Search Results (20)

Forecasting nuclear energy production is essential for market operations such as security, economic efficiency, resource optimization, grid stability, and the integration of renewable energy sources. Forecasting approaches allow nuclear power plants to operate consistently, contributing to the overall reliability and long-term viability of the energy system. It is noted that energy systems researchers are increasingly interested in machine learning models used to face the challenge of time series forecasting. This study evaluates a hybrid ensemble learning of three time series forecasting models including least-squares support vector regression, gated recurrent unit, and long short-term memory models applied to nuclear power time series forecasting on the dataset of French power plants from 2009 to 2020. Furthermore, this research evaluates forecasting results in which approaches are directed towards the optimized RreliefF (Robust relief Feature) selection algorithm using a hyperparameter optimization based on tree-structured Parzen estimator and following an ensemble learning approach, showing promising results in terms of performance metrics. The suggested ensemble learning model, which combines deep learning and the RreliefF algorithm using a hold-out, outperforms the other nine forecasting models in this study according to performance criteria such as 75% for the coefficient of determination, a root squared error average of 0.108, and an average absolute error of 0.080. Full article

The study developed a large emission area of flexible blue organic light-emitting diodes (BOLED) on a polyethylene terephthalate/ Indium tin oxide (PET/ITO) substrate using a polycyclic skeleton ν-DABNA Thermally Activated Delayed Fluorescence (TADF) material. Initially, a 1 × 1 cm² blue OLED was fabricated to optimize the layer thickness. The blue OLED structure consisted of PET/ITO/HATCN/TAPC/UBH-21:ν-DABNA/TPBi/LiF/Al. However, as the emission area increased to 3.5 × 3.5 cm², the current density decreased due to the resistance of PET/ITO, leading to luminance non-uniformity. To address this issue, auxiliary Au lines were added to the ITO anode to enhance current injection. Despite this, when the Au lines reached a thickness of 30 nm, average light emission was disrupted. To improve the luminescence characteristics of large-area PET/ITO OLEDs, a capping and planarization layer of PEDOT:PSS was applied. Grid uniformity revealed a significant increase in overall luminance uniformity from 74.1% to 87.4% with the addition of auxiliary Au lines. Further increases in grid line density slightly reduced uniformity but enhanced brightness, resulting in brighter, flexible, large-area blue OLED lighting panels. Full article

Short-term load forecasting (STLF) plays a crucial role in the planning, management, and stability of a country’s power system operation. In this study, we have developed a novel approach that can simultaneously predict the load demand of different regions in Bangladesh. When making predictions for loads from multiple locations simultaneously, the overall accuracy of the forecast can be improved by incorporating features from the various areas while reducing the complexity of using multiple models. Accurate and timely load predictions for specific regions with distinct demographics and economic characteristics can assist transmission and distribution companies in properly allocating their resources. Bangladesh, being a relatively small country, is divided into nine distinct power zones for electricity transmission across the nation. In this study, we have proposed a hybrid model, combining the Convolutional Neural Network (CNN) and Gated Recurrent Unit (GRU), designed to forecast load demand seven days ahead for each of the nine power zones simultaneously. For our study, nine years of data from a historical electricity demand dataset (from January 2014 to April 2023) are collected from the Power Grid Company of Bangladesh (PGCB) website. Considering the nonstationary characteristics of the dataset, the Interquartile Range (IQR) method and load averaging are employed to deal effectively with the outliers. Then, for more granularity, this data set has been augmented with interpolation at every 1 h interval. The proposed CNN-GRU model, trained on this augmented and refined dataset, is evaluated against established algorithms in the literature, including Long Short-Term Memory Networks (LSTM), GRU, CNN-LSTM, CNN-GRU, and Transformer-based algorithms. Compared to other approaches, the proposed technique demonstrated superior forecasting accuracy in terms of mean absolute performance error (MAPE) and root mean squared error (RMSE). The dataset and the source code are openly accessible to motivate further research. Full article

