Search

Air pollution poses significant risks to human health and the environment, necessitating effective air quality management strategies. This study presents a novel approach to air quality management by integrating an autoencoder (AE) with a convolutional neural network (CNN) algorithm in Tehran city of Iran. One of the primary and vital problems in deep learning is model complexity, and the complexity of a model is affected by data distribution, data complexity, and information volume. AE provide a helpful way to denoise input data and make building deep learning models much more efficient. The proposed methodology enables spatial modeling and risk mapping of six air pollutants, namely, particulate matter 2.5 (PM2.5), particulate matter 10 (PM10), sulfur dioxide (SO2), nitrogen dioxide (NO2), ozone (O3), and carbon monoxide (CO). For air pollution modelling, data from a spatial database containing the annual average of six pollutants from 2012 to 2022 was utilized. The model considered various parameters influencing air pollution: altitude, humidity, distance to industrial areas, NDVI (normalized difference vegetation index), population density, rainfall, distance to the street, temperature, traffic volume, wind direction, and wind speed. The risk map accuracy was assessed using the area under the receiver operating characteristic (ROC) curve for six pollutants. Among them, NO2, PM10, CO, PM2.5, O3, and SO2 exhibited the highest accuracy with values of 0.964, 0.95, 0.896, 0.878, 0.877, and 0.811, respectively, in the risk map generated by the CNN-AE model. The findings demonstrated the CNN-AE model's impressive precision when generating the pollution risk map. [ABSTRACT FROM AUTHOR]

Locations prone to landslides must be identified and mapped to prevent landslide-related damage and casualties. Machine learning approaches have proven effective for such tasks and have thus been widely applied. However, owing to the rapid development of data-driven approaches, deep learning methods that can exhibit enhanced prediction accuracies have not been fully evaluated. Several researchers have compared different methods without optimizing them, whereas others optimized a single method using different algorithms and compared them. In this study, the performances of different fully optimized methods for landslide susceptibility mapping within the landslide-prone Kermanshah province of Iran were compared. The models, i.e., convolutional neural networks (CNNs), deep neural networks (DNNs), and support vector machine (SVM) frameworks were developed using 14 conditioning factors and a landslide inventory containing 110 historical landslide points. The models were optimized to maximize the area under the receiver operating characteristic curve (AUC), while maintaining their stability. The results showed that the CNN (accuracy = 0.88, root mean square error (RMSE) = 0.37220, and AUC = 0.88) outperformed the DNN (accuracy = 0.79, RMSE = 0.40364, and AUC = 0.82) and SVM (accuracy = 0.80, RMSE = 0.42827, and AUC = 0.80) models using the same testing dataset. Moreover, the CNN model exhibiting the highest robustness among the three models, given its smallest AUC difference between the training and testing datasets. Notably, the dataset used in this study had a low spatial accuracy and limited sample points, and thus, the CNN approach can be considered useful for susceptibility assessment in other landslide-prone regions worldwide, particularly areas with poor data quality and quantity. The most important conditioning factors for all models were rainfall and the distances from roads and drainages. [ABSTRACT FROM AUTHOR]

Information about water resources is crucial for sustainable development, and this issue is considered to be one of the most important concerns worldwide due to rapid industrialization and population growth. Countries in the semiarid region of the western Asia, like Iran, are dependent on groundwater resources so access to these resources is vital. This study maps surface spring potential on the Nourabad–Koohdasht Plain of Iran using a deep-learning algorithm called convolutional neural network (CNN), and the result was compared to predictions made with five advanced data-mining models: logistic model tree (LMT), LMT hybridized with bagging (BA-LMT), LMT hybridized with dagging (DA-LMT), LMT hybridized with random subspace (RS-LMT), and LMT hybridized with AdaBoost (AB-LMT). Frequency ratio was used to assess the strengths of relationships of each subclass layer to groundwater presence and evidential belief function revealed their effects on model uncertainty. The locations of 2463 springs were determined and showed that the northern part of the plain has the highest groundwater potential based on the density of springs. The data representing each of the spring locations were used for prediction modeling. Receiver operating characteristic (ROC) and area under the ROC curve (AUC) were used to evaluate the strengths of the predictions produced by the models. The results show that CNN (AUC = 0.885) provided the best prediction of spring locations. AB-LMT (AUC = 0.877) was second best, and BA-LMT (AUC = 0.876), DA-LMT (AUC = 0.856), RS-LMT (AUC = 0.846), and the standalone LMT model (AUC = 0.827) followed in rank. It can be concluded that the hybrid LMT models increased the predictive strength of the standalone LMT model when used to predict spring locations. These hybrid modeling methods may be used to improve sustainable groundwater management in the study region and in other regions as well. [ABSTRACT FROM AUTHOR]

Coastal aquifer vulnerability assessment (CAVA) studies are essential for mitigating the effects of seawater intrusion (SWI) worldwide. In this research, the vulnerability of the coastal aquifer in the Lahijan region of northwest Iran was investigated. A vulnerability map (VM) was created applying hydrogeological parameters derived from the original GALDIT model (OGM). The significance of OGM parameters was assessed using the mean decrease accuracy (MDA) method, with the current state of SWI emerging as the most crucial factor for evaluating vulnerability. To optimize GALDIT weights, we introduced the biogeography-based optimization (BBO) and gray wolf optimization (GWO) techniques to obtain to hybrid OGM-BBO and OGM-GWO models, respectively. Despite considerable research focused on enhancing CAVA models, efforts to modify the weights and rates of OGM parameters by incorporating deep learning algorithms remain scarce. Hence, a convolutional neural network (CNN) algorithm was applied to produce the VM. The area under the receiver-operating characteristic curves for OGM-BBO, OGM-GWO, and VMCNN were 0.794, 0.835, and 0.982, respectively. According to the CNN-based VM, 41% of the aquifer displayed very high and high vulnerability to SWI, concentrated primarily along the coastline. Additionally, 32% of the aquifer exhibited very low and low vulnerability to SWI, predominantly in the southern and southwestern regions. The proposed model can be extended to evaluate the vulnerability of various coastal aquifers to SWI, thereby assisting land use planers and policymakers in identifying at-risk areas. Moreover, deep-learning-based approaches can help clarify the associations between aquifer vulnerability and contamination resulting from SWI. [ABSTRACT FROM AUTHOR]

Recently, the increasing prevalence of solar energy in power and energy systems around the world has dramatically increased the importance of accurately predicting solar irradiance. However, the lack of access to data in many regions and the privacy concerns that can arise when collecting and transmitting data from distributed points to a central server pose challenges to current predictive techniques. This study proposes a global solar radiation forecasting approach based on federated learning (FL) and convolutional neural network (CNN). In addition to maintaining input data privacy, the proposed procedure can also be used as a global supermodel. In this paper, data related to eight regions of Iran with different climatic features are considered as CNN input for network training in each client. To test the effectiveness of the global supermodel, data related to three new regions of Iran named Abadeh, Jarqavieh, and Arak are used. It can be seen that the global forecasting supermodel was able to forecast solar radiation for Abadeh, Jarqavieh, and Arak regions with 95%, 92%, and 90% accuracy coefficients, respectively. Finally, in a comparative scenario, various conventional machine learning and deep learning models are employed to forecast solar radiation in each of the study regions. The results of the above approaches are compared and evaluated with the results of the proposed FL-based method. The results show that, since no training data were available from regions of Abadeh, Jarqavieh, and Arak, the conventional methods were not able to forecast solar radiation in these regions. This evaluation confirms the high ability of the presented FL approach to make acceptable predictions while preserving privacy and eliminating model reliance on training data. [ABSTRACT FROM AUTHOR]

