Convolutional LSTM Architecture for Precipitation Nowcasting Using Satellite Data

There has been a lot of research on flight delays. But it is more useful and difficult to estimate the departure delay time especially three hours before the scheduled time of departure, from which passengers can reasonably plan their travel time and the airline and airport staff can schedule flights more reasonably. In this paper, we develop a Spatio-temporal Graph Dual-Attention Neural Network (SGDAN) to learn the departure delay time for each flight with real-time conditions at three hours before the scheduled time of departure. Specifically, it first models the air traffic network as graph sequences, what is, using a heterogeneous graph to model a flight and its adjacent flights with the same departure or arrival airport in a special time interval, and using a sequence to model the flight and its previous flights that share the same aircraft. The main contributions of this paper are using heterogeneous graph-level attention to learn the influence between the flight and its adjacent flight together with sequence-level attention to learn the influence between the flight and its previous flight in the flight sequence. With aggregating features from the learned influence from both graph-level and sequence-level attention, SGDAN can generate node embedding to estimate the departure delay time. Experiments on a real-world large-scale data set show that SGDAN produces better results than state-of-the-art models in the accurate flight delay time estimation task.

Download Full-text

Airfoil buffet aerodynamics at plunge and pitch excitation based on long short-term memory neural network prediction

CEAS Aeronautical Journal ◽

10.1007/s13272-021-00550-6 ◽

2021 ◽

Author(s):

R. Zahn ◽

C. Breitsamter

Keyword(s):

Neural Network ◽

Short Term Memory ◽

Training Data ◽

Good Prediction ◽

Short Term ◽

Data Set ◽

Term Memory ◽

Identification Approach ◽

Long Short Term Memory ◽

Transonic Buffet

AbstractIn the present study, a nonlinear system identification approach based on a long short-term memory (LSTM) neural network is applied for the prediction of transonic buffet aerodynamics. The identification approach is applied as a reduced-order modeling (ROM) technique for an efficient computation of time-varying integral quantities such as aerodynamic force and moment coefficients. Therefore, the nonlinear identification procedure as well as the generalization of the ROM are presented. The training data set for the LSTM–ROM is provided by performing forced-motion unsteady Reynolds-averaged Navier–Stokes simulations. Subsequent to the training process, the ROM is applied for the computation of the aerodynamic integral quantities associated with transonic buffet. The performance of the trained ROM is demonstrated by computing the aerodynamic loads of the NACA0012 airfoil investigated at transonic freestream conditions. In contrast to previous studies considering only a pitching excitation, both the pitch and plunge degrees of freedom of the airfoil are individually and simultaneously excited by means of an user-defined training signal. Therefore, strong nonlinear effects are considered for the training of the ROM. By comparing the results with a full-order computational fluid dynamics solution, a good prediction capability of the presented ROM method is indicated. However, compared to the results of previous studies including only the airfoil pitching excitation, a slightly reduced prediction performance is shown.

Download Full-text

Nowcasting heavy precipitation over the Netherlands using a 13-year radar archive: a machine learning approach

10.5194/egusphere-egu21-12814 ◽

2021 ◽

Author(s):

Eva van der Kooij ◽

Marc Schleiss ◽

Riccardo Taormina ◽

Francesco Fioranelli ◽

Dorien Lugt ◽

...

Keyword(s):

Machine Learning ◽

The Netherlands ◽

Heavy Rainfall ◽

Predictive Performance ◽

Heavy Precipitation ◽

Early Warning Systems ◽

Training Data ◽

Short Term ◽

Data Set ◽

Radar Images

Accurate short-term forecasts, also known as nowcasts, of heavy precipitation are desirable for creating early warning systems for extreme weather and its consequences, e.g. urban flooding. In this research, we explore the use of machine learning for short-term prediction of heavy rainfall showers in the Netherlands.We assess the performance of a recurrent, convolutional neural network (TrajGRU) with lead times of 0 to 2 hours. The network is trained on a 13-year archive of radar images with 5-min temporal and 1-km spatial resolution from the precipitation radars of the Royal Netherlands Meteorological Institute (KNMI). We aim to train the model to predict the formation and dissipation of dynamic, heavy, localized rain events, a task for which traditional Lagrangian nowcasting methods still come up short.We report on different ways to optimize predictive performance for heavy rainfall intensities through several experiments. The large dataset available provides many possible configurations for training. To focus on heavy rainfall intensities, we use different subsets of this dataset through using different conditions for event selection and varying the ratio of light and heavy precipitation events present in the training data set and change the loss function used to train the model.To assess the performance of the model, we compare our method to current state-of-the-art Lagrangian nowcasting system from the pySTEPS library, like S-PROG, a deterministic approximation of an ensemble mean forecast. The results of the experiments are used to discuss the pros and cons of machine-learning based methods for precipitation nowcasting and possible ways to further increase performance.

