Search

Nowadays, for controlling crime, surveillance cameras are typically installed in all public places to ensure urban safety and security. However, automating Human Activity Recognition (HAR) using computer vision techniques faces several challenges such as lowlighting, complex spatiotemporal features, clutter backgrounds, and inefficient utilization of surveillance system resources. Existing attempts in HAR designed straightforward networks by analyzing either spatial or motion patterns resulting in limited performance while the dual streams methods are entirely based on Convolutional Neural Networks (CNN) that are inadequate to learning the long-range temporal information for HAR. To overcome the above-mentioned challenges, this paper proposes an optimized dual stream framework for HAR which mainly consists of three steps. First, a shots segmentation module is introduced in the proposed framework to efficiently utilize the surveillance system resources by enhancing the lowlight video stream and then it detects salient video frames that consist of human. This module is trained on our own challenging Lowlight Human Surveillance Dataset (LHSD) which consists of both normal and different levels of lowlighting data to recognize humans in complex uncertain environments. Next, to learn HAR from both contextual and motion information, a dual stream approach is used in the feature extraction. In the first stream, it freezes the learned weights of the backbone Vision Transformer (ViT) B-16 model to select the discriminative contextual information. In the second stream, ViT features are then fused with the intermediate encoder layers of FlowNet2 model for optical flow to extract a robust motion feature vector. Finally, a two stream Parallel Bidirectional Long Short-Term Memory (PBiLSTM) is proposed for sequence learning to capture the global semantics of activities, followed by Dual Stream Multi-Head Attention (DSMHA) with a late fusion strategy to optimize the huge features vector for accurate HAR. To assess the strength of the proposed framework, extensive empirical results are conducted on real-world surveillance scenarios and various benchmark HAR datasets that achieve 78.6285%, 96.0151%, and 98.875% accuracies on HMDB51, UCF101, and YouTube Action, respectively. Our results show that the proposed strategy outperforms State-of-the-Art (SOTA) methods. The proposed framework gives superior performance in HAR, providing accurate and reliable recognition of human activities in surveillance systems.

The method examines the challenging issues in tracking multiple autonomous ground vehicles (AGVs) using a motion capture system (MoCap) accessible by a wireless network such as standard IEEE 802.11a WIFI protocol. A state-of-art technology such as global positioning system may have limitations in their accuracy and restricted lines of sight. Tracking in these environments entail various complexities such as target (AGV) occlusion, clutter, and the electromagnetic interference that can interrupt communication between AGV and ground-based control station. We present a novel idea that exploits IR signals emitted by the MoCap to fetch position data reflected from AGVs. MoCap uses the Motive software to manipulate the position data necessary for the tracking filter. This method adopted the fixed-interval smoothing based on the joint integrated track splitting filter (FIsJITS) for detecting and tracking the vehicles. FIsJITS obtains track state estimation in both forward and backward directions within fixed time measurement intervals. The multi-track backward predictions are fused in the forward-path track to obtain a-priori smoothing predictions, followed by a smoothing state estimation. This approach also calculates the smoothing target existence probability (TEP) to reinforce target detection, allowing a tracking system to simultaneously track multiple vehicles efficiently within cluttered environments. For a seamless data processing, we utilize the WIFI through a wireless access point (WAP) to transfer the data from MoCap to the computer system where Motive and tracking system softwares are running simultaneously. We conducted the real-time experiments to demonstrate the tracking performance of the proposed AGV system outperforming existing algorithms.

Human Activity Recognition (HAR) plays a crucial role in communication and the Internet of Things (IoT), by enabling vision sensors to understand and respond to human behavior more intelligently and efficiently. Existing deep learning models are complex to deal with the low illumination, diverse viewpoints, and cluttered backgrounds, which require substantial computing resources and are not appropriate for edge devices. Furthermore, without an effective video analysis technique it processes entire frames, resulting inadequate performance. To address these key challenges, a cloud-assisted IoT computing framework is proposed for HAR in uncertain low-lighting environments, which is mainly composed of two tiers: edge and cloud computing. Initially, a lightweight Convolutional Neural Network (CNN) model is developed which is responsible to enhance the low-light frames, followed by the human detection algorithm to process the selective frames, thus enabling efficient resource utilization. Next, these refined frames are then transmitted to the cloud for accurate HAR, where dual stream CNN and transformer fusion network extract both short-and long-range spatiotemporal discriminative features followed by proposed Optimized Parallel Sequential Temporal Network (OPSTN) with squeeze and excitation attention to efficiently learn HAR in complex scenarios. Finally, extensive experiments are conducted over three challenging HAR datasets to deeply examine the proposed framework from various perspectives such as complex activity recognition, lowlighting, etc., where the results are outperformed compared with the state-of-art methods.