By accurate predicting of pipe bursts, it is possible to schedule pipe maintenance, rehabilitation and improve level of services in Water Distribution Networks (WDNs). In recent years, artificial intelligence techniques have been used vastly in pipe burst prediction. These techniques need big input dataset for deep learning but big dataset might not be available in old WDNs. This paper aimed to compare the performance of multivariate adaptive regression splines (MARS) and regressive convolution neural network with least square support vector regression (RCNN-LSSVR) for predicting of Pipe Burst Rate (PBR) with limited dataset. Pipes age, diameter, depth of installation, length, average and maximum hydraulic pressure are considered as effective parameters of PBR. Data collected from real-case study in Iran include 158 cases for polyethylene and 124 cases for asbestos cement pipes during 2012–2019. The results indicate that RCNN-LSSVR model has a great performance of PBR prediction on limited dataset. [ABSTRACT FROM AUTHOR]

Despite the importance of the prediction of land susceptibility to gully erosion, there is a lack of research studies adopting the deep‐learning approach. This study aimed to predict gully susceptibility hotspots using hybridized deep‐learning models and evaluate their efficiency. Field records of gully occurrences in a gully‐prone region, the Talwar watershed (6468 km2), eastern Kurdistan province, Iran, were used to generate a gully inventory dataset. A total of 14 geomorphometric, environmental, and topo‐hydrological gully drivers were selected as predictor variables. The hybridized models were developed using convolutional neural network (NNC) and metaheuristic procedures, including the gray wolf optimizer (GWO) and the imperialist competitive algorithm (ICA). The validity of the resulting outputs was investigated based on the area under the receiver operating characteristic (ROC) curve. Results revealed that the NNC‐GWO had the highest efficiency in the validation step (AUC = 97.2%), whereas the NNC‐ICA was the second‐best model (AUC = 95.1%). The standalone NNC model showed the lowest accuracy (AUC = 91.2%) in predicting gully susceptibility hotspots compared to NNC‐GWO and NNC‐ICA. Thus, both hybridized models had better predictive performance for identifying gully susceptibility in comparison with the standalone NNC model. Furthermore, according to the NNC‐GWO model, about 0.2% (1294.8 ha) and 0.05% (235.2 ha) of the study area were identified as high and very high gully susceptibility classes. In addition, the application of the standalone NNC led to an overestimation of the susceptibility degree for gully initiation. This study supports researchers efforts to increase the model's performance when working in the land degradation domain. [ABSTRACT FROM AUTHOR]

Improving the accuracy of photovoltaic (PV) power forecasting is crucial to ensure more effective use of energy resources. Improvements are especially important for regions for which historical solar radiation data do not exist. This paper proposes a cyber-secure forecasting model called federated deep learning (FDL) to forecast PV power generation in various regions across Iran. The training process in each client is done by a convolutional neural network (CNN). Then, a generalizable global supermodel is generated based on the features extracted in each client, which has the ability to generalize and forecast for regions where there is no training data. Preserve data privacy and ideal performance against cyber-attacks are prominent features of the proposed method. The use of the proposed method is illustrated with a case study for Iran. The proposed FDL network is designed with 9 clients and three different scenarios were developed to test the robustness of the suggested method. In the first scenario, the PV power generation forecasting is done via the proposed technique and other conventional methods. The performance accuracy (R 2) of the generated global supermodel in this scenario for PV power generation forecasting in the regions of Khomein, Meybod, Varzaneh, Taleghan, and Shiraz are obtained as 0.981, 0.989, 0.986, 0.983, and 0.987 respectively. However, it was observed that other conventional deep learning-based models such as CNN and long short-term memory were not able to provide any forecasting for these regions. The second scenario models the scaling attack as a specific pattern of false data injection attack, to evaluate the performance of forecasting models against the data integrity attack. In the third scenario, cyber-attack detection is performed based on data visualization and image processing procedures. The results presented in different scenarios emphasize the high accuracy and generalizability of the global cyber-secure supermodel in PV power generation forecasting in different regions of Iran. [ABSTRACT FROM AUTHOR]

Forest fires burn various natural ecosystems worldwide and can harm the environment and human life. Accordingly, real-time monitoring of this phenomenon and early decision-making warning is vital. Active remote sensing systems, such as Synthetic Aperture Radar (SAR) sensors, provide an excellent opportunity for burn mapping because they can penetrate through clouds and smoke day and night. In this study, the potential of Sentinel-1 SAR data was investigated by deploying a deep convolutional neural network (CNN) based framework to map burn progression dynamically, in both supervised and transfer manners. Accordingly, an optimized deep architecture was designated to use SAR data features with high sensitivity to map the burned areas. The proposed method includes three main steps: 1) extraction of SAR indices, 2) training deep CNN model with a limited number of scenes and 3) assessing the transferability of the CNN for estimating burn progression for any unseen scene. Sentinel-1 SAR indices of log-ratio, radar burn difference (RBD), and difference of dual-polarization SAR vegetation index (ΔDPSVI) were obtained to be fed to the CNN. To validate the efficiency of the proposed approach, two fire events, i.e. the Derazno fire in Iran (2021) and the Rossomanno-Grottascura-Bellia fire in Italy (2017), were considered. For the scenes including training samples, the proposed method improved the overall accuracies (OAs) of classical machine learning techniques (i.e. SVM and RF) significantly (more than 4%). However, the improvement was minor when compared to a CNN using only log-ratio as the input channel (log-ratio CNN). For the scenes without training samples (unseen dates), the investigated transferred model performed substantially better (3% higher OA) compared to the other machine learning methods and the log-ratio CNN. This finding approves that the obtained SAR indices boost the transferability of the CNN model for burn progression mapping at unseen scenes. [ABSTRACT FROM AUTHOR]

Climate change and its side effects on weather variables and water resources is one of the recent international concerns. This research aims to project future changes in weather variables due to climate change for different emission scenarios of RCP2.6, RCP4.5 and RCP8.5 to assess and compare climate change impact on river flow variation based on hybrid application of deep learning and LARS-WG6 in different climates of Iran (CSA: cold semi-arid; HT: humid temperate; CA: cold arid). For this purpose, best combination of daily rainfall and discharges data with different lags were selected as inputs in rainfall-runoff modeling using Convolutional Neural Networks (CNN) model. Predicted meteorological data by the LARS-WG6 in the future (2021-2040) were then used as inputs for the selected CNN model and consequently runoff was predicted. Obtained findings showed that projected rainfall changes in 2021-2040 compared to the base period in scenarios of RCP2.6, 4.5 and 8.5 will be +14, +11, +6% in CA, +8, +1, +2% in CSA and +3.7, -1, -2.6% in HT regions, respectively. Whereas, maximum discharge in CA climate in scenario RCP2.6 is projected to raise about 18%, in the HT climate shows a reduce of about 5% in RCP8.5 and in CSA will decrease by about 5% in RCP2.6. Considering the effect of climate as well as seasonal changes in each region on the amount and direction of the climate change impacts on meteorological and hydrologic variables changes, assessing climate change impact in each region is necessary to extract adaptation strategies in water resources management. [ABSTRACT FROM AUTHOR]

