Performance Of Proposed Model Research Articles

CPU–GPU heterogeneous code acceleration of a finite volume Computational Fluid Dynamics solver

This research focuses on accelerating the finite-volume Computational Fluid Dynamics (CFD) solver, SENSEI, through concurrent CPU–GPU heterogeneous computing, leveraging multiple CPUs and GPUs. An overview of SENSEI, its discretization, and the heterogeneous computing workflow utilizing MPI and OpenACC are provided. A performance model for CPU–GPU heterogeneous computing, incorporating ghost cell exchange usually applied in stencil computations, is introduced to estimate the performance. This work explores the scaling performance of CPU–GPU heterogeneous computing in comparison to a pure multi-CPU/GPU setup using a supersonic inlet test case. The results emphasize the advantages of leveraging both CPU and GPU computational power concurrently. Highlighting the collaborative synergy of CPUs and GPUs as co-workers, this work shows improved performance over exclusive use of pure CPUs or GPUs, with the proposed performance model ensuring a fair estimation of these advantages. This work concludes by offering suggestions for application users interested in accelerating scientific computing simulations and providing feedback for hardware architects designing improved CPU–GPU heterogeneous systems for heterogeneous computing.

Reliability analysis in distribution system by deep belief neural network

Rapid increase in the usage of intermittent renewable energy, ongoing changes in electrical power system structure and operational needs posing growing problems while ensuring adequate service reliability and retaining the quality of power. Power system reliability is a pertinent factor to consider while planning, designing, and operating distribution systems. utilities are obligated to offer their customers uninterrupted electrical service at the least cost while maintaining a satisfactory level of service quality. The important metrics for gauging the effect of distributed renewable energy on distribution networks is reliability analysis. Reliability analysis in distribution systems involves evaluating the performance and robustness of electrical distribution networks. An artificial intelligence approach is implemented in this paper to improve reliability analysis with dispersed generations in distribution network. Deep belief neural networks (DBNNs) are a type of artificial neural network that can be used for various tasks, including analyzing complex data such as those found in power distribution systems. This paper integrated a DBNN using a particle swarm optimization (PSO) technique. The proposed model performance is assessed using mean square error, mean absolute error, root mean square error, and R squared error. The findings reveal that reliability analysis with this novel technique is more accurate.

SADCNN-ORBM: a hybrid deep learning model based citrus disease detection and classification

Citrus disease has a significant influence on agricultural productivity these days, so technology based on artificial intelligence has been developed for creating computer vision (CV) models. By spotting disease in its early stages and enabling necessary productivity actions, CV in agriculture improves the production of agricultural goods. In this paper, we developed a CV-based citrus disease detection model called the self-attention dilated convolutional neural network optimized restricted Boltzmann machine (SADCNN-ORBM) model, which consists of two crucial parts: a SADCNN for disease segmentation and an ORBM optimized by the self-adaptive coati optimization (SACO) algorithm to improve the classification performance of diseases, which successfully divides the disease type into three groups: anthracnose, melanose, and brown spot. Numerous feature sets, such as texture features, three-channel red, green, blue (RGB) features, local binary pattern (LBP) features, and speeded-up robust features (SURF) features, are combined and given as input into the classification layer in the proposed model. We compare our proposed model's performance with existing methods by using several evaluation metrics. The findings demonstrate the SADCNN-ORBM model's superiority in precisely recognizing and classifying citrus illnesses, outperforming all available techniques.

Non-destructive detection of defective maize kernels using hyperspectral imaging and convolutional neural network with attention module

Rapid, effective and non-destructive detection of the defective maize kernels is crucial for their high-quality storage in granary. Hyperspectral imaging (HSI) coupled with convolutional neural network (CNN) based on spectral and spatial attention (Spl-Spal-At) module was proposed for identifying the different types of maize kernels. The HSI data within 380–1000 nm of six classes of sprouted, heat-damaged, insect-damaged, moldy, broken and healthy kernels was collected. The CNN-Spl-At, CNN-Spal-At and CNN-Spl-Spal-At models were established based on the spectra, images and their fusion features as inputs for the recognition of different kernels. Further compared the performances of proposed models and conventional models were built by support vector machine (SVM) and extreme learning machine (ELM). The results indicated that the recognition ability of CNN with attention series models was significantly better than that of SVM and ELM models and fused features were more conducive to expressing the appearance of different kernels than single features. And the CNN-Spl-Spal-At model had an optimal recognition result with high average classification accuracy of 98.04 % and 94.56 % for the training and testing sets, respectively. The recognition results were visually presented on the surface image of kernels with different colors. The CNN-Spl-Spal-At model was built in this study could effectively detect defective maize kernels, and it also had great potential to provide the analysis approaches for the development of non-destructive testing equipment based on HSI technique for maize quality.