Download Full-text

Development and Evaluation of a River-Basin-Scale High Spatio-Temporal Precipitation Data Set Using the WRF Model: A Case Study of the Heihe River Basin

Remote Sensing ◽

10.3390/rs70709230 ◽

2015 ◽

Vol 7 (7) ◽

pp. 9230-9252 ◽

Cited By ~ 13

Author(s):

Xiaoduo Pan ◽

Xin Li ◽

Guodong Cheng ◽

Hongyi Li ◽

Xiaobo He

Keyword(s):

River Basin ◽

Wrf Model ◽

Heihe River ◽

Precipitation Data ◽

Data Set ◽

River Basin Scale ◽

Basin Scale ◽

Spatio Temporal ◽

The Heihe River Basin

Download Full-text

Crack-Depth Determination by a Neural Network with a Synthetic Training Data Set

Review of Progress in Quantitative Nondestructive Evaluation ◽

10.1007/978-1-4615-2848-7_102 ◽

1993 ◽

pp. 803-810 ◽

Cited By ~ 1

Author(s):

M. Takadoya ◽

M. Notake ◽

M. Kitahara ◽

J. D. Achenbach ◽

Q. C. Guo ◽

...

Keyword(s):

Neural Network ◽

Crack Depth ◽

Training Data ◽

Data Set ◽

Depth Determination ◽

Synthetic Training Data

Download Full-text

Bootstrapping a Neural Morphological Generator from Morphological Analyzer Output for Inuktitut

10.33011/computel.v2i.455 ◽

2019 ◽

Vol 2 (1) ◽

Author(s):

Jeffrey Micher

Keyword(s):

Neural Network ◽

Training Data ◽

Data Set ◽

Set Size ◽

The Neural Network ◽

Surface Character ◽

Finite State ◽

Character Sequences ◽

Finite State Transducer

We present a method for building a morphological generator from the output of an existing analyzer for Inuktitut, in the absence of a two-way finite state transducer which would normally provide this functionality. We make use of a sequence to sequence neural network which “translates” underlying Inuktitut morpheme sequences into surface character sequences. The neural network uses only the previous and the following morphemes as context. We report a morpheme accuracy of approximately 86%. We are able to increase this accuracy slightly by passing deep morphemes directly to output for unknown morphemes. We do not see significant improvement when increasing training data set size, and postulate possible causes for this.

Download Full-text

Modeling of daily runoff from a small agricultural watershed using artificial neural network with resampling techniques

Journal of Hydroinformatics ◽

10.2166/hydro.2014.106 ◽

2014 ◽

Vol 17 (1) ◽

pp. 56-74 ◽

Cited By ~ 11

Author(s):

Gurjeet Singh ◽

Rabindra K. Panda ◽

Marc Lamers

Keyword(s):

Neural Network ◽

Artificial Neural Network ◽

Agricultural Watershed ◽

Training Data ◽

Optimum Number ◽

Training Process ◽

Data Set ◽

Reported Study ◽

Artificial Neural ◽

Artificial Neural Network Ann

The reported study was undertaken in a small agricultural watershed, namely, Kapgari in Eastern India having a drainage area of 973 ha. The watershed was subdivided into three sub-watersheds on the basis of drainage network and land topography. An attempt was made to relate the continuously monitored runoff data from the sub-watersheds and the whole-watershed with the rainfall and temperature data using the artificial neural network (ANN) technique. The reported study also evaluated the bias in the prediction of daily runoff with shorter length of training data set using different resampling techniques with the ANN modeling. A 10-fold cross-validation (CV) technique was used to find the optimum number of hidden neurons in the hidden layer and to avoid neural network over-fitting during the training process for shorter length of data. The results illustrated that the ANN models developed with shorter length of training data set avoid neural network over-fitting during the training process, using a 10-fold CV method. Moreover, the biasness was investigated using the bootstrap resampling technique based ANN (BANN) for short length of training data set. In comparison with the 10-fold CV technique, the BANN is more efficient in solving the problems of the over-fitting and under-fitting during training of models for shorter length of data set.

Download Full-text

Assessment of Alternative Ways to Integrate Weather Predictions in Photovoltaic Generation Forecasting.

10.5194/egusphere-egu21-16091 ◽

2021 ◽

Author(s):

Kevin Bellinguer ◽

Robin Girard ◽

Guillaume Bontron ◽

Georges Kariniotakis

Keyword(s):