The prediction of rainfall is essential for monitoring droughts and floods. The purpose of this paper is to develop a deep learning model for predicting monthly rainfall. The new model is used to predict rainfall in the Kashan plain of Iran. This study combines a deep learning model with an artificial neural network (ANN) model to predict rainfall. In this study, a convolutional neural network (CONV) is used as a deep learning model. The paper also introduces a new activation function called E-Tanh to develop ANN models. The new model has two main advantages. The model automatically determines key features. In addition, the new activation function can enhance the precision of ANN models. Lagged rainfall values are inserted into the models to predict rainfall. This study uses a bat optimization algorithm to choose inputs. At the training level, the mean absolute percentage errors (MAPES) of CONV-ANN-ANN-E-Tanh, CONV, and ANN-E-Tanh were 0.5%, 1%, and 2%, respectively. At the testing level, the MAPEs of CONV-ANN -E-Tanh, CONV, and ANN-E-Tanh were 1%, 3%, and 4%, respectively. The E-Tanh performed better than other activation functions based on error function values. Also, the CONV-ANN-E-Tanh can reduce CPU time. Our results show that the new hybrid model is a reliable tool for simulating complex phenomena. [ABSTRACT FROM AUTHOR]

Monitoring the changes in forest cover has become an important tool for forest management due to its impact on climate change, desertification, soil erosion, and flooding. The Zagros Mountains forest-steppe is a temperate broadleaf and mixed forest ecoregion in Western Asia, and the present study analyzes the changes in Zagros forests from 1986 to 2019 using remote sensing techniques. Three Landsat images, acquired in 1986, 2002, and 2019, were used to map land cover classification in the study area. Classification kappa coefficients were 85.2, 94.1, and 94.7 for three images, respectively. Focusing on the class level, the results show that between 1986 and 2002, the primary cause of forest loss in the study area was the conversion of 7.8 thousand ha of oak forests to agricultural land. Since 2002, 0.8 thousand ha of oak forest cover have been changed to barrenlands. The decrease of Zagros forests is a main threat to the rich biodiversity found in the West of Iran. To avoid worsening soil erosion, restoration programs should prioritize planting patterns on rainfall and hill slopes, texturally dissimilar soils, and regeneration in forest gaps, particularly at high elevations and steep slopes. It is recommended that future research should use higher resolution datasets and encompass more expansive landscape units, such as the whole Zagros Mountain forests. [ABSTRACT FROM AUTHOR]

Snow avalanches impose a considerable threat to infrastructure and human safety in snow bound mountain areas. Nevertheless, the spatial prediction of snow avalanches has received little research attention in many vulnerable parts of the world, particularly in developing countries. The present study investigates the applicability of a stand alone convolutional neural network (CNN) model, as a deep learning approach, along with two metaheuristic algorithms including grey wolf optimization (CNN-GWO) and imperialist competitive algorithm (CNN-ICA) in snow avalanche modelling in the Darvan watershed, Iran. The analysis was based on thirteen potential drivers of avalanche occurrence and an inventory map of previously documented avalanche occurrences. The efficiency of models' performance was evaluated by Area Under the Receiver Operating Characteristic curve (AUC) and the Root Mean Square Error (RMSE). The CNN-ICA model yielded the highest accuracy in both training (AUC = 0.982, RMSE = 0.067) and validation (AUC = 0.972, RMSE = 0.125) steps, followed by the CNN-GWO model (AUC of 0.975 for training, RMSE of 0.18 for training, AUC of 0.968 for validation, RMSE of 0.157 for validation). However, the standalone CNN model showed lower goodness-of- fit (AUC = 0.864, RMSE = 0.22) and predictive performance (AUC = 0.811, RMSE = 0.330). The approach utilized in this study is broadly applicable for identifying areas where avalanche hazard is likely to be high and where mitigation measures or corresponding land use planning should be prioritized. [ABSTRACT FROM AUTHOR]

This research paper reports the process and results of a project to automatically classify historical and non-historical buildings using airborne and satellite imagery. The case study area is the center of Yazd, the most important historical site in Iran. New computational scientific methods and accessibility to satellite images have created more opportunities to work on automated historical architecture feature recognition. Building on this, a convolutional neural network (CNN) is the main method for the classification task of the project. The most distinctive features of the historical houses in Iran are central courtyards. Based on this characteristic, the objective of the research is recognizing and labeling the houses as historical buildings by a CNN model. As a result, the trained model is tested by a validation dataset and has an accuracy rate of around 98%. In Sum, the reported project is one of the first works on deep learning methods in historical Iranian architecture study and one of the first efforts to use automated remote sensing techniques for recognizing historical courtyard houses in aerial images. [ABSTRACT FROM AUTHOR]

Landslides are one of the most destructive natural phenomena in the world, which occur mostly in mountainous areas and cause damage to the economic sectors, agricultural lands, residential areas and infrastructures of any country, and also threaten the lives and property of human beings. Therefore, landslide susceptibility mapping (LSM) can play a critical role in identifying prone areas and reducing the damage caused by landslides in each area. In the present study, deep learning algorithms including convolutional neural network (CNN) and long short-term memory (LSTM) were used to identify landslide prone areas in Ardabil province, Iran. Then 312 landslide locations were identified and randomly divided into train and test datasets, and according to previous studies and environmental conditions in the study area, twelve factors affecting the occurrence of landslides were selected. The ratio of the importance of each influential factor in landslide occurrence was obtained through information gain ranking filter method and it was found that land-use and profile curvature had the highest and lowest impacts, respectively. Afterwards, LSMs were generated using CNN and LSTM algorithms. In the next step, the performance of the models was evaluated based on the area under curve (AUC) value of receiver operating characteristics curve and the root mean square error (RMSE) method. The AUC values for CNN and LSTM models were 0.821 and 0.832, respectively. Furthermore, the RMSE values in the CNN model for each of the training and testing dataset were 0.121 and 0.132, respectively. The RMSE values in the LSTM model for each of the training and testing dataset were 0.185 and 0.188, respectively. Therefore, it can be concluded that LSTM performance is slightly better than CNN; but in general, both models have close performance and the accuracy of both models is acceptable. [ABSTRACT FROM AUTHOR]