Fish Type and Disease Classification Using Deep Learning Model Based Customized CNN with Resnet 50 Technique

Aquaculture is a critical source of seafood production, addressing the global demand for fish products. Suggesting a Deep learning-based classification technique for fishes specifically Indian Major Carp (IMC) as Mrigala, Catla and Rohu is the major objective of this paper along with detecting the disease among them. This world inside hydrosphere has their own discrete living manner. Yet they are not untouched by diseases; fishes mostly affected when young carry pathogens which cause various infections naturally or due to environmental pollutants including chemical and hazardous waste. This paper proposed the classification and prediction of diseases of fishes in aquaculture using Deep Learning based customized Convolutional Neural Network with ResNet-50 model. The proposed model performance metric compared with recent state-of-art techniques. ResNet-50 classifies accurately the IMC and type of disease the fishes are suffering from.

Design of a Novel Chaotic Horse Herd Optimizer and Application to MPPT for Optimal Performance of Stand-Alone Solar PV Water Pumping Systems

A significant part of agricultural farms in the Kingdom of Saudi Arabia (KSA) are in off-grid sites where there is a lack of sufficient water supply despite its availability from groundwater resources in several regions of the country. Since abundant agricultural production is mainly dependent on water, farmers are forced to pump water using diesel generators. This investigation deals with the increase in the effectiveness of a solar photovoltaic water pumping system (SPVWPS). It investigated, from a distinct perspective, the nonlinear behavior of photovoltaic modules that affects the induction motor-pump because of the repeated transitions between the current and the voltage. A new chaotic Horse Herd Optimization (CHHO)-based Maximum Power Point Tracking technique (MPPT) is proposed. This algorithm integrates the capabilities of chaotic search methods to solve the model with a boost converter to maximize power harvest while managing the nonlinear and unpredictable dynamical loads. The analytical modeling for the proposed SPVWPS components and the implemented control strategies of the optimal duty cycle of the DC–DC chopper duty cycle and the Direct Torque Control (DTC) of the Induction Motor (IM) has been conducted. Otherwise, the discussions and evaluations of the proposed model performance in guaranteeing the maximum water flow rate and the operation at MPP of the SPVWPS under partial shading conditions (PSC) and changing weather conditions have been carried out. A comparative study with competitive algorithms was conducted, and the proposed control system’s accuracy and its significant appropriateness to improve the tracking ability for SPVWPS application have been proven in steady and dynamic operating climates and PSC conditions.

Tensor-Based Viterbi Algorithms for Collaborative Cloud-Edge Cyber-Physical-Social Activity Prediction

With the rapid development and application of smart city, Cyber-Physical-Social Systems (CPSS) as its superset is becoming increasingly important, and attracts extensive attentions. For satisfying the smart requirements of CPSS design, a cloud-edge collaborative CPSS framework is first proposed in this paper. Then Coupled-Hidden-Markov-Model (CHMM) and tensor algebra are used to improve existing activity prediction methods for providing CPSS with more intelligent decision support. There are three key features (timing, periodicity and correlation) implied in CPSS data from multi-edge, which affects the accuracy of activity prediction. Thus, these features are synthetically integrated into improved Tensor-based CHMMs (T-CHMMs) to enhance the prediction accuracy. Based on the multi-edge CPSS data, three Tensor-based Viterbi Algorithms (TVA) are correspondingly proposed to solve the prediction problem for T-CHMMs. Compared with traditional matrix-based methods, the proposed TVA could more accurately compute the optimal hidden state sequences under given observation sequences. Finally, the comprehensive performances of proposed models and algorithms are validated on three open datasets by self-comparison and other-comparison. The experimental results show that the proposed methods is superior to the compared three classical methods in terms of F1 measure, average precision and average recall.

