• Energy Research

The precise prediction of power estimates of wind–solar renewable energy sources becomes challenging due to their intermittent nature and difference in intensity between day and night. Machine-learning algorithms are non-linear mapping functions to approximate any given function from known input–output pairs and can be used for this purpose. This paper presents an artificial neural network (ANN)-based method to predict hybrid wind–solar resources and estimate power generation by correlating wind speed and solar radiation for real-time data. The proposed ANN allows optimization of the hybrid system’s operation by efficient wind and solar energy production estimation for a given set of weather conditions. The proposed model uses temperature, humidity, air pressure, solar radiation, optimum angle, and target values of known wind speeds, solar radiation, and optimum angle. A normalization function to narrow the error distribution and an iterative method with the Levenberg–Marquardt training function is used to reduce error. The experimental results show the effectiveness of the proposed approach against the existing wind, solar, or wind–solar estimation methods. It is envisaged that such an intelligent yet simplified method for predicting wind speed, solar radiation, and optimum angle, and designing wind–solar hybrid systems can improve the accuracy and efficiency of renewable energy generation.

Facial emotion recognition (FER) is an important and developing topic of research in the field of pattern recognition. The effective application of facial emotion analysis is gaining popularity in surveillance footage, expression analysis, activity recognition, home automation, computer games, stress treatment, patient observation, depression, psychoanalysis, and robotics. Robot interfaces, emotion-aware smart agent systems, and efficient human–computer interaction all benefit greatly from facial expression recognition. This has garnered attention as a key prospect in recent years. However, due to shortcomings in the presence of occlusions, fluctuations in lighting, and changes in physical appearance, research on emotion recognition has to be improved. This paper proposes a new architecture design of a convolutional neural network (CNN) for the FER system and contains five convolution layers, one fully connected layer with rectified linear unit activation function, and a SoftMax layer. Additionally, the feature map enhancement is applied to accomplish a higher detection rate and higher precision. Lastly, an application is developed that mitigates the effects of the aforementioned problems and can identify the basic expressions of human emotions, such as joy, grief, surprise, fear, contempt, anger, etc. Results indicate that the proposed CNN achieves 92.66% accuracy with mixed datasets, while the accuracy for the cross dataset is 94.94%.

Technology’s expansion has contributed to the rise in popularity of social media platforms. Twitter is one of the leading social media platforms that people use to share their opinions. Such opinions, sometimes, may contain threatening text, deliberately or non-deliberately, which can be disturbing for other users. Consequently, the detection of threatening content on social media is an important task. Contrary to high-resource languages like English, Dutch, and others that have several such approaches, the low-resource Urdu language does not have such a luxury. Therefore, this study presents an intelligent threatening language detection for the Urdu language. A stacking model is proposed that uses an extra tree (ET) classifier and Bayes theorem-based Bernoulli Naive Bayes (BNB) as the based learners while logistic regression (LR) is employed as the meta learner. A performance analysis is carried out by deploying a support vector classifier, ET, LR, BNB, fully connected network, convolutional neural network, long short-term memory, and gated recurrent unit. Experimental results indicate that the stacked model performs better than both machine learning and deep learning models. With 74.01% accuracy, 70.84% precision, 75.65% recall, and 73.99% F1 score, the model outperforms the existing benchmark study.

With the fast and rapid pace of developments in wireless technology, energy consumption has become a problem of great significance for future networks. Over the past few years, several energy monitoring policies have been initiated to promote energy efficiency emphasizing the active and important role of consumers to realize this goal. This study resolves the energy consumption issues of broadband networks and determines the energy usage associated with high spectrum allocation in future broadband networks by leveraging clustering from the data mining domain. For analyzing the overall patterns of energy consumption in broadband networks, this study segments the broadband networks based on the similarities of their electrical load profiles and the proportion of energy usage per hour (%) as a common framework and divides the users into different groups. The prime objective for the segmentation is to provide personalized recommendations to each group to reduce the energy consumption and associated costs thereby fostering energy efficiency measures and improving consumer engagement. The segmentation is obtained by an iterative process based on computational clusters calculation which is finalized by a post clustering analysis. Post clustering analysis involves visualization and statistical data mining techniques to analyze the energy consumption patterns and reallocating to a more appropriate group. The K-Means clustering technique is utilized for this purpose which provides the best prediction accuracy of 98.46% for energy load profiles at the spectrum of 100GHz. The energy consumption segmentation of the consumers provides knowledge and a better understanding of the consumer for optimizing energy consumption for future broadband.

3D printing is emerging as a technology for future production due to its support for human life. Increasingly more printed products include many applications. Developers and companies have expressed their ambition to develop the next generation to bring 3D printers to most families. However, energy efficiency is a big challenge for such devices. In this research, we investigated the power of components given by measurements on commercial 3D printers. We then built a compact model to estimate the energy of 3D printers and proposed an energy-saving strategy, primarily focused on the heating process. We separated thermal plates into two independent temperature sections to cut wasted energy costs when printing specially shaped objects and small prints. In order to reduce power dissipation, the printing process needs to be installed at high ambient temperatures. Experimental results show that our method reduces 23% of total power consumption in comparison to the current commercial device.

Accurate solar and photovoltaic (PV) power forecasting is essential for optimizing grid integration, managing energy storage, and maximizing the efficiency of solar power systems. Deep learning (DL) models have shown promise in this area due to their ability to learn complex, non-linear relationships within large datasets. This study presents a systematic literature review (SLR) of deep learning applications for solar PV forecasting, addressing a gap in the existing literature, which often focuses on traditional ML or broader renewable energy applications. This review specifically aims to identify the DL architectures employed, preprocessing and feature engineering techniques used, the input features leveraged, evaluation metrics applied, and the persistent challenges in this field. Through a rigorous analysis of 26 selected papers from an initial set of 155 articles retrieved from the Web of Science database, we found that Long Short-Term Memory (LSTM) networks were the most frequently used algorithm (appearing in 32.69% of the papers), closely followed by Convolutional Neural Networks (CNNs) at 28.85%. Furthermore, Wavelet Transform (WT) was found to be the most prominent data decomposition technique, while Pearson Correlation was the most used for feature selection. We also found that ambient temperature, pressure, and humidity are the most common input features. Our systematic evaluation provides critical insights into state-of-the-art DL-based solar forecasting and identifies key areas for upcoming research. Future research should prioritize the development of more robust and interpretable models, as well as explore the integration of multi-source data to further enhance forecasting accuracy. Such advancements are crucial for the effective integration of solar energy into future power grids.

Data center servers located in thermal hotspot regions receive inlet air at a higher than the set temperature and thus generate comparatively high outlet temperature. Consequently, there is a rise in energy that is consumed to cool down the servers that otherwise would undergo reliability hazards. The workload deployment across the servers should be resilient to thermal hotspots to ensure smooth performance. In a heterogeneous data center environment, an equally important fact is the placement of the servers in a thermal hotspot-aware manner to lower the peak outlet temperatures. These approaches can be applied proactively with the help of outlet temperature prediction. This paper presents the hotspot adaptive workload deployment algorithm (HAWDA) and hotspot aware server relocation algorithm (HASRA) based on thermal profiling regarding outlet temperature prediction. HAWDA deploys workload on servers in a thermal-efficient manner and HASRA optimizes the server location in thermal hotspot regions to lower the peak outlet temperatures. Performance comparison is carried out to analyze the efficacy of HAWDA against the TASA and GRANITE algorithms. Results suggest that HAWDA provides average peak utilization of the servers similar to GRANITE and TASA without additional burden on the cooling mechanism, with and without server relocation, as HAWDA minimizes the peak outlet temperature.