Abstract Bio‐active food coating having natural antioxidants has attained great attention these days. Polysaccharides extracted from bacteria, fungi, and plants are considered rich in antioxidant biomolecules. Chickpea hull which is a food waste material contains a substantial quantity of antioxidants and bioactive compounds. In this study, chitosan (CS)‐based chickpea hull polysaccharides (CHPS) edible coating of cherry tomatoes was successfully fabricated. Cherry tomatoes were characterized in terms of physico‐chemical characteristics and shelf life. In comparison to the control, it was discovered that the CS‐incorporated CHPS coatings were successful at lowering the respiratory activity, total soluble solids, total polyphenols, firmness, weight loss, lycopene content, and vitamin C as well as improving the fruit's overall acceptability. The dose dependence of each of these effects was noticed. Conclusively, using CS‐incorporated CHPS coatings could preserve the shelf life of cherry tomatoes. A useful and different approach to enhance the postharvest quality of cherry tomatoes is to utilize CS‐CHPS composite coatings.

Fuzzy graphs are very important when we are trying to understand and study complex systems with uncertain and not exact information. Among different types of fuzzy graphs, cubic fuzzy graphs are special due to their ability to represent the membership degree of both vertices and edges using intervals and fuzzy numbers, respectively. To figure out how things are connected in cubic fuzzy graphs, we need to know about cubic α-strong, cubic β-strong and cubic δ-weak edges. These concepts better help in making decisions, solving problems and analyzing things like transportation, social networks and communication systems. The applicability of connectivity and comprehension of cubic fuzzy graphs have urged us to discuss connectivity in the domain of cubic fuzzy graphs. In this paper, the terms partial cubic α-strong and partial cubic δ-weak edges are introduced for cubic fuzzy graphs. The bounds and exact expression of connectivity index for several cubic fuzzy graphs are estimated. The average connectivity index for cubic fuzzy graphs is also defined and some results pertaining to these concepts are proved in this paper. The results demonstrate that removing some vertices or edges may cause a change in the value of connectivity index or average connectivity index, but the change will not necessarily be related to both values. This paper also defines the concepts of partial cubic connectivity enhancing node and partial cubic connectivity reducing node and some related results are proved. Furthermore, the concepts of cubic α-strong, cubic β- strong, cubic δ-weak edge, partial cubic α-strong and partial cubic δ-weak edges are utilized to identify areas most affected by a tsunami resulting from an earthquake. Finally, the research findings are compared with the existing methods to demonstrate their suitability and creativity.

Abstract Microbial oils have gained massive attention because of their significant role in industrial applications. Currently plants and animals are the chief sources of medically and nutritionally important fatty acids. However, the ever-increasing global demand for polyunsaturated fatty acids (PUFAs) cannot be met by the existing sources. Therefore microbes, especially fungi, represent an important alternative source of microbial oils being investigated. Mucor circinelloides—an oleaginous filamentous fungus, came to the forefront because of its high efficiency in synthesizing and accumulating lipids, like γ-linolenic acid (GLA) in high quantity. Recently, mycelium of M. circinelloides has acquired substantial attraction towards it as it has been suggested as a convenient raw material source for the generation of biodiesel via lipid transformation. Although M. circinelloides accumulates lipids naturally, metabolic engineering is found to be important for substantial increase in their yields. Both modifications of existing pathways and re-formation of biosynthetic pathways in M. circinelloides have shown the potential to improve lipid levels. In this review, recent advances in various important metabolic aspects of M. circinelloides have been discussed. Furthermore, the potential applications of M. circinelloides in the fields of antioxidants, nutraceuticals, bioremediation, ethanol production, and carotenoids like beta carotene and astaxanthin having significant nutritional value are also deliberated.