Objective: The purpose of this study is to analyze the texture characteristics of chronic non-bacterial osteomyelitis (CNO) bone lesions, identified as areas of altered signal intensity on short tau inversion recovery (STIR) sequences, and to distinguish them from bone marrow growth-related changes through Machine Learning (ML) and Deep Learning (DL) analysis. Materials and methods: We included a group of 66 patients with confirmed diagnosis of CNO and a group of 28 patients with suspected extra-skeletal systemic disease. All examinations were performed on a 1.5 T MRI scanner. Using the opensource 3D Slicer software version 4.10.2, the ROIs on CNO lesions and on the red bone marrow were sampled. Texture analysis (TA) was carried out using Pyradiomics. We applied an optimization search grid algorithm on nine classic ML classifiers and a Deep Learning (DL) Neural Network (NN). The model’s performance was evaluated using Accuracy (ACC), AUC-ROC curves, F1-score, Positive Predictive Value (PPV), Mean Absolute Error (MAE) and Root-Mean-Square Error (RMSE). Furthermore, we used Shapley additive explanations to gain insight into the behavior of the prediction model. Results: Most predictive characteristics were selected by Boruta algorithm for each combination of ROI sequences for the characterization and classification of the two types of signal hyperintensity. The overall best classification result was obtained by the NN with ACC = 0.91, AUC = 0.93 with 95% CI 0.91–0.94, F1-score = 0.94 and PPV = 93.8%. Between classic ML methods, ensemble learners showed high model performance; specifically, the best-performing classifier was the Stack (ST) with ACC = 0.85, AUC = 0.81 with 95% CI 0.8–0.84, F1-score = 0.9, PPV = 90%. Conclusions: Our results show the potential of ML methods in discerning edema-like lesions, in particular by distinguishing CNO lesions from hematopoietic bone marrow changes in a pediatric population. The Neural Network showed the overall best results, while a Stacking classifier, based on Gradient Boosting and Random Forest as principal estimators and Logistic Regressor as final estimator, achieved the best results between the other ML methods. Full article

Monitoring rainfall in the Brazilian Legal Amazon (BLA), which comprises most of the largest tropical rainforest and largest river basin on the planet, is extremely important but challenging. The size of the area and land cover alone impose difficulties on the operation of a rain gauge network. Given this, we aimed to evaluate the performance of nine databases that estimate rainfall in the BLA, four from gridded analyses based on pluviometry (Xavier, CPC, GPCC and CRU), four based on remote sensing (CHIRPS, IMERG, CMORPH and PERSIANN-CDR), and one from reanalysis (ERA5Land). We found that all the bases are efficient in characterizing the average annual cycle of accumulated precipitation in the BLA, but with a predominantly negative bias. Parameters such as Pearson’s correlation (r), root-mean-square error (RMSE) and Taylor diagrams (SDE), applied in a spatial analysis for the entire BLA as well as for six pluviometrically homogeneous regions, showed that, based on a skill ranking, the data from Xavier’s grid analysis, CHIRPS, GPCC and ERA5Land best represent precipitation in the BLA at monthly, seasonal and annual levels. The PERSIANN-CDR data showed intermediate performance, while the IMERG, CMORPH, CRU and CPC data showed the lowest correlations and highest errors, characteristics also captured in the Taylor diagrams. It is hoped that this demonstration of hierarchy based on skill will subsidize climate studies in this region of great relevance in terms of biodiversity, water resources and as an important climate regulator. Full article