Linear Model ◽

Relevant Information ◽

Training Data ◽

Short Term ◽

Data Set ◽

Photovoltaic Generation ◽

Weather Information ◽

Weather Situation ◽

Conditioned Learning ◽

Pv Generation

<div> In recent years, the share of photovoltaic (PV) power in Europe has grown: the installed capacity increased from around 10 GW in 2008 to nearly 119 GW in 2018 [1]. Due to the intermittent nature of PV generation, new challenges arise regarding economic profitability and the safe operation of the power network. To overcome these issues, a special effort is made to develop efficient PV generation forecasting tools. &#160; For short-term PV production forecasting, past production observations are typically the main drivers. In addition, spatio-temporal (ST) inputs such as Satellite-Derived Surface Irradiance (SDSI) provide relevant information regarding the weather situation in the vicinity of the farm. Moreover, the literature shows us that Numerical Weather Predictions (NWPs) provide relevant information regarding weather trends. &#160; NWPs can be integrated in the forecasting process in two different ways. The most straightforward approach considers NWPs as explanatory input variables to the forecasting models. Thus, the atmosphere dynamics are directly carried by the NWPs. The alternative considers NWPs as state variables: weather information is used to filter the training data set to obtain a coherent subset of PV production observations measured under similar weather conditions as the PV production to be predicted. This approach is based on analog methods and makes the weather dynamics to be implicitly contained in the PV production observations. This conditioned learning approach permits to perform local regressions and is adaptive in the sense that the model training is conditioned to the weather situation. The specialized literature focuses on spot NWPs which permits to find situations that evolve in the same way but does not preserve ST patterns. In this context, the addition of SDSI features cannot make the most of the conditioning process. Ref. [3] proposes to use geopotential fields, which are wind drivers, as analog predictors. &#160; In this work, we propose the following contributions to the state of the art: We investigate the influence of spot NWPs on the performances of an auto-regressive (AR) and a random forest models according to the two above-mentioned approaches: either as additional explanatory features and/or as analog features. The analogy score proposed by [2] is used to find similar weather situations, then the model is trained over the associated PV production observations. The results highlight that the linear model performs better with the conditioned approach while the non-linear model obtains better performances when fed with explanatory features. Then, the similarity score is extended to gridded NWPs data through the use of a principal component analysis. This method allows to condition the learning to large-scale weather information. A comparison between spot and gridded NWPs conditioned approaches applied with AR model highlights that gridded NWPs improves the influence of SDSI over forecasting performances. &#160; The proposed approaches are evaluated using 9 PV plants in France and for a testing period of 12 months. &#160; References [1]&#160;&#160;&#160;&#160;&#160; IRENA - https://www.irena.org/Statistics/Download-Data [2]&#160;&#160;&#160;&#160;&#160; Alessandrini, Delle Monache, et al. An analog ensemble for short-term probabilistic solar power forecast. Applied Energy, 2015. https://doi.org/10.1016/j.apenergy.2015.08.011 [3]&#160;&#160;&#160;&#160;&#160; Bellinguer, Girard, Bontron, Kariniotakis. Short-term Forecasting of Photovoltaic Generation based on Conditioned Learning of Geopotential Fields. 2020, UPEC. https://doi.org/10.1109/UPEC49904.2020.9209858 </div>

Download Full-text

A Neural Network for Modeling Multicategorical Parcel Use Change

Geographic Information Systems ◽

10.4018/978-1-4666-2038-4.ch078 ◽

2013 ◽

pp. 1297-1308

Author(s):

Kang Shou Lu ◽

John Morgan ◽

Jeffery Allen

Keyword(s):

Neural Network ◽

Land Use ◽

Multinomial Logistic Regression ◽

Slight Variation ◽

Ann Model ◽

Temporal Prediction ◽

Percentage Points ◽

Spatio Temporal ◽

Coastal County ◽

Artificial Neural Network Ann

This paper presents an artificial neural network (ANN) for modeling multicategorical land use changes. Compared to conventional statistical models and cellular automata models, ANNs have both the architecture appropriate for addressing complex problems and the power for spatio-temporal prediction. The model consists of two layers with multiple input and output units. Bayesian regularization was used for network training in order to select an optimal model that avoids over-fitting problem. When trained and applied to predict changes in parcel use in a coastal county from 1990 to 2008, the ANN model performed well as measured by high prediction accuracy (82.0-98.5%) and high Kappa coefficient (81.4-97.5%) with only slight variation across five different land use categories. ANN also outperformed the benchmark multinomial logistic regression by average 17.5 percentage points in categorical accuracy and by 9.2 percentage points in overall accuracy. The authors used the ANN model to predict future parcel use change from 2007 to 2030.

Download Full-text

A Neural Network for Modeling Multicategorical Parcel Use Change

Emerging Methods and Multidisciplinary Applications in Geospatial Research ◽

10.4018/978-1-4666-1951-7.ch008 ◽

2012 ◽

pp. 131-142

Author(s):

Kang Shou Lu ◽

John Morgan ◽

Jeffery Allen

Keyword(s):

Neural Network ◽

Land Use ◽

Multinomial Logistic Regression ◽

Slight Variation ◽

Ann Model ◽

Temporal Prediction ◽

Percentage Points ◽

Spatio Temporal ◽

Coastal County ◽

Artificial Neural Network Ann

This paper presents an artificial neural network (ANN) for modeling multicategorical land use changes. Compared to conventional statistical models and cellular automata models, ANNs have both the architecture appropriate for addressing complex problems and the power for spatio-temporal prediction. The model consists of two layers with multiple input and output units. Bayesian regularization was used for network training in order to select an optimal model that avoids over-fitting problem. When trained and applied to predict changes in parcel use in a coastal county from 1990 to 2008, the ANN model performed well as measured by high prediction accuracy (82.0-98.5%) and high Kappa coefficient (81.4-97.5%) with only slight variation across five different land use categories. ANN also outperformed the benchmark multinomial logistic regression by average 17.5 percentage points in categorical accuracy and by 9.2 percentage points in overall accuracy. The authors used the ANN model to predict future parcel use change from 2007 to 2030.

Download Full-text