Background: T-box family members are transcription factors characterized by highly conserved residues corresponding to the DNA-binding domain known as the T-box. TBX2 has been implicated in several developmental processes, such as coordinating cell fate, patterning, and morphogenesis of a wide range of tissues and organs, including lungs, limbs, heart, kidneys, craniofacial structures, and mammary glands.Methods: In the present study, we have clinically and genetically characterized a proband showing a severe form of chondrodysplasia with developmental delay. Whole-exome sequencing (WES), Sanger sequencing, and 3D protein modeling were performed in the present investigation.Results: Whole-exome sequencing revealed a novel nonsense variant (c.529A>T; p.Lys177*; NM_005994.4) in TBX2. 3D-TBX2 protein modeling revealed a substantial reduction of the mutated protein, which might lead to a loss of function (LOF) or nonsense-mediated decay (NMD).Conclusion: This study has not only expanded the mutation spectrum in the gene TBX2 but also facilitated the diagnosis and genetic counseling of related features in affected families.

Carotenoids are lipid-soluble compounds that are present in nature, including plants and microorganisms such as fungi, certain bacteria, and algae. In fungi, they are widely present in almost all taxonomic classifications. Fungal carotenoids have gained special attention due to their biochemistry and the genetics of their synthetic pathway. The antioxidant potential of carotenoids may help fungi survive longer in their natural environment. Carotenoids may be produced in greater quantities using biotechnological methods than by chemical synthesis or plant extraction. The initial focus of this review is on industrially important carotenoids in the most advanced fungal and yeast strains, with a brief description of their taxonomic classification. Biotechnology has long been regarded as the most suitable alternative way of producing natural pigment from microbes due to their immense capacity to accumulate these pigments. So, this review mainly presents the recent progress in the genetic modification of native and non-native producers to modify the carotenoid biosynthetic pathway for enhanced carotenoid production, as well as factors affecting carotenoid biosynthesis in fungal strains and yeast, and proposes various extraction methods to obtain high yields of carotenoids in an attempt to find suitable greener extraction methods. Finally, a brief description of the challenges regarding the commercialization of these fungal carotenoids and the solution is also given.

Canthaxanthin is a reddish-orange xanthophyll with strong antioxidant activity and higher bioavailability than carotenes, primarily used in food, cosmetics, aquaculture, and pharmaceutical industries. The spiking market for natural canthaxanthin promoted researchers toward genetic engineering of heterologous hosts for canthaxanthin production. Mucor circinelloides is a dimorphic fungus that produces β-carotene as the major carotenoid and is considered as a model organism for carotenogenic studies. In this study, canthaxanthin-producing M. circinelloides strain was developed by integrating the codon-optimized β-carotene ketolase gene (bkt) of the Haematococcus pluvialis into the genome of the fungus under the control of strong promoter zrt1. First, a basic plasmid was constructed to disrupt crgA gene, a negative regulator of carotene biosynthesis resulted in substantial β-carotene production, which served as the building block for canthaxanthin by further enzymatic reaction of the ketolase enzyme. The genetically engineered strain produced a significant amount (576 ± 28 μg/g) of canthaxanthin, which is the highest amount reported in Mucor to date. Moreover, the cell dry weight of the recombinant strain was also determined, producing up to more than 9.0 g/L, after 96 h. The mRNA expression level of bkt in the overexpressing strain was analyzed by RT-qPCR, which increased by 5.3-, 4.1-, and 3-folds at 24, 48, and 72 h, respectively, compared with the control strain. The canthaxanthin-producing M. circinelloides strain obtained in this study provided a basis for further improving the biotechnological production of canthaxanthin and suggested a useful approach for the construction of more valuable carotenoids, such as astaxanthin.

Education is a corner stone for the social, economic, cultural and political development of any society. This research articulates the impact of Higher Education in the process of peacebuilding and violence free coexistence of post 9/11 context in Khyber Pakhtunkhwa (KPK) prior and after Federally Administered Tribal Area (FATA) amalgamation. The study critically reviews the educational structure, policies, emphasis, application of educational goals, achievements from rural to Higher education and protecting measures enabling the youth to get easy access to education. Since the creation of Pakistan emphasis laid down on the free and compulsory education for all but failed to address the nascent challenges surfaced in the form of violent extremism, sectarianism, intolerance, and lingual discrepancies. Although the literacy rate has been increased with the passage of time but the quality of education and practical application remains under darkness. The study has analyzed the role of higher education for community development, peace building in post conflicted areas such as Bosnia, Rwanda, Sudan, and Spain with the adoption of global approaches of stabilization, state-building, and civic training. But does this happen in Khyber Pakhtunkhwa? It has been established through empirical analyses that the framework of higher education has lain dormant and beyond the limits of easy access for the man in the street. The higher education role has been ignored in peace building and reconstruction era. This research in its findings proposes that the structure and curriculum of higher education be revisited and expanded promptly in Khyber Pakhtunkhwa (KPK) particularly within the newly merged areas for the restoration of social justice, cultural diversity, educational growth, state-building and peacebuilding community.