Rice is one of the most essential and strategic food sources globally. Accordingly, policymakers and planners often consider a special place in the agricultural economy and economic development for this essential commodity. Typically, a sample survey is carried out through field observations and farmers' consultations to estimate annual rice yield. Studies show that these methods lead to many errors and are time-consuming and costly. Satellite remote sensing imagery is widely used in agriculture to provide timely, high-resolution data and analytical capabilities. Earth observations with high spatial and temporal resolution have provided an excellent opportunity for monitoring and mapping crop fields. This study used the time series of dual-pol synthetic aperture radar (SAR) images of Sentinel-1 and multispectral Sentinel-2 images from Sentinel-1 and Sentinel-2 ESA's Copernicus program to extract rice cultivation areas in Mazandaran province in Iran. A novel multi-channel streams deep feature extraction method was proposed to simultaneously take advantage of SAR and optical imagery. The proposed framework extracts deep features from the time series of NDVI and original SAR images by first and second streams. In contrast, the third stream integrates them into multi-levels (shallow to deep high-level features); it extracts deep features from the channel attention module (CAM), and group dilated convolution. The efficiency of the proposed method was assessed on approximately 129,000 in-situ samples and compared to other state-of-the-art methods. The results showed that combining NDVI time series and SAR data can significantly improve rice-type mapping. Moreover, the proposed methods had high efficiency compared with other methods, with more than 97% overall accuracy. The performance of rice-type mapping based on only time-series SAR images was better than only time-series NDVI datasets. Moreover, the classification performance of the proposed framework in mapping the Shirodi rice type was better than that of the Tarom type. [ABSTRACT FROM AUTHOR]

Predicting tides and water levels had always been such an important topic for researchers and professionals since the study of tidal level has pivotal role in supporting marine economy, port construction projects and maritime transportation. Tidal water levels are a combination of astronomical (deterministic part) and non-astronomical (stochastic part) water levels. In this study, we combined Harmonic Analysis (HA) with three Deep Neural Networks (DNNs), namely the Long-Short Term Memory (LSTM), Convolution Neural Network (CNN), and Multilayer Perceptron (MLP). The HA method is used for predicting the astronomical components, while DNNs are used to predict the non-astronomical water level. We have used tide gauge data from three stations along the southern coastline of Iran to demonstrate the effectiveness and accuracy of our model. We utilized RMSE, MAE, R2 (r-squared), and MAPE to evaluate the performance of the model. Finally, The LSTM network shown superior performance in most of the cases, although other networks also show good results. All three DNNs have R2 of 0.99, and the RMSE, MAE, and MAPE indicate that errors are low. [ABSTRACT FROM AUTHOR]

This research introduces a new combined modelling approach for mapping soil salinity in the Minab plain in southern Iran. This study assessed the uncertainty (with 95% confidence limits) and interpretability of two deep learning (DL) models (deep boltzmann machine—DBM) and a one dimensional convolutional neural networks (1DCNN)—long short-term memory (LSTM) hybrid model (1DCNN-LSTM) for mapping soil salinity by applying DeepQuantreg and game theory (Shapely Additive exPlanations (SHAP) and permutation feature importance measure (PFIM)), respectively. Based on stepwise forward regression (SFR)—a technique for controlling factor selection, 18 of 47 potential controls were selected as effective factors. Inventory maps of soil salinity were generated based on 476 surface soil samples collected for measuring electrical conductivity (ECe). Based on Taylor diagrams, both DL models performed well (RMSE < 20%), but the 1DCNN-LSTM hybrid model performed slightly better than the DBM model. The uncertainty range associated with the ECe values predicted by both models estimated using DeepQuantilreg were similar (0–25 dS/m for the 1DCNN-LSTM hybrid model and 2–27 dS/m for DBM model). Based on the SFR and PFIM (permutation feature importance measure)—a measure in game theory, four controls (evaporation, sand content, precipitation and vertical distance to channel) were selected as the most important factors for soil salinity in the study area. The results of SHAP (Shapely Additive exPlanations)—the second measure used in game theory—suggested that five factors (evaporation, vertical distance to channel, sand content, cation exchange capacity (CEC) and digital elevation model (DEM)) have the strongest impact on model outputs. Overall, the methodology used in this study is recommend for applications in other regions for mapping environmental problems. [ABSTRACT FROM AUTHOR]

Despite extensive research, recognition of Persian and Arabic manuscripts is still a challenging problem due to the complicated and irregular nature of writing, wide vocabulary, and diversity of handwritings. In Persian and Arabic words, letters are joined together, and signs such as dots are placed above or below letters. In the proposed approach, the words are first decomposed into their constituent subwords to enhance the recognition accuracy. Then the signs of subwords are extracted to develop a dictionary of main subwords and signs. The dictionary is then employed to train a classifier. Since the proposed recognition approach is based on unsigned subwords, the classifier may make a mistake in recognizing some subwords of a word. To overcome this, a new subword fusion algorithm is proposed based on the similarity of the main subwords and signs. Here, convolutional neural networks (CNNs) are utilized to train the classifier. An autoencoder (AE) network is employed to extract appropriate features. Thus, a hybrid network is developed and named AECNN. The known Iranshahr dataset, including nearly 17000 images of handwritten names of 503 cities of Iran, was employed to analyze and test the proposed approach. The resultant recognition accuracy is 91.09%. Therefore, the proposed approach is much more capable than the other methods known in the literature. [ABSTRACT FROM AUTHOR]

Copyright of Iranian Journal of Information Processing & Management is the property of Iranian Information & Documentation Center (IRANDOC) and its content may not be copied or emailed to multiple sites or posted to a listserv without the copyright holder's express written permission. However, users may print, download, or email articles for individual use. This abstract may be abridged. No warranty is given about the accuracy of the copy. Users should refer to the original published version of the material for the full abstract. (Copyright applies to all Abstracts.)

This study performs a GIS-based landslide susceptibility assessment using a convolutional neural network, CNN, in a study area of the Gorzineh-khil region, northeastern Iran. For this assessment, a 15-layered CNN was programmed in the Python high-level language for susceptibility mapping. In this regard, as far as the landside triggering factors are concerned, it was concluded that the geomorphologic/topographic parameters (i.e., slope curvature, topographical elevation, slope aspect, and weathering) and water condition parameters (hydrological gradient, drainage pattern, and flow gradient) are the main triggering factors. These factors provided the landside dataset, which was input to the CNN. We used 80% of the dataset for training and the remaining 20% for testing to prepare the landslide susceptibility map of the study area. In order to cross-validate the resulting map, a loss function, and common classifiers were considered: support vector machines, SVM, k-nearest neighbor, k-NN, and decision tree, DT. An evaluation of the results of the susceptibility assessment revealed that the CNN led the other classes in terms of 79.0% accuracy, 73.0% precision, 75.0% recall, and 77.0% f1-score, and, hence, provided better accuracy and the least computational error when compared to the other models. [ABSTRACT FROM AUTHOR]