Manual therapy (MT) techniques typically incorporate localised touch on the skin with the application of specific kinetic forces. The contribution of localised touch to the effectiveness of MT techniques has not been evaluated. This study investigated the immediate effects of MT versus localisation training (LT) on pain intensity and range of movement (ROM) for neck pain. In this single-blind randomised controlled trial thirty eligible neck pain volunteers (23 females and 7 males), aged 28.63 ± 12.49 years, were randomly allocated to MT or to a motionless (LT) group. A single three-minute treatment session was delivered to each group’s cervico-thoracic area. The LT involved tactile sensory stimulation applied randomly to one out of a nine-block grid. Subjects were asked to identify the number of the square being touched, reflecting a different location on the region of skin. MT involved three-minute anteroposterior (AP) glides and sustained natural apophyseal glides (SNAG) techniques. Pre- and post-intervention pain intensity were assessed using a pressure pain threshold (PPT) algometer and the numeric pain rating scale (NPRS). Neck ROM was recorded with a bubble inclinometer. Improvements in ROM and self-reported pain were recorded in both groups (p < 0.001) without differences in NPRS, ROM or PPT scores between groups (p > 0.05). Tactile sensory training (localisation) was as effective as MT in reducing neck pain, suggesting a component of MT’s analgesic effect to be related with the element of localised touch rather than the forces induced during passive movements. Full article

Climate change and global warming drive many governments and scientists to investigate new renewable and green energy sources. Special attention is on solar panel technology, since solar energy is considered one of the primary renewable sources and solar panels can be installed in domestic neighborhoods. Photovoltaic (PV) power prediction is essential to match supply and demand and ensure grid stability. However, the PV system has assertive stochastic behavior, requiring advanced forecasting methods, such as machine learning and deep learning, to predict day-ahead PV power accurately. Machine learning models need a rich historical dataset that includes years of PV power outputs to capture hidden patterns between essential variables to predict day-ahead PV power production accurately. Therefore, this study presents a framework based on the transfer learning method to use reliable trained deep learning models of old PV plants in newly installed PV plants in the same neighborhoods. The numerical results show the effectiveness of transfer learning in day-ahead PV prediction in newly established PV plants where a sizable historical dataset of them is unavailable. Among all nine models presented in this study, the LSTM models have better performance in PV power prediction. The new LSTM model using the inadequate dataset has 0.55 mean square error (MSE) and 47.07% weighted mean absolute percentage error (wMAPE), while the transferred LSTM model improves prediction accuracy to 0.168 MSE and 32.04% wMAPE. Full article

An accurate and fast building load prediction model is critically important for guiding building energy system design, optimizing operational parameters, and balancing a power grid between energy supply and demand. A physics-based simulation tool is traditionally used to provide the building load demand; however, it is constrained by its complex model development process and requirement for engineering judgments. Machine learning algorithms (i.e., data-driven models) based on big data can bridge this gap. In this study, we used the massive energy data generated by a physics-based tool (EnergyPlus) to develop three data-driven models (i.e., LightGBM, random forest (RF), and long-short term memory (LSTM)) and compared their prediction performances. The physics-based models were developed using office prototype building models as baselines, and ranges were provided for selected key input parameters. Three different input feature dimensions (i.e., six-, nine-, and fifteen-input feature selections) were investigated, aiming to meet different demands for practical applications. We found that LightGBM significantly outperforms the RF and LSTM algorithms, not only with respect to prediction accuracy but also in regard to computation cost. The best prediction results show that the coefficient of variation of the root mean squared error (CVRMSE), squared correction coefficient (R²), and computation time are 5.25%, 0.9959, and 7.0 s for LightGBM, respectively, evidently better than the values for the algorithms based on RF (18.54%, 0.9482, and 44.6 s) and LSTM (22.06%, 0.9267, and 758.8 s). The findings demonstrate that a data-driven model is able to avoid the process of establishing a complicated physics-based model for predicting a building’s thermal load, with similar accuracy to that of a physics-based simulation tool. Full article