A Hybrid Model of Variational Mode Decomposition and Long Short-Term Memory for Next-Hour Wind Speed Forecasting in a Hot Desert Climate

Advancements in technology, policies, and cost reductions have led to rapid growth in wind power (WP) production. One of the major challenges in wind energy production is the instability of WP generation due to weather changes. Efficient power grid management requires accurate power output forecasting. New wind energy forecasting methods based on deep learning (DL) are delivering competitive performance versus traditional methods, like numerical weather prediction (NWP), statistical models and machine learning (ML) models. This is truer for short-term prediction. Since there is a relationship between methods, climates and forecasting complexity, forecasting methods do not always perform the same depending on the climate and terrain of the data source. This paper presents a novel model that combines the variational mode decomposition (VMD) method with a long short-term memory (LSTM) model for next-hour wind speed (WS) prediction in a hot desert climate, such as the climate in Saudi Arabia. The proposed model performance is compared to two other hybrid models, six DL models and four ML models using different feature sets. Also, the proposed model is tested on data from different climates, Caracas and Toronto. The proposed model showed a forecast skill (FS) between 61% and 74% based on mean absolute error (MAE), 64% and 72% based on root mean square error (RMSE), and 59% and 68% based on mean absolute percentage error (MAPE) for locations in Saudi Arabia.

Nonlinear Effects in the Asymmetric Copula-Based Stochastic Frontier Model

Several stochastic frontier production analysis problems are not linear regression problems; two error components (inefficiency and noise) may not have symmetric dependence. In nonlinear problems, the data are divided into two or more regimes according to turning points or change points; the relationships between the output and all the inputs are allowed to differ between regimes. In the problem of two-error dependence, the inefficiency component generally plays a more dominant role than the noise component, thereby the output of each firm is mainly influenced by the inefficiency; in such cases, the symmetric dependence is not suitable. It is, therefore, desirable to come up with a new nonlinear asymmetric copula-based stochastic frontier model (SFM). Two nonlinear structures, namely kink and threshold and skew-normal copula, are suggested to SFM to deal with these problems. The Monte Carlo simulations and analysis of a real data set are employed to evaluate the accuracy and performance of proposed models. The results show that the proposed models show a higher performance when compared with the conventional linear SFM with symmetric two-error dependence.

Productivity analysis of agricultural UAVs by field crop spraying

The article presents an analysis of the performance of agricultural UAVs when spraying field crops. Based on an analysis of the current state of technology for using UAVs in precision agriculture, a model for calculating UAV performance has been developed, based on the time structure of the crop spraying process, including the time of spraying, replenishment of mineral fertilizers or pesticides, and UAV flight. It is noted that for an agricultural UAV, when used in a precision farming system, the equation of existence must relate the take-off mass, the functional components of the take-off mass and the parameters of the sprayer, ensuring differentiated application of pesticides and fertilizers. Based on the proposed performance model of agricultural UAVs, recommendations are given for choosing the type of UAV for spraying crops. This model allows us to generalize methods for testing the production performance of agricultural UAVs, build models of production efficiency and provide proposals for technological solutions to improve these performance characteristics.

Intelligent Deep Learning and Softmax Routing for Energy-Efficient Wireless Sensor Networks in Public Space Design

The increasing usage of several nodes to transfer the massive volume of data to the remotes in wireless sensor networks is a challenging task to reduce the loss. The high volumes of data transmission in wireless sensor networks (WSN) can surpass their capacity, resulting in congestion, latency issues, and packet loss. However, computational intelligence (CI) models can aid in managing and creating intelligent networks in WSN. The WSN congestion issues result in information loss and increased energy usage. CI-based models have been used to resolve this issue, reducing the latency. This paper proposes SoftMax Routing with Deep Neural Network (SRDNN) for efficient routing in WSN. This will route the data packets by choosing the high energy and lower load. It consists of two parts, such as the construction of the routing path, which determines the residual energy of the node. It is analyzed using SoftMax routing to decide whether the node is efficient in energy. The route request and reply established various paths between the source and destination. The path with minimum buffer space and maximum bandwidth is chosen in the optimal routing. The simulation results under the metrics such as energy consumption, data loss rate, throughput, and delay show the proposed model performance.