Floods, as one of the natural hazards, can affect the environment, damage the infrastructures, and threaten human lives. Due to climate change and anthropogenic activities, floods occur in high frequency all over the world. Therefore, mapping of the flood areas is of prime importance in disaster management. This research presents a novel framework for flood area mapping based on heterogeneous remote sensing (RS) datasets. The proposed framework fuses the synthetic aperture radar (SAR), optical, and altimetry datasets for mapping flood areas, and it is applied in three main steps: (1) preprocessing, (2) deep feature extraction based on multiscale residual kernel convolution and convolution neural network's (CNN) parameter optimization by fusing the datasets, and (3) flood detection based on the trained model. This research exploits two large-scale area datasets for mapping the flooded areas in Golestan and Khuzestan provinces, Iran. The results show that the proposed methodology has a high performance in flood area detection. The visual and numerical analyses verify the effectiveness and ability of the proposed method to detect the flood areas with an overall accuracy (OA) higher than 98% in both study areas. Finally, the efficiency of the designed architecture was verified by hybrid-CNN and 3D-CNN methods. [ABSTRACT FROM AUTHOR]

Lake Urmia is located in the northwest of Iran and shared between the provinces of West Azarbaijan and East Azarbaijan. In the last two decades, there has been a considerable decline in the lake's water level. Satellite altimetry (SA) together with the advanced precise orbital positioning system has reached a high accuracy in the measurement of the water level height, but increasing the accuracy of waveform retracking (WR) is a challenging issue. In this study, wavelet decomposition and convolutional neural network were used for the WR with 50%, 55%, and 60% training scenarios and the threshold method was used for the 1992–2019 period. The training of 55% has the best result with a ± 0.027 m root mean square error. The water level has decreased by approximately 7 m from 1994 to 2018 and its overall trend is downward. The proposed method has been able to increase the WR accuracy by up to 30%. The gravity recovery and climate experiment and the annual monitoring of the water level station have also been used for the SA verification, which have a significant correlation of 0.66 and 0.96 with SA, respectively. [ABSTRACT FROM AUTHOR]

From a watershed management perspective, streamflow need to be predicted accurately using simple, reliable, and cost-effective tools. Present study demonstrates the first applications of a novel optimized deep-learning algorithm of a convolutional neural network (CNN) using BAT metaheuristic algorithm (i.e., CNN-BAT). Using the prediction powers of 4 well-known algorithms as benchmarks – multilayer perceptron (MLP-BAT), adaptive neuro-fuzzy inference system (ANFIS-BAT), support vector regression (SVR-BAT) and random forest (RF-BAT), the CNN-BAT model is tested for daily streamflow (Qt) prediction in the Korkorsar catchment in northern Iran. Fifteen years of daily rainfall (Rt) and streamflow data from 1997 to 2012 were collected and used for model development and evaluation. The dataset was divided into two groups for building and testing models. The correlation coefficient (r) between rainfall and streamflow with and without antecedent events (i.e., Rt-1, Rt-2, etc.) (as the input variables) and Qt (as the output variable) served as the basis for constructing different input scenarios. Several quantitative and visually-based evaluation metrics were used to validate and compare the model's performance. The results indicate that Rt was the most effective input variable on Qt prediction and the integration of Rt, Rt-1, and Qt-1 was the optimal input combination. The evaluation metrics show that the CNN-BAT algorithm outperforms the other algorithms. The Friedman and Wilcoxon signed-rank test indicates that the prediction power of CNN-BAT algorithm is significantly/statistically different from the other developed algorithms. [ABSTRACT FROM AUTHOR]

This paper introduces an improved convolutional neural network based on the conventional U-Net for simulating built-up land expansion. The proposed method hires a pixel-wise semantic segmentation approach considering the spatial drivers affecting urbanization as data cubes. Independent variables including altitude, slope, and distance from barren, crop, greenery, roads, and urban areas for 1998, 2008, and 2018 were considered as covariates for the simulation of built-up land expansion in Tehran and Karaj regions in Iran. The proposed method was compared with the random forest (RF) algorithm as the baseline model. Evaluation using the area under the total operating characteristic indicated the superiority of our modified U-Net (0.87) over the RF (0.82) algorithm. Furthermore, evaluation using the percent correct metric indicated that our proposed model is capable of learning neighborhood effects effectively leading to simulate built-up land expansion accurately, independent from applying a cellular automata (CA) model. Therefore, the modified U-Net independent from the CA which can consider the neighborhood effects is recommended for the simulation of built-up land expansion precisely. [ABSTRACT FROM AUTHOR]

Background: Artificial intelligence (AI) models have provided advanced applications to many scientific areas, including the prediction of the pathologic grade of tumors, utilizing radiology techniques. Gliomas are among the malignant brain tumors in human adults, and their efficient diagnosis is of high clinical significance. Objectives: Given the contribution of AI to medical diagnoses, we investigated the role of deep learning in the differential diagnosis and grading of human brain gliomas. Methods: This study developed a new AI diagnostic model, i.e., EfficientNetB0, to grade and classify human brain gliomas, using sequences from magnetic resonance imaging (MRI). Results: We validated the newAI model, using a standard dataset (BraTS-2019) and demonstrated that the AI components, i.e., convo-lutional neural networks and transfer learning, provided excellent performance for classifying and grading glioma images at 98.8% accuracy. Conclusions: The proposed model, EfficientNetB0, is capable to classify and grade glioma from MRI sequences at high accuracy, validity, and specificity. It can provide better performance and diagnostic results for human glioma images than models developed by previous studies. [ABSTRACT FROM AUTHOR]

Landslides are considered as one of the most devastating natural hazards in Iran, causing extensive damage and loss of life. Landslide susceptibility maps for landslide prone areas can be used to plan for and mitigate the consequences of catastrophic landsliding events. Here, we developed a deep convolutional neural network (CNN–DNN) for mapping landslide susceptibility, and evaluated it on the Isfahan province, Iran, which has not previously been assessed on such a scale. The proposed model was trained and validated using training (80%) and testing (20%) datasets, each containing relevant data on historical landslides, field records and remote sensing images, and a range of geomorphological, geological, environmental and human activity factors as covariates. The CNN–DNN model prediction accuracy was tested using a wide range of statistics from the confusion matrix and error indices from the receiver operating characteristic (ROC) curve. The CNN–DNN model was evaluated comprehensively by comparing it to several state-of-the-art benchmark machine learning techniques including the support vector machine (SVM), logistic regression (LR), Gaussian naïve Bayes (GNB), multilayer perceptron (MLP), Bernoulli Naïve Bayes (BNB) and decision tree (DT) classifiers. The CNN–DNN model for landslide susceptibility mapping was found to predict more accurately than the benchmark algorithms, with an AUC = 90.9%, IRs = 84.8%, MSE = 0.17, RMSE = 0.40, and MAPE = 0.42. The map provided by the CNN–DNN clearly revealed a high-susceptibility area in the west and southwest, related to the main Zagros trend in the province. These findings can be of great utility for landslide risk management and land use planning in the Isfahan province. [ABSTRACT FROM AUTHOR]

Prostate cancer is one of the most common and fatal diseases among men, and its early diagnosis can have a significant impact on the treatment process and prevent mortality. Since it does not have apparent clinical symptoms in the early stages, it is difficult to diagnose. In addition, the disagreement of experts in the analysis of magnetic resonance images is also a significant challenge. In recent years, various research has shown that deep learning, especially convolutional neural networks, has appeared successfully in machine vision (especially in medical image analysis). In this research, a deep learning approach was used on multi-parameter magnetic resonance images, and the synergistic effect of clinical and pathological data on the accuracy of the model was investigated. The data were collected from Trita Hospital in Tehran, which included 343 patients (data augmentation and learning transfer methods were used during the process). In the designed model, four different types of images are analyzed with four separate ResNet50 deep convolutional networks, and their extracted features are transferred to a fully connected neural network and combined with clinical and pathological features. In the model without clinical and pathological data, the maximum accuracy reached 88%, but by adding these data, the accuracy increased to 96%, which shows the significant impact of clinical and pathological data on the accuracy of diagnosis., (© 2024. The Author(s).)