Traditional power generating technologies rely on fossil fuels, which contribute to worldwide environmental issues such as global warming and climate change. As a result, renewable energy sources (RESs) are used for power generation where battery energy storage systems (BESSs) are widely used to store electrical energy for backup, match power consumption and generation during peak hours, and promote energy efficiency in a pollution-free environment. Accurate battery state of health (SOH) prediction is critical because it plays a key role in ensuring battery safety, lowering maintenance costs, and reducing BESS inconsistencies. The precise power consumption forecasting is critical for preventing power shortage and oversupply, and the complicated physicochemical features of batteries dilapidation cannot be directly acquired. Therefore, in this paper, a novel hybrid architecture called ‘CL-Net’ based on convolutional long short-term memory (ConvLSTM) and long short-term memory (LSTM) is proposed for multi-step SOH and power consumption forecasting. First, battery SOH and power consumption-related raw data are collected and passed through a preprocessing step for data cleansing. Second, the processed data are fed into ConvLSTM layers, which extract spatiotemporal features and form their encoded maps. Third, LSTM layers are used to decode the encoded features and pass them to fully connected layers for final multi-step forecasting. Finally, a comprehensive ablation study is conducted on several combinations of sequential learning models using three different time series datasets, i.e., national aeronautics and space administration (NASA) battery, individual household electric power consumption (IHEPC), and domestic energy management system (DEMS). The proposed CL-Net architecture reduces root mean squared error (RMSE) up to 0.13 and 0.0052 on the NASA battery and IHEPC datasets, respectively, compared to the state-of-the-arts. These experimental results show that the proposed architecture can provide robust and accurate SOH and power consumption forecasting compared to the state-of-the-art.

Renewable energy (RE) power plants are deployed globally because the renewable energy sources (RESs) are sustainable, clean, and environmentally friendly. However, the demand for power increases on a daily basis due to population growth, technology, marketing, and the number of installed industries. This challenge has raised a critical issue of how to intelligently match the power generation with the consumption for efficient energy management. To handle this issue, we propose a novel architecture called ‘AB-Net’: a one-step forecast of RE generation for short-term horizons by incorporating an autoencoder (AE) with bidirectional long short-term memory (BiLSTM). Firstly, the data acquisition step is applied, where the data are acquired from various RESs such as wind and solar. The second step performs deep preprocessing of the acquired data via several de-noising and cleansing filters to clean the data and normalize them prior to actual processing. Thirdly, an AE is employed to extract the discriminative features from the cleaned data sequence through its encoder part. BiLSTM is used to learn these features to provide a final forecast of power generation. The proposed AB-Net was evaluated using two publicly available benchmark datasets where the proposed method obtains state-of-the-art results in terms of the error metrics.

Nowadays, for efficient energy management, local demand-supply matching in power grid is emerging research domain. However, energy demand is increasing day by day in many countries due to rapid growth of the population and most of their work being reliant on electronic devices. This problem has highlighted the significance of effectively matching power demand with supply for optimal energy management. To resolve this issue, we present an intelligent deep learning framework that integrates Atrous Convolutional Layers (ACL) with Residual Gated Recurrent Units (RGRU) to establish balance between the demand and supply. Moreover, it accurately predicts short-term energy and delivers a systematic method of communication between consumers and energy distributors as well. To cope with the varying nature of electricity data, first data acquisition step is performed where data are collected from various sources such as smart meters and solar plants. In the second step a pre-processing method is applied on raw data to normalize and clean the data. Next, the refined data are passed to ACL for spatial feature extraction. Finally, a sequential learning model RGRU is used that learns from complicated patterns for the final output. The proposed model obtains the smallest values of Mean Square Error (MSE) including 0.1753, 0.0001, 0.0177 over IHEPC, KCB, and Solar datasets, respectively, which manifests better performance as compared to existing approaches.

