission Reflection Radiometer (ASTER) on NASA's Terra platform has been widely used in geological and other science studies. In support of ASTER studies, a library of natural and man-made materials was compiled as the ASTER Spectral Library v1.2 and made available from http://speclib.jpl.nasa.gov. The library is a collection of contributions in a standard format with ancillary data from the Jet Propulsion Laboratory (JPL), Johns Hopkins University (JHU) and the United States Geological Survey (USGS). A new version of the library (v2.0) is now available online or via CD, which includes major additions to the mineral and rock spectra. The ASTER library provides a comprehensive collection of over 2300 spectra of a wide variety of materials covering the wavelength range 0.4-15.4 µm.     

Suitability of different neural networks in daily flow forecasting

Alternative forms of neural networks have been applied to forecast daily river flows on a continuous basis with the purpose of understanding how recent architectures like ANFIS, GRNN and RBF compare with traditional FFBP when monsoon-fed rivers involving significant statistical bias are involved. The forecasts are made at a location called Rajghat along river Narmada in India. Division of yearly data into four seasons and development of separate networks accordingly was found to be more useful than a single network applicable for the entire year. When a variety of error criteria were viewed together the most satisfactory network for all seasons was the radial basis function, which showed better performance then FFBP, GRNN and ANFIS. The FFBP network was found to be equally acceptable as the RBF in seasons other than the monsoon. Generally the peak flows were more satisfactorily modeled by the RBF than FFBP, GRNN and ANFIS. The relatively simpler handling of data non-linearity in FFBP was more attractive than complex ones of ANFIS and GRNN. The representative statistical model, namely response surface method, yielded highly unsatisfactory results compared to any ANN model involved in this study, confirming that the complexity of ANNs is really necessary to model daily river flows.     

Processing of Hyperion data set for detection of indicative minerals using a hybrid method in Dost-Bayli,Iran

In recent years hyperspectral remote sensing has played an important role in discovering of the earth surface and unmixing is an indispensable part of the hyperspectral data analysis. The most challenging stage in the spectral unmixing is determination of endmembers (EMs). Because of the absence of pure pixels, common methods based on pure pixel assumption do not yield accurate results. On the other hand, Hyperion hyperspectral data acquired by National Aeronautics and Space Administration (NASA)’s Earth Observing-1 (EO-1) system, is available widely but with a lower signal-to-noise ratio (SNR) in comparison to airborne spectrometers. Therefore, the methods with less sensitivity to noise amount will be more efficient in processing of the Hyperion data. Minimum Volume Constrained Nonnegative Matrix Factorization (MVC-NMF) algorithm is then an appropriate technique for EM detection in this case. It, however, has shortcomings in dealing with large data. To fix this problem the first module of Optical Real-time Adaptive Spectral Identification System (ORASIS) was applied for data reduction before running of the MVC-NMF algorithm. The modified technique was then investigated on a set of noisy synthetic data that the outcomes proved its functionality. The Hyperion image of Dost-Bayli area located in the Ardabil province in northwestern Iran, was then unmixed by the mentioned approach. To validate the accuracy of detected minerals, 20 surface samples were collected from the study area and analysed by X-ray diffraction (XRD) for detection of their mineralogical constituents and spectrometry to create a native spectral library. However, both native and United States Geological Survey (USGS) spectral libraries were applied in identification of estimated EMs. The signatures of the obtained EMs by hybrid method were appropriately similar to reference spectra. The mineral abundances maps were therefore generated by linear spectral unmixing (LSU), which have proper consistency with XRD results.     

Biodiversity and ecosystem informatics

The field of Biodiversity and Ecosystem Informatics (BDEI) brings together computer scientists, biologists, natural resource managers, and others who wish to solve real-world challenges while advancing the underlying ecological, computer, and information sciences. The potential for synergies among these disciplines is high, because our need to understand complex, ecosystem-scale processes requires the solution to many groundbreaking technological problems. Fortunately, we are beginning to see increased support for applied computer science and information technology research in the context of environmental problem-solving. In July, 2001, the National Science Foundation (NSF), in collaboration with the United States Geological Survey (USGS), and the National Aeronautics and Space Administration (NASA), invited proposals for high-risk, small-scale planning and incubation activities to catalyze innovation and rapid advances in this new research community. The papers included in this special issue are selected, peer-reviewed summaries from principal investigators involved in this first NSF BDEI effort. These papers provide an overview of this emerging area and remind us that computer and information science and engineering play a crucial role in creating the technologies from which advances in the natural sciences evolve.     