Cloud computing is currently the most cost-effective means of providing commercial and consumer IT services online. However, it is prone to new flaws. An economic denial of sustainability attack (EDoS) specifically leverages the pay-per-use paradigm in building up resource demands over time, culminating in unanticipated usage charges to the cloud customer. We present an effective approach to mitigating EDoS attacks in cloud computing. To mitigate such distributed attacks, methods for detecting them on different cloud computing smart grids have been suggested. These include hard-threshold, machine, and deep learning, support vector machine (SVM), K-nearest neighbors (KNN), random forest (RF) tree algorithms, namely convolutional neural network (CNN), and long short-term memory (LSTM). These algorithms have greater accuracies and lower false alarm rates and are essential for improving the cloud computing service provider security system. The dataset of nine injection attacks for testing machine and deep learning algorithms was obtained from the Cyber Range Lab at the University of New South Wales (UNSW), Canberra. The experiments were conducted in two categories: binary classification, which included normal and attack datasets, and multi-classification, which included nine classes of attack data. The results of the proposed algorithms showed that the RF approach achieved accuracy of 98% with binary classification, whereas the SVM model achieved accuracy of 97.54% with multi-classification. Moreover, statistical analyses, such as mean square error (MSE), Pearson correlation coefficient (R), and the root mean square error (RMSE), were applied in evaluating the prediction errors between the input data and the prediction values from different machine and deep learning algorithms. The RF tree algorithm achieved a very low prediction level (MSE = 0.01465) and a correlation R² (R squared) level of 92.02% with the binary classification dataset, whereas the algorithm attained an R² level of 89.35% with a multi-classification dataset. The findings of the proposed system were compared with different existing EDoS attack detection systems. The proposed attack mitigation algorithms, which were developed based on artificial intelligence, outperformed the few existing systems. The goal of this research is to enable the detection and effective mitigation of EDoS attacks. Full article

Textile-reinforced mortar (TRM) is an effective method for confining concrete elements to elevate the axial load resistance and upgrade the overall performance of concrete. TRM is a promising alternative to carbon-fiber-reinforced polymers (CFRP) which are commonly used to strengthen concrete and are known to be expensive since they require a huge amount of energy in processing these materials. Green technologies can be applied in this process, following the same TRM principles of confinement, replacing conventional cement or epoxy-based mortars and synthetic textiles towards sustainable concrete strengthening technology. This is through the utilization of a geopolymer mortar reinforced with short banana fibers (BF) and long BFs as textiles. Geopolymer mortar presented in this paper is composed of fly ash and silica fume as the binder, sand as the filler, sodium hydroxide (NaOH) and sodium silicate (Na₂SiO₃) as the activator and BFs as the reinforcement and textile. Geopolymerization generates significantly less carbon dioxide (CO₂) while BFs are known for having attractive mechanical properties, are cost effective and abundant in nature, and thus the use of this fiber will significantly minimize the huge waste produced from banana plantations after a one-time fruit harvest. The geotextile or geogrid used to wrap the concrete cylinder samples is made up of 2 mm-long BF yarns with weights ranging from 150 to 450 grams per square meter that varies with grid sizes from 10 mm, 15 mm to 25 mm for both orthogonal directions considering the lightweight characteristic of BFs. Twelve TRM designs were used to strengthen the concrete cylinders with three samples each. TRM design parameters vary in the thicknesses of the geopolymer mortar covering and the size of the geotextile grids. Eighteen of the geotextiles used were coated with a polymer to protect the fibers while the other eighteen geotextiles remained uncoated. A total of thirty-nine concrete cylinders with 150 mm base diameter and 300 mm height cured within 28 days were prepared, for which 36 cylinders were confined with green TRM with different parameters while three of the plain concrete cylinders served as the control specimens. This is to maximize the investigation on the potential of green TRM in confining concrete and to determine the variations in compressive strengths and mode of failures of confined and unconfined concrete specimens. Results highlighted notable enhancement in the mechanical properties of the modified plain concrete after 28 days of TRM curing using a universal testing machine (UTM). Likewise, a confinement theory of the optimum TRM design was modeled mathematically to evaluate the effects of concrete confinement and overall load carrying capacity enhancement gained from additional strength transferred by the TRM to the concrete element. Full article