Automated snake image identification is important from different points of view, most importantly, snake bite management. Auto-identification of snake images might help the avoidance of venomous snakes and also providing better treatment for patients. In this study, for the first time, it's been attempted to compare the accuracy of a series of state-of-the-art machine learning methods, ranging from the holistic to neural network algorithms. The study is performed on six snake species in Lar National Park, Tehran Province, Iran. In this research, the holistic methods [k-nearest neighbors (kNN), support vector machine (SVM) and logistic regression (LR)] are used in combination with a dimension reduction approach [principle component analysis (PCA) and linear discriminant analysis (LDA)] as the feature extractor. In holistic methods (kNN, SVM, LR), the classifier in combination with PCA does not yield an accuracy of more than 50%, But the use of LDA to extract the important features significantly improves the performance of the classifier. A combination of LDA and SVM (kernel = 'rbf') is achieved to a test accuracy of 84%. Compared to holistic methods, convolutional neural networks show similar to better performance, and accuracy reaches 93.16% using MobileNetV2. Visualizing intermediate activation layers in VGG model reveals that just in deep activation layers, the color pattern and the shape of the snake contribute to the discrimination of snake species. This study presents MobileNetV2 as a powerful deep convolutional neural network algorithm for snake image classification that could be used even on mobile devices. This finding pave the road for generating mobile applications for snake image identification. [ABSTRACT FROM AUTHOR]

Understanding the spatial distribution of soil salinity is required to conserve land against degradation and desertification. Against this background, this study is the first attempt to predict soil salinity in the Jaghin basin, in southern Iran, by applying and comparing the performance of four deep learning (DL) models (deep convolutional neural networks—DCNNs, dense connected deep neural networks—DenseDNNs, recurrent neural networks-long short-term memory—RNN-LSTM and recurrent neural networks-gated recurrent unit—RNN-GRU) and six shallow machine learning (ML) models (bagged classification and regression tree—BCART, cforest, cubist, quantile regression with LASSO penalty—QR-LASSO, ridge regression—RR and support vectore machine—SVM). To do this, 49 environmental landsat8-derived variables including digital elevation model (DEM)-extracted covariates, soil-salinity indices, and other variables (e.g., soil order, lithology, land use) were mapped spatially. For assessing the relationships between soil salinity (EC) and factors controlling EC, we collected 319 surficial (0–5 cm depth) soil samples for measuring soil salinity on the basis of electrical conductivity (EC). We then selected the most important features (covariates) controlling soil salinity by applying a MARS model. The performance of the DL and shallow ML models for generating soil salinity spatial maps (SSSMs) was assessed using a Taylor diagram and the Nash Sutcliff coefficient (NSE). Among all 10 predictive models, DL models with NSE ≥ 0.9 (DCNNs was the most accurate model with NSE = 0.96) were selected as the four best models, and performed better than the six shallow ML models with NSE ≤ 0.83 (QR-LASSO was the weakest predictive model with NSE = 0.50). Based on DCNNs-, the values of the EC ranged between 0.67 and 14.73 dS/m, whereas for QR-LASSO the corresponding EC values were 0.37 to 19.6 dS/m. Overall, DL models performed better than shallow ML models for production of the SSSMs and therefore we recommend applying DL models for prediction purposes in environmental sciences. [ABSTRACT FROM AUTHOR]

Plant pests and diseases are a major threat to human food security. In large farms, accurate and timely human diagnosis is not possible due to time consuming and possible misdiagnosis. Therefore, for immediate, automatic, appropriate and accurate detection of agricultural pests, the use of image processing and artificial intelligence, including deep learning can be very useful. In this study, convolutional neural network models have been developed to identify three common citrus pests in northern Iran such as citrus leafminer, sooty mold and pulvinaria using images of infected leaves, through deep learning methods. For this purpose, Resnet50 and VGG16 architectures are trained as well-known convolutional neural networks, applying the transfer learning method on 1774 images of infected citrus leaves, accumulated from natural and field conditions. In the training phase, data augmentation is used to increase the number of training samples, and to improve the generalizability of the classifiers. For experimental analysis, cross validation strategy is used to evaluate the accuracy of the convolutional neural network. In this strategy, all images are tested without any overlap between training and test sets. Based on the results, the accuracies of Resnet 50 and VGG 16 models are evaluated as 96.05 and 89.34%, respectively. Hence, the Resnet 50 model can convert the above method into a very suitable early consulting or warning system. [ABSTRACT FROM AUTHOR]

• A modern scheme for multi-temporal forecasting of short-term SPEI indices. • Coupling a new multi-processing empirical Fourier-based with ensemble DeepRVFL. • Accurate determination of antecedent information using SelectKBest to feed models. • Potential assessment of EDF-DeepRVFL model in very-hot and semi-humid climate zones. • Superiority of the DeepRVFL to the CNN-Bi-RNN and RVFL in the hybrid framework. The development of advanced technologies based on computer aid models in the domain of crops and agriculture productively is a modern advancement. Machine learning (ML) based forecasting of the short-term drought indicators (e.g., Standardized Precipitation Evapotranspiration Index (SPEI)) based on individual signals is a complex process that involves several factors including data quality and availability, inherent signal complexity, non-stationarity of climate, and uncertainty in ML models. Recent achievements in the field of Fourier-based signal processing integrated with advanced deep learning approaches have made it possible to produce very accurate intelligent frameworks for multi-temporal drought indicators and fill this gap. In this research, a new ultramodern complementary intelligent framework comprised of the SelectKbest feature selection (FS), Empirical Fourier Decomposition (EFD), and deep ensemble random vector functional link (Deep RVFL) was developed for multi-temporal monthly forecasting of short-term drought indicators for three and six months (SPEI3 and SPEI6) for two different very hot and semi-humid climate zones of Iran.. For this purpose, the most influential time lags associated with each drought indicator were indicated using the SelectKbest FS in each zone. Afterwards, the individual SPEI signals were decomposed by the EFD technique imposing the most important lagged components to feed the ML approaches. Here, a new hybrid architecture deep learning model, namely a convolutional neural network coupled with a bidirectional gated recurrent unit (CNN-Bi-GRU) designed for multi-temporal drought forecasting in the next one-and three- months. Two advanced approaches, introducing the convolutional neural network coupled with bidirectional recurrent neural network (CNN-Bi-RNN), and Random vector function link (RVFL) were adopted to validate the main model (EFD-DeepRVFL) in complementary and standalone counterpart forms. The robustness of all the models was examined using several metrics such as coefficient of determination (R2), root mean square error (RMSE), reliability, and squared Chi-square Distance (SquD). The comprehensive assessment of the outcomes of hybrid schemes revealed that EFD-DeepRVFL owing to superior performance (R2|SPEI3(t + 1) = 0.953, R2|SPEI3(t + 3) = 0.837, R2|SPEI6(t + 1) = 0.962, and R2|SPEI6(t + 3) = 0.887) at Ahvaz station and (R2|SPEI3(t + 1) = 0.964, R2|SPEI3(t + 3) = 0.863, R2|SPEI6(t + 1) = 0.935, and R2|SPEI6(t + 3) = 0.839) at Kermanshah station outperformed the EFD-RVFL and CNN-Bi-RNN, respectively. The developed expert system provides early warning of drought conditions, as a decision-making tool, crop yield prediction, and water resources risk assessment. [ABSTRACT FROM AUTHOR]