Deep learning based classification of COVID-19 severity using hierarchical deep maxout model

COVID-19 is a huge pandemic disease and is gradually growing into our daily lives. It has affected us all in different ways, which the World Health Organization classified as a pandemic in 2020. A number of deep learning methods have been developed to identify COVID-19 in chest X-rays, but very few techniques are used to estimate the severity of COVID-19. Additionally, it is very complex to effectively train the data, resulting in slower convergence and time complexities. Therefore, to overcome these drawbacks, this study proposes a unique model to classify the severity of Covid-19 using a deep learning method. The pre-processing step includes image enhancement and noise reduction using the Amended Bilateral Filter (ABF). The relevant features are then extracted using an ensemble-based deep learning model, which includes the DenseNet-201, ResNet-152 and VGG-19 models, which are integrated via an ensemble average approach. The most optimal features are selected using the Modified Archer Fish Optimization Algorithm (Mod-AFO). The proposed optimization algorithm exhibits high convergence and reduced model complexity. COVID-19 severity classification can be effectively categorized using a Multihead attention-assisted hierarchical deep-maxout network model (Multi-HDM). The proposed classification model is comparatively faster in severity estimation and high in performance. Using Python, the proposed model performances are analyzed and compared with the existing models to demonstrate the superiority of the proposed model. The proposed model achieved 98.2% accuracy, 98.5% sensitivity, 96.4% specificity and 96.4% precision in COVID-19 severity estimation using the COVID-19 Chest X-ray images and Lung masks dataset.

Vehicle light communication design using an efficient learning model

Vehicle communication is preferable in vehicular visible light communication (VVLC) owing to its low complexity, secure, interference-free characteristics, line of sight exploitation, vehicle light emitting diodes, and vehicle light propagation characteristics. The channels of VVLC are stochastic and give less accuracy during path loss prediction without considering external environmental factors. There is no wireless channel adopted for channel-based frequency response and node multi-access. This work concentrates on modeling a machine learning-based model for VVLC channel modeling and enhancing the accuracy by specifically concentrating on frequency response via multiple variables like environmental factors and vehicle mobility. The proposed model considers variables with the adoption of an Extreme learning model (ELM) with feed-forward neural networks (FFNN). This model performs information processing over the VLC channel to enhance prediction. The accuracy of the proposed model relies on channel modeling and multi-access relays among nodes. The proposed model performance is evaluated with prevailing approaches to produce distance prediction and response among the relay nodes. Some metrics like BER, SNR, MSE, received power, impulse, and frequency response, and received optical power are evaluated and compared with other approaches.

Wind Power Deviation Charge Reduction using Machine Learning

High penetration of wind power plants in power systems resulted in various challenges such as frequent system imbalances due to highly uncertain and variable wind generation. Additional spinning reserves and specific balancing products such as flexible ramp products are used to handle such frequent imbalances. Incorporation of these ancillary services leads to increased total operational costs. Increased operational costs should be transferred to wind power producers as it is caused by wind power plants. This leads to penalizing the wind power producers for the deviation of power generation from forecasts, called deviation charges. These deviation charges can be reduced by improving the forecasting accuracy. Existing forecasting models show performance in terms of error matrices. Such error matrices do not indicate the financial loss associated with it. This can be overcome by expressing forecasting performance in terms of deviation charge and it will directly encourage wind power producers to improve forecasting accuracy or arrange reserves to accommodate the error. This paper proposes a backpropagation-based artificial neural network model for reducing deviation charges in this context. An analysis is conducted on the data collected from the Bonneville Power Administration (BPA) Balancing Area. Seasonal analysis (Spring, Summer, Fall, and Winter) is conducted to show the performance of the proposed model throughout the year. The proposed model performance is compared with linear regression and ARIMA models. The comparison shows that the proposed ANN model gives the least deviation charges in the Spring, Summer, and Winter seasons and deviation charges in the Fall season are higher than the ARIMA model.