Due to severe climate change impact on electricity consumption, as well as new trends in smart grids (such as the use of renewable resources and the advent of prosumers and energy commons), medium-term and long-term electricity load forecasting has become a crucial need. Such forecasts are necessary to support the plans and decisions related to the capacity evaluation of centralized and decentralized power generation systems, demand response strategies, and controlling the operation. To address this problem, the main objective of this study is to develop and compare precise district level models for predicting the electrical load demand based on machine learning techniques including support vector machine (SVM) and Random Forest (RF), and deep learning methods such as non-linear auto-regressive exogenous (NARX) neural network and recurrent neural networks (Long Short-Term Memory—LSTM). A dataset including nine years of historical load demand for Bruce County, Ontario, Canada, fused with the climatic information (temperature and wind speed) are used to train the models after completing the preprocessing and cleaning stages. The results show that by employing deep learning, the model could predict the load demand more accurately than SVM and RF, with an R-Squared of about 0.93–0.96 and Mean Absolute Percentage Error (MAPE) of about 4–10%. The model can be used not only by the municipalities as well as utility companies and power distributors in the management and expansion of electricity grids; but also by the households to make decisions on the adoption of home- and district-scale renewable energy technologies. Full article

The goal of the paper is to develop an online forecasting procedure to be adopted within the H2020 InteGRIDy project, where the main objective is to use the photovoltaic (PV) forecast for optimizing the configuration of a distribution network (DN). Real-time measurements are obtained and saved for nine photovoltaic plants in a database, together with numerical weather predictions supplied from a commercial weather forecasting service. Adopting several error metrics as a performance index, as well as a historical data set for one of the plants on the DN, a preliminary analysis is performed investigating multiple statistical methods, with the objective of finding the most suitable one in terms of accuracy and computational effort. Hourly forecasts are performed each 6 h, for a horizon of 72 h. Having found the random forest method as the most suitable one, further hyper-parameter tuning of the algorithm was performed to improve performance. Optimal results with respect to normalized root mean square error (NRMSE) were found when training the algorithm using solar irradiation and a time vector, with a dataset consisting of 21 days. It was concluded that adding more features does not improve the accuracy when adopting relatively small training sets. Furthermore, the error was not significantly affected by the horizon of the forecast, where the 72-h horizon forecast showed an error increment of slightly above 2% when compared to the 6-h forecast. Thanks to the InteGRIDy project, the proposed algorithms were tested in a large scale real-life pilot, allowing the validation of the mathematical approach, but taking also into account both, problems related to faults in the telecommunication grids, as well as errors in the data exchange and storage procedures. Such an approach is capable of providing a proper quantification of the performances in a real-life scenario. Full article

Since the onset of agriculture, soils have lost their organic carbon to such an extent that the soil functions of many croplands are threatened. Hence, there is a strong demand for mapping and monitoring critical soil properties and in particular soil organic carbon (SOC). Pilot studies have demonstrated the potential for remote sensing techniques for SOC mapping in croplands. It has, however, been shown that the assessment of SOC may be hampered by the condition of the soil surface. While growing vegetation can be readily detected by means of the well-known Normalized Difference Vegetation Index (NDVI), the distinction between bare soil and crop residues is expressed in the shortwave infrared region (SWIR), which is only covered by two broad bands in Landsat or Sentinel-2 imagery. Here we tested the effect of thresholds for the Cellulose Absorption Index (CAI), on the performance of SOC prediction models for cropland soils. Airborne Prism Experiment (APEX) hyperspectral images covering an area of 240 km² in the Belgian Loam Belt were used together with a local soil dataset. We used the partial least square regression (PLSR) model to estimate the SOC content based on 104 georeferenced calibration samples (NDVI < 0.26), firstly without setting a CAI threshold, and obtained a satisfactory result (coefficient of determination (R²) = 0.49, Ratio of Performance to Deviation (RPD) = 1.4 and Root Mean Square Error (RMSE) = 2.13 g kgC⁻¹ for cross-validation). However, a cross comparison of the estimated SOC values to grid-based measurements of SOC content within three fields revealed a systematic overestimation for fields with high residue cover. We then tested different CAI thresholds in order to mask pixels with high residue cover. The best model was obtained for a CAI threshold of 0.75 (R² = 0.59, RPD = 1.5 and RMSE = 1.75 g kgC⁻¹ for cross-validation). These results reveal that the purity of the pixels needs to be assessed aforehand in order to produce reliable SOC maps. The Normalized Burn Ratio (NBR2) index based on the SWIR bands of the MSI Sentinel 2 sensor extracted from images collected nine days before the APEX flight campaign correlates well with the CAI index of the APEX imagery. However, the NBR2 index calculated from Sentinel 2 images under moist conditions is poorly correlated with residue cover. This can be explained by the sensitivity of the NBR2 index to both soil moisture and residues. Full article