Generalized regression neural network in modelling river sediment yield

The majority of the artificial neural network applications in water resources involve the employment of feed forward back propagation method (FFBP). In this study another ANN algorithm, generalized regression neural network, GRNN, was used in river suspended sediment estimation. Generalized regression neural network does not require an iterative training procedure as in back propagation method. The GRNN simulations do not face the frequently encountered local minima problem in FFBP applications and GRNN does not generate estimates physically not plausible. The neural networks are trained using daily river flow and suspended sediment data belonging to Juniata Catchment in USA. The suspended sediment estimations provided by two ANN algorithms are compared with conventional sediment rating curve and multi linear regression method results. The mean squared error and the determination coefficient are used as comparison criteria. Also the estimated and observed sediment sums are examined in addition to two previously mentioned performance criteria. The ANN estimations are found significantly superior to conventional method results.     

Development of an autonomous bridge deck inspection robotic system

The threat to safety of aging bridges has been recognized as a critical concern to the general public due to the poor condition of many bridges in the United States. Currently, the bridge inspection is conducted manually, and it is not efficient to identify bridge condition deterioration in order to facilitate implementation of appropriate maintenance or rehabilitation procedures. In this paper, we report a new development of the autonomous mobile robotic system for bridge deck inspection and evaluation. The robot is integrated with several nondestructive evaluation (NDE) sensors and a navigation control algorithm to allow it to accurately and autonomously maneuver on the bridge deck to collect visual images and conduct NDE measurements. The developed robotic system can reduce the cost and time of the bridge deck data collection and inspection. For efficient bridge deck monitoring, the crack detection algorithm to build the deck crack map is presented in detail. The impact‐echo (IE), ultrasonic surface waves (USW), and electrical resistivity (ER) data collected by the robot are analyzed to generate the delamination, concrete elastic modulus, corrosion maps of the bridge deck, respectively. The presented robotic system has been successfully deployed to inspect numerous bridges in more than ten different states in the United States.     

Public transportation trip flow modeling with generalized regression neural networks

Artificial neural networks (ANNs) are one of the recently explored advanced technologies, which show promise in the area of transportation engineering. The presented study comprised the employment of this seldom used ANN method, generalized regression neural network (GRNN), in comparison to both a frequently applied neural network training algorithm, feed-forward back-propagation (FFBP), and a stochastic model of auto-regressive structure for the purpose of forecasting daily trip flows, which is an essential component in demand analysis. The study is carried out under the motivation of knowing that modeling daily trips for available transportation modes will facilitate the arrangement for effective public infrastructure investments and the cited papers in the literature did not make use of and handle any comparison with GRNN method. The ANN predictions are found to be quite close to the observations as reflected in the selected performance criteria. The selected stochastic model performance is quite poor compared with ANN results. It is seen that the GRNN did not provide negative forecasts in contrast to FFBP applications. Besides, the local minima problem faced by FFBP algorithm is not encountered in GRNNs.     

Generalized Regression Neural Networks and Feed Forward Neural Networks for prediction of scour depth around bridge piers

In this study, Generalized Regression Neural Networks (GRNN) and Feed Forward Neural Networks (FFNN) approaches are used to predict the scour depth around circular bridge piers. Hundred and sixty five data collected from various experimental studies, are used to predict equilibrium scour depth. The model consisting of the combination of dimensional data involving the input variables is constructed. The performance of the models in training and testing sets are compared with observations. Then, the model is also tested by Multiple Linear Regression (MLR) and empirical formula. The results of all approaches are compared in order to get more reliable comparison. The results indicated that GRNN can be applied successfully for prediction of scour depth around circular bridge piers.     