One of the most important features of a reliable power system is its capability to supply the demand continuously. This continuous supply has been maintained by the transient stability of the system against large disturbances. The study of this type of stability is examined through different indicators. One of the common indicators to evaluate the transient stability of the power system is the well-known Critical Clearing Time (CCT) index. Conventional methods for calculating CCT have presented good accuracy, however, their computational cost is very high which makes them not suitable for real-time applications and real large-scale networks. Considering their ability to feature extraction of big data, deep neural networks can be utilized as reliable tools to cover these deficiencies. In this regard, to cover the shortcomings of the conventional methods, this paper proposes a method based on deep Convolutional Neural Networks (CNN) to estimate the CCT index in real-time power system applications. Moreover, to analyze a realistic case, nonlinear limiters of the excitation systems which have a considerable effect on transient stability index are considered in this study. Thanks to the using of several deep layers and the comprehensive established database, the accuracy of proposed method is appropriately high. Numerical studies on IEEE standard networks as well as a real case in Iran Power Grid (IPG) represents the advantages of the proposed method. • A deep-learning-based convolutional neural network is designed to estimate CCT index of power system. • Nonlinear limiters of the excitation system are taken into account. • A real case study in Iran grid is considered for simulation. • As a result, the transient stability of system is accurately determined in on-line manner. [ABSTRACT FROM AUTHOR]

This study examined coastal aquifer vulnerability to seawater intrusion (SWI) in the Shiramin area in northwest Iran. Here, six types of hydrogeological data layers existing in the traditional GALDIT framework (TGF) were used to build one vulnerability map. Moreover, a modified traditional GALDIT framework (mod-TGF) was prepared by eliminating the data layer of aquifer type from the GALDIT model and adding the data layers of aquifer media and well density. To the best of our knowledge, there is a research gap to improve the TGF using deep learning algorithms. Therefore, this research adopted the Convolutional Neural Network (CNN) as a new deep learning algorithm to improve the mod-TGF framework for assessing the coastal aquifer vulnerability. Based on the findings, the CNN model could increase the performance of the mod-TGF by >30 %. This research can be a reference for further aquifer vulnerability studies. [Display omitted] • Traditional GALDIT framework (TGF) were used to build one vulnerability map. • A modified TGF vulnerability index was prepared by removing/adding data layer • Convolutional neural network (CNN) were applied to improve the mod-TGF model • CNN precisely identify the relation between aquifer vulnerability and contaminations • The findings of this study are useful to comprehend the coastal aquifer vulnerability situation [ABSTRACT FROM AUTHOR]

Awareness of uniaxial compressive strength (UCS) as a key rock formation parameter for the design and development of gas and oil field plays. It plays an essential role in the selection of the drill bits and stability of the wellbore's wall. Precise prediction of UCS before or during the drilling, especially in exploration wellbores, is necessary to improve the drilling speed and reduce the instability of the wellbore walls. UCS predictor machine-learning (ML) models are developed in this study using drilling parameters recorded during drilling using least-squares support-vector machine (LSSVM) and multi-layer extreme learning machine (MELM) algorithms hybridized with cuckoo optimization algorithm (COA), particle swarm optimization (PSO) and genetic algorithm (GA) optimizers. In addition, stand-alone LSSVM and convolutional neural network (CNN) models without optimizer enhancements are evaluated. Drilling and petrophysical data recorded for two wells (A and B) from the Rag-e-Safid oil field in southwest Iran were compiled to form the studied dataset. UCS was initially calculated numerically based on data from laboratory tests from petrophysical logs. The Well A dataset was pre-processed to remove outlying data records by applying the quantile regression algorithm. That analysis indicated that 9 data records should be removed from the Well A dataset. A decision tree model was employed for feature selection purposes to identify the more influential variables with respect to UCS. Depth, weight on the drill bit (WOB), drill-string rotation speed (RPM), rate of penetration (ROP), and torque (Trq) were the variables identified as being highly influential on UCS values. Application of the ML models on the training data subset (75% of Well A data records) revealed that the MELM-COA algorithm achieved the lowest root mean squared error (4.6945 MPa) and a higher coefficient of determination (0.9873) value than the other models when predicting UCS in the Well A training and validation data subsets. The Well-A-trained MELM-COA model confirmed its generalizability within the studied field by generating low UCS prediction errors when applied to the independent Well B testing dataset. [ABSTRACT FROM AUTHOR]

In recent years, methods based on deep convolutional neural networks have made great progress in the field of image super-resolution reconstruction. However, mainstream approaches generally establish many network layers, leading to high computational costs and memory usage that are unsuitable for resource-limited edge devices. To alleviate this issue, a lightweight inverted residual attention network (IRAN) is proposed to obtain better super-resolution reconstruction performance with lower parameters and computation. The main structure of the IRAN consists of a series of inverted residual attention groups (IRAGs), which are mainly composed of several inverted residual attention blocks (IRABs). IRAB effectively reduces the network parameters and computation while extracting depth features by introducing the inverted residual structure and uses the simple channel attention mechanism to learn the important channel feature information. In addition, an enhanced spatial attention mechanism is introduced at the beginning and end of IRAG to further improve the reconstruction performance of the network. The experimental results show that compared with the mainstream lightweight networks, not only the peak signal-to-noise ratio and structural similarity of image quality metrics are better but also the parameters and the computational effort of the network are lower. [ABSTRACT FROM AUTHOR]

Background: This study aimed to develop a deep convolutional neural network (CNN) for automatic classification of pubertal growth spurts using cervical vertebral maturation (CVM) staging based on the lateral cephalograms of an Iranian subpopulation., Material and Methods: Cephalometric radiographs were collected from 1846 eligible patients (aged 5-18 years) referred to the orthodontic department of Hamadan University of Medical Sciences. These images were labeled by two experienced orthodontists. Two scenarios, including two- and three-class (pubertal growth spurts using CVM), were considered as the output for the classification task. The cropped image of the second to fourth cervical vertebrae was used as input to the network. After the preprocessing, the augmentation step, and hyperparameter tuning, the networks were trained with initial random weighting and transfer learning. Finally, the best architecture among the different architectures was determined based on the accuracy and F-score criteria., Results: The CNN based on the ConvNeXtBase-296 architecture had the highest accuracy for automatically assessing pubertal growth spurts based on CVM staging in both three-class (82% accuracy) and two-class (93% accuracy) scenarios. Given the limited amount of data available for training the target networks for most of the architectures in use, transfer learning improves predictive performance., Conclusions: The results of this study confirm the potential of CNNs as an auxiliary diagnostic tool for intelligent assessment of skeletal maturation staging with high accuracy even with a relatively small number of images. Considering the development of orthodontic science toward digitalization, the development of such intelligent decision systems is proposed., (Copyright © 2023 World Federation of Orthodontists. Published by Elsevier Inc. All rights reserved.)