Static Hilbert convex set clustering for web services aggregation

<span>Web services' high levels of duplicate textual structures have caused network bottlenecks and congestion. Clustering and then aggregating similar web services as one compressed message can potentially achieve network traffic reduction. In this paper, a static Hilbert clustering as new model for clustering web services based on convex set similarity is proposed. Mathematically, the proposed model calculates similarity among simple object access protocol (SOAP) messages and then cluster them based on higher similarity values. Next, each cluster is aggregated as a compact message. The experiments have explained the proposed model performance as it has outperformed the convention clustering strategies in both compression ratio and clustering time. The best results have been achievable by the proposed model has reached up to (15) with fixed-length and up to (21) with Huffman.</span>

Using machine learning ensemble method for detection of energy theft in smart meters

AbstractElectricity theft is a primary concern for utility providers, as it leads to substantial financial losses. To address the issue, a novel extreme gradient boosting (XGBoost)‐based model utilizing the consumers’ electricity consumption patterns for analysis is proposed for electricity theft detection (ETD). To remove the imbalance in the real‐world electricity consumption dataset and ensure an even distribution of theft and non‐theft data instances, six different artificially created theft attacks were used. Moreover, the utilization of the XGBoost algorithm for classification, especially to identify malicious instances of electricity theft, yielded commendable accuracy rates and a minimal occurrence of false positives. The proposed model identifies electricity theft specific to the regions, utilizing electricity consumption parameters, and other variables as input features. The authors’ model outperformed existing benchmarks like k‐neural networks, light gradient boost, random forest, support vector machine, decision tree, and AdaBoost. The simulation results using the false attacks for balancing the dataset have improved the proposed model's performance, achieving a precision, recall, and F1‐score of 96%, 95%, and 95%, respectively. The results of the detection rate and the false positive rate (FPR) of the proposed XGBoost‐based detection model have achieved 96% and 3%, respectively.

Adapting the pre-trained convolutional neural networks to improve the anomaly detection and classification in mammographic images

Mortality from breast cancer (BC) is among the top causes of cancer death in women. BC can be effectively treated when diagnosed early, improving the likelihood that a patient will survive. BC masses and calcification clusters must be identified by mammography in order to prevent disease effects and commence therapy at an early stage. A mammography misinterpretation may result in an unnecessary biopsy of the false-positive results, lowering the patient’s odds of survival. This study intends to improve breast mass detection and identification in order to provide better therapy and reduce mortality risk. A new deep-learning (DL) model based on a combination of transfer-learning (TL) and long short-term memory (LSTM) is proposed in this study to adequately facilitate the automatic detection and diagnosis of the BC suspicious region using the 80–20 method. Since DL designs are modelled to be problem-specific, TL applies the knowledge gained during the solution of one problem to another relevant problem. In the presented model, the learning features from the pre-trained networks such as the squeezeNet and DenseNet are extracted and transferred with the features that have been extracted from the INbreast dataset. To measure the proposed model performance, we selected accuracy, sensitivity, specificity, precision, and area under the ROC curve (AUC) as our metrics of choice. The classification of mammographic data using the suggested model yielded overall accuracy, sensitivity, specificity, precision, and AUC values of 99.236%, 98.8%, 99.1%, 96%, and 0.998, respectively, demonstrating the model’s efficacy in detecting breast tumors.

Uncertainty Quantification of Deep Learning–Based Statistical Downscaling of Climatic Parameters