A satellite data set for tropical forest area change assessment

A database of largely cloud-free (less than 2.5% of all sites have more than 5% cloud cover), geo-referenced 20 km?×?20 km sample sites of 30 m resolution optical satellite imagery have been prepared for the 1990 and 2000 epochs. This spans the tropics with a systematic sample located at the degree confluence points of the geographic grid. The resulting 4016 sample pairs are to be used to measure changes in the area of forest cover between the two epochs. The primary data source was the National Aeronautics and Space Administration's (NASA's) global land survey (GLS) data sets. Visual screening of GLS images at all 4016 confluence points from each date identified 2868 suitable pairs where no better alternatives exist (71.6% of the sample). Better alternatives could be found for 26.6% of the sample, substituting cloudy or missing GLS data sets at one or the other epoch or both (GLS-1990 or GLS-2000). Gaps were filled from the United States Geological Survey (USGS) Landsat archives (1070 samples), data from other Landsat archives (53 samples) or with alternatives to Landsat, that is, 15 samples from Satellite Pour l'Observation de la Terre (SPOT). This increased the effective number of sample pairs to 3945 representing 98% of all target samples. No suitable image pairs were found for 71 confluence points, which were not randomly distributed, but mostly concentrated in the Congo basin, where around 15% of the region remains un-sampled. Variations in date of image acquisition and geometric fidelity are documented. Results highlight the importance of combining systematic data-processing schemes with targeted image acquisition and archiving strategies for global scale applications such as deforestation monitoring and shows that by replacing cloudy or missing GLS data with alternative imagery, the overall coverage of the sample sites within the ecological zones ‘Tropical rainforest’ and ‘Tropical mountain system’ can be improved by 16%.     

Contour line and geographic feature extraction from USGS color topographical paper maps

This paper presents a method for the extraction of contour lines and other geographic information from scanned color images of topographical maps. Although topographic maps are available from many suppliers, this work focuses on United States Geological Survey (USGS) maps. The extraction of contour lines, which are shown with brown color on USGS maps, is a difficult process due to aliasing and false colors induced by the scanning process and due to closely spaced and intersecting/overlapping features inherent to the map. These difficulties render simple approaches such as clustering ineffective. The proposed method overcomes these difficulties using a multistep process. First, a color key set, designed to comprehend color aliasing and false colors, is generated using an eigenvector line-fitting technique in RGB space. Next, area features, representing vegetation and bodies of water, are extracted using RGB color histogram analysis in order to simplify the next stage. Then, linear features corresponding to roads and rivers including contours, are extracted using a valley seeking algorithm operating on a transformed version of the original map. Finally, an A* search algorithm is used to link valleys together to form linear features and to close the gaps caused by intersecting features. The performance of the algorithm is tested on a number of USGS topographic map samples.     

Investigation of heavy-mineral deposits using multispectral satellite data

Mineral deposit mapping is essential for sustainable and eco-friendly exploitation of natural resources. The south Tamil Nadu coast of India is rich in minerals. Currently the beach sands are extracted for export entirely in raw form without any value addition. Due to unsustainable sand mining, there are negative environment impacts, which lead to various coastal hazards such as erosion, salinization and sea-water intrusion. In order to initiate the focus on mapping of mineral deposits along this area, standardized hyperspectral analysis has been carried out using Landsat satellite data and Environment for Visualising Images (ENVI) software. The selected endmembers are identified by comparing the spectral signatures with predefined spectral plots from the United States Geological Survey (USGS) spectral library. Finally the endmembers are mapped with ENVI's spectral angle mapper (SAM). The minerals which show significant variation in reflectance at different spectral bands can be effectively mapped by using multispectral data. Ground verifications performed to assess the accuracy of classification were mostly in agreement with the obtained results. This study has opened up new areas for inland heavy mineral exploitation and leads to eco-friendly exploitation of natural resources along the study area. It also illustrates the high potential of multispectral satellite data for exploration and mapping of mineral resources.     

Planimetric accuracy of Ikonos 1 m panchromatic orthoimage products and their utility for local government GIS basemap applications

A detailed assessment of the planimetric accuracy of high-resolution panchromatic Ikonos orthoimage products for three different test sites is presented. The main objective was to evaluate the potential of Ikonos orthoimage products for use as digital image basemaps in local government Geographic Information Systems (GIS) systems. For maximum utility and adoption, a planimetric accuracy of 3–5?m and a circular error at 90% probability (CE90) is considered nominal for local government planning and management applications. Low-precision, low-cost georeferenced Ikonos image products (Geo, 50?m CE90) were orthorectified using third party commercially available software and various custom Digital Elevation Models (DEMs). The planimetric accuracies of the resulting custom Ikonos orthoimages were found to vary between 2–4?m CE90. In addition, United States Geological Survey (USGS) DEMs were also used to orthorectify georeferenced Ikonos image products. Planimetric accuracies of 2–3?m CE90 were obtained from georeferenced Ikonos using USGS DEMs with RMS vertical accuracies of the order of 2?m. The approach demonstrated here can be used to deliver up-to-date, cost-effective orthoimages from Ikonos Geo products that yield planimetric accuracies suitable for use as digital image basemaps by local governments.