Satellite remote sensing is a practical technique to estimate global precipitation with adequate spatiotemporal resolution in ungauged regions. However, the performance of satellite-based precipitation products is variable and uncertain for the Tibetan Plateau (TP) because of its complex terrain and climate conditions. In this study, we evaluated the abilities of nine widely used satellite-based precipitation products over the Eastern Tibetan Plateau (ETP) and quantified precipitation dynamics over the entire TP. The evaluation was carried out from three aspects, i.e., magnitude agreement, occurrence consistency, and elevation dependency, from grid-cell to regional scales. The results show that the nine satellite-based products exhibited different agreement with gauge-based reference data with median correlation coefficients ranging from 0.15 to 0.95. Three products (climate hazards group infrared precipitation with stations (CHIRPS), multi-source weighted-ensemble precipitation (MSWEP), and tropical rainfall measuring mission multi-satellite precipitation analysis (TMPA)) generally presented the best performance with the reference data, even in complex terrain regions, given their root mean square errors (RMSE) of less than 25 mm/mon. The climate prediction center merged analysis of precipitation (CMAP) product has relatively coarse spatial resolution, but it also exhibited good performance with a bias of less than 20% in watershed scale. Two other products (precipitation estimation from remotely sensed information using artificial neural networks-cloud classification system (PER-CCS) and climate prediction center morphing technique-raw (CMORPH-RAW)) overestimated precipitation with median RMSEs of 87 mm/mon and 45 mm/mon, respectively. All the precipitation products generally exhibited better agreement with the reference data for rainy season and lower-elevation regions. All of the products captured precipitation occurrence well, with hit event over 60%, and similar percentages of missed and false event. According to the evaluation, the four products (CHIRPS, MSWEP, TMPA, and CMAP) revealed that the annual precipitation over the TP fluctuated between 333 mm/yr and 488 mm/yr during the period 2003 to 2015. The study indicates the importance of integration of multiple data sources and post-processing (e.g., gauge data fusion and elevation correction) for satellite-based products and have implications for selection of suitable precipitation products for hydrological modeling and water resources assessment for the TP. Full article

High-resolution meteorological data products are crucial for agrometeorological studies. Here, we study the accuracy of an important gridded dataset, the near-surface temperature dataset from the 5 km × 5 km resolution China dataset of meteorological forcing for land surface modeling (published by the Beijing Normal University). Using both the gridded dataset and the observed temperature data from 590 meteorological stations, we calculate nine universal meteorological indices (mean, maximum, and minimum temperatures of daily, monthly, and annual data) and five agricultural thermal indices (first frost day, last frost day, frost-free period, and ≥0 °C and ≥10 °C active accumulated temperature, i.e., AAT0 and AAT10) of the 11 temperature zones over mainland China. Then, for each meteorological index, we calculate the root mean square errors (RMSEs), correlation coefficient and climate trend rates of the two datasets. The results show that the RMSEs of these indices are usually lower in the north subtropical, mid-subtropical, south subtropical, marginal tropical and mid-tropical zones than in the plateau subfrigid, plateau temperate, and plateau subtropical mountains zones. Over mainland China, the AAT0, AAT10, and mean and maximum temperatures calculated from the gridded data show the same climate trends with those derived from the observed data, while the minimum temperature and its derivations (first frost day, last frost day, and frost-free period) show the opposite trends in many areas. Thus, the mean and maximum temperature data derived from the gridded dataset are applicable for studies in most parts of China, but caution should be taken when using the minimum temperature data. Full article