Abstract This study investigated the uncertainty involved in statistical downscaling of hydroclimatic time series obtained by artificial neural networks (ANNs). Phase 6 of the Coupled Model Intercomparison Project (CMIP6) general circulation model (GCM) Canadian Earth System Model, version 5 (CanESM5), was used as large-scale predictor data for downscaling temperature and precipitation parameters. Two ANNs, feed forward and long short-term memory (LSTM), were utilized for statistical downscaling. To quantify the uncertainty of downscaling, prediction intervals (PIs) were estimated via the lower upper bound estimation (LUBE) method. To assess performance of proposed models in different climate regimes, data from the Tabriz and Rasht stations in Iran were employed. The calibrated models via historical GCM data were used for future projections via the high-forcing and fossil fuel–driven development scenario shared socioeconomic pathway (SSP) 5-8.5. Projections were compared with the Canadian Regional Climate Model 4 (Can-RCM4) projections via the same scenario. Results indicated that both LSTM-based point predictions and PIs are more accurate than the feedforward neural network (FFNN)-based predictions, with an average of 55% higher Nash–Sutcliffe efficiency (NSE) for point predictions and 25% lower coverage width criterion (CWC) for PIs. Projections suggested that Tabriz is going to experience a warmer climate with an increase in average temperature of 2° and 5°C for near and far futures, respectively, and a drier climate with a 20% decrease in precipitation until 2100. Future projections for the Rasht station, however, suggested a more uniform climate with less seasonal variability. Average precipitation will increase by up to 25% and 70% until near and far future periods, respectively. Ultimately, point predictions show that the average temperature in Rasht will increase by 1°C until the near future and then be a constant average temperature until the far future. Significance Statement The downscaling of hydroclimatic parameters is subject to uncertainty. The best way is to provide an area with the highest contingency of them as a prediction interval. The reduction width of such an interval leads to increased confidence in explaining and predicting these processes. We proposed and applied a deep learning–based machine learning method for both point prediction and prediction interval estimation of temperature and precipitation parameters for the future over two different climatic regions. The results show the superiority of such a machine learning–based prediction interval estimation for quantification of the downscaling uncertainty.

A novel stacking framework with PSO optimized SVM for effective disease classification

Disease diagnosis is very important in the medical field. It is essential to diagnose chronic diseases such as diabetes, heart disease, cancer, and kidney diseases in the early stage. In recent times, ensembled-based approaches giving effective predictive performance than individual classifiers and gained attention in assisting doctors with early diagnosis. But one of the challenges in these approaches is dealing with class-imbalanced data and improper configuration of ensemble classifiers with optimized parameters. In this paper, a novel 3-level stacking approach with ADASYN oversampling technique with PSO Optimized SVM meta-model (Stacked-ADASYN-PSO) is proposed. Our proposed Stacked-ADASYN-PSO model uses base models such as Logistic regression(LR), K-Nearest neighbor (KNN), Support Vector Machine (SVM), Decision Tree (DT), and Multi-Layer Perceptron (MLP) in layer-0. In layer-1 three meta classifiers namely LR, KNN, and Bagging DT are used. In layer-2 PSO optimized SVM used as the final meta-model to combine the previous layer predictions. To evaluate the robustness of the proposed model It is tested on five benchmark disease datasets from the UCI machine learning repository. These results are compared with state-of-the-art ensemble models and non-ensemble models. Results demonstrated that the proposed model performance is superior in terms of AUC, accuracy, specificity, and precision. We have performed statistical analysis using paired T-tests with a 95% confidence level and our proposed stacking model is significantly differs when compared to base classifiers.

Pedestrian Crossing Intention Prediction Model Considering Social Interaction between Multi-Pedestrians and Multi-Vehicles

Accurate pedestrian crossing intention prediction is critical for autonomous vehicles and advanced driver assistance systems. In multi-pedestrian and multi-vehicle interaction scenes, the social interaction between pedestrians and other traffic participants is ubiquitous, which affects pedestrian crossing decisions and the accuracy of prediction. However, previous studies on pedestrian crossing intention lack comprehensive consideration and mathematical modeling of the social interaction. We propose a “social interaction force” (SIF) to identify and quantify social interaction behaviors and combine the hidden Markov model (HMM) to predict pedestrian crossing intentions 1.0 s ahead. Firstly, a large dataset of pedestrian-vehicle interaction samples is collected from two views, and high-dimensional features are extracted for pedestrian intention prediction. Next, the concept of SIF is proposed to quantitatively characterize the influence of other pedestrians and vehicles on pedestrian crossing decisions, including “pedestrian interaction force” and “pedestrian-vehicle interaction force.” Finally, SIF, pedestrian features, and road structure features are input into HMM. Sliding time windows are applied to the HMM to achieve dynamic prediction of pedestrian intention sequences. Experimental results show that the recognition accuracy of the proposed model is 0.976, and the accuracy of 1.0 s ahead prediction is 0.932 with guaranteed prediction speed. The proposed model performance is superior to that of the most prevalent models developed thus far.