• Wind power generation forecasting based on Deep Federated Learning. • A cyber-secure supermodel for wind power forecasting in critical regions. • A novel Image Processing approach was applied for cyber-attack detection. • Performance of the proposed model was evaluated under a false data injection attack. • The presented results confirmed the superiority of the proposed model. Accurate wind power forecasting is one of the most important operations within the economic dispatch problem to increase the performance of power and energy systems. Accordingly, this study proposes a cyber-resilient hybrid approach based on the Federated Learning and Convolutional Neural Network (CNN) procedure for short-term wind power generation forecasting in different regions of Iran. Generalizability, data independence, forecasting for regions where no training data is available, and preserving the security and privacy of data are prominent features of the proposed method. The federated network was designed with an architecture of 9 clients to perform the training process and extract the salient features from the data associated with each region in each client via the CNN technique. Then, the generalized global supermodel is produced based on the extracted features in each client to forecast the wind power in new and unknown regions such as Mahshahr, Bojnord, and Lootak that had no training data available and had no effect on global supermodel generation. Various scenarios were developed to test the robustness of the suggested methodology. In the first scenario, wind power forecasting is performed based on the suggested technique. In this scenario, the accuracy of the generalized supermodel to forecast wind power generation in each of the Mahshahr, Bojnord, and Lootak regions is 84%, 85%, and 74%, respectively. The second scenario models the scaling attack by changing the wind speed parameters to evaluate the performance of forecasting models against the data integrity attack. In this scenario, an evaluation of the forecast results based on various performance metrics is conducted highlighting the accuracy reduction of the forecast model, due to the damage caused by cyber-attacks on the input data. In the third scenario, the detection of cyber-attack is done based on the image processing-based technique. The presented results emphasize the accurate performance and high generalizability of the cyber-resilient global supermodel in forecasting wind power in various regions of Iran. [ABSTRACT FROM AUTHOR]

Aquifer vulnerability mapping to pollution is topical research activity, and common frameworks such as the basic DRASTIC framework (BDF) suffer from the inherent subjectivity. This paper formulates an artificial intelligence modeling strategy based on Convolutional Neural Network (CNN) to decrease subjectivity. This formulation considers three definitions of intrinsic, specific, and total vulnerabilities. Accordingly, three CNN models are trained and tested to calculate IVI, SVI, and TVI, respectively referring to the intrinsic, specific, and total vulnerability indices. The formulation is applied in an unconfined aquifer northwest of Iran and delineates hotspots within the aquifer. The area under curve (AUC) values derived by the receiver operating curves evaluate the vulnerability indices versus nitrate concentrations. The AUC values for BDF, IVI, SVI, and TVI are 0.81, 0.91, 0.95, and 0.95, respectively. Therefore, CNNs significantly improve the results compared to BDF, but IVI, SVI, and TVI have approximately identical performances. However, the visual comparison between their results provides evidence that significant differences exist between the spatial patterns despite identical AUC values. •Convolutional Neural Network (CNN) was formulated for mapping aquifer vulnerability. •Intrinsic/specific/total vulnerabilities were delineated in an unconfined aquifer. •The inherent subjectivities with the DRASTIC framework was reduced by CNNs. •CNNs identify hotspots within the study area and increase the modeling performances. • Three types of vulnerabilities have different patterns despite identical performance. [ABSTRACT FROM AUTHOR]

Accurate prediction of pore pressure (PP) is among the most critical concerns to the design of drilling operation because of the remarkable role of this parameter in preventing particular drilling problems such as wellbore instability, drilling pipe stuck, mud loss, kicks, and even blow outs. Given the limitations of PP measurement through in-hole well tests, a number of analytic and intelligent techniques have been developed to estimate the PP from conventionally available petrophysical logs at offset wells. In this contribution, analytic equations are combined with intelligent algorithms (IAs) in an integrated workflow for estimating the PP. For this purpose, we collected the required data from two wells (herein referred to as Wells A and B) penetrating a carbonate reservoir in two fields in southwestern Iran. The collected data included full-set petrophysical log data at a total of 2850 points as well as 15 measured PPs using the RFT tool. In order to model and validate the results, the data from Well A was used to train the model, with the Well-B data used for validation. Once finished with data collection, a noise attenuation stage was implemented through median filtering. Subsequently, PP estimation was practiced using a couple of popular analytic models, namely modified Eaton's, Bowers', and compressibility models, with the results compared to the measured PPs. Next, a feature selection phase was conducted where depth (Depth), gamma ray log (CGR), density log (RHOB), resistivity log (RT), pore compressibility (Cp), and slowness log (DT) were selected as the most effective parameters for estimating the PP out of the 8 parameters studied at Well A. Feature selection was performed using the second version of nondominated-sorting genetic algorithm (NSGA-II) combined with multilayer perceptron (MLP) neural network (NN). Next, deep learning techniques, simple form of the least square support vector machine (LSSVM) and its hybrid forms with particle swarm optimization (PSO), cuckoo optimization algorithm (COA), and genetic algorithm (GA), and multilayer extreme learning machine (MELM) hybridized with the PSO, COA, and GA were used to estimate the PP based on the data at Well A, with the results then validated using the data at Well B. Results of the training and testing phases showed that, among the 9 models considered in this research, the best results were produced by the CNN model followed by MELM-COA, and LSSVM-COA, corresponding to root-mean-square errors (RMSEs) of 0.1072, 0.1175, and 0.1237 and determination coefficients (R2) of 0.9884, 0.9860, and 0.9844, respectively, indicating the higher accuracy and generalizability of these models compared to other investigated models. Evaluation of these models on the validation data from Well B further remarked the superiority of the CNN model, as per an RMSE and R2 of 0.1066 and 0.9806, respectively. Indeed, the better performance of the CNN model than the other models in both the training and validation phases reflects the high generalizability of this model in the range of the studied data. In general, the good performance of the intelligent models in similar formation along two wells – where the analytic models rather failed to exhibit consistently good performance – proves the superiority of the IAs over conventional analytic models. This methodology is strongly recommended provided more diverse data is available at in larger amounts. • Large dataset was used to predict pore pressure in two wells from carbonate reservoirs. • Pore pressure modeling was done using hybrid algorithms and deep learning neural network. • Convolutional neural network was superior to other hybrid algorithms. [ABSTRACT FROM AUTHOR]