The use of electrical energy is directly proportional to the increase in global population, both concerning growing industrialization and rising residential demand. The need to achieve a balance between electrical energy production and consumption inspires researchers to develop forecasting models for optimal and economical energy use. Mostly, the residential and industrial sectors use metering sensors that only measure the consumed energy but are unable to manage electricity. In this paper, we present a comparative analysis of a variety of deep features with several sequential learning models to select the optimized hybrid architecture for energy consumption prediction. The best results are achieved using convolutional long short-term memory (ConvLSTM) integrated with bidirectional long short-term memory (BiLSTM). The ConvLSTM initially extracts features from the input data to produce encoded sequences that are decoded by BiLSTM and then proceeds with a final dense layer for energy consumption prediction. The overall framework consists of preprocessing raw data, extracting features, training the sequential model, and then evaluating it. The proposed energy consumption prediction model outperforms existing models over publicly available datasets, including Household and Korean commercial building datasets.

In the current technological era, energy-efficient buildings have a significant research body due to increasing concerns about energy consumption and its environmental impact. Designing an appropriate energy-efficient building depends on its layout, such as relative compactness, overall area, height, orientation, and distribution of the glazing area. These factors directly influence the cooling load (CL) and heating load (HL) of residential buildings. An accurate prediction of these load facilitates a better management of energy consumption and enhances the living standards of inhabitants. Most of the traditional machine learning (ML)-based approaches are designed for single-output (SO) prediction, which is a tedious task due to separate training processes for each output with low performance. In addition, these approaches have a high level of nonlinearity between input and output, which need more enhancement in terms of robustness, predictability, and generalization. To tackle these issues, we propose a novel framework based on gated recurrent unit (GRU) that reliably predicts the CL and HL concurrently. To the best of our knowledge, we are the first to propose a multi-output (MO) sequential learning model followed by utility preprocessing under the umbrella of a unified framework. A comprehensive set of ablation studies on ML and deep learning (DL) techniques is done over an energy efficiency dataset, where the proposed model reveals an incredible performance as compared to other existing models.

Multi-criteria decision-making (MCDM) plays a vibrant role in decision-making, and the characteristic object method (COMET) acts as a powerful tool for decision-making of complex problems. COMET technique allows using both symmetrical and asymmetrical triangular fuzzy numbers. The COMET technique is immune to the pivotal challenge of rank reversal paradox and is proficient at handling vagueness and hesitancy. Classical COMET is not designed for handling uncertainty data when the expert has a problem with the identification of the membership function. In this paper, symmetrical and asymmetrical normalized interval-valued triangular fuzzy numbers (NIVTFNs) are used for decision-making as the solution of the identified challenge. A new MCDM method based on the COMET method is developed by using the concept of NIVTFNs. A simple problem of MCDM in the form of an illustrative example is given to demonstrate the calculation procedure and accuracy of the proposed approach. Furthermore, we compare the solution of the proposed method, as interval preference, with the results obtained in the Technique for Order of Preference by Similarity to Ideal solution (TOPSIS) method (a certain preference number).

The goal of this article is to establish some fractional proportional integral inequalities for convex functions by employing proportional fractional integral operators. In addition, we establish some classical integral inequalities as the special cases of our main findings.

Movies have become one of the major sources of entertainment in the current era, which are based on diverse ideas. Action movies have received the most attention in last few years, which contain violent scenes, because it is one of the undesirable features for some individuals that is used to create charm and fantasy. However, these violent scenes have had a negative impact on kids, and they are not comfortable even for mature age people. The best way to stop under aged people from watching violent scenes in movies is to eliminate these scenes. In this paper, we proposed a violence detection scheme for movies that is comprised of three steps. First, the entire movie is segmented into shots, and then a representative frame from each shot is selected based on the level of saliency. Next, these selected frames are passed from a light-weight deep learning model, which is fine-tuned using a transfer learning approach to classify violence and non-violence shots in a movie. Finally, all the non-violence scenes are merged in a sequence to generate a violence-free movie that can be watched by children and as well violence paranoid people. The proposed model is evaluated on three violence benchmark datasets, and it is experimentally proved that the proposed scheme provides a fast and accurate detection of violent scenes in movies compared to the state-of-the-art methods.

In the recent era, bioactive compounds from plants have received great attention because of their vital health-related activities, such as antimicrobial activity, antioxidant activity, anticoagulant activity, anti-diabetic activity, UV protection, antiviral activity, hypoglycemia, etc. Previous studies have already shown that polysaccharides found in plants are not likely to be toxic. Based on these inspirational comments, most research focused on the isolation, identification, and bioactivities of polysaccharides. A large number of biologically active polysaccharides have been isolated with varying structural and biological activities. In this review, a comprehensive summary is provided of the recent developments in the physical and chemical properties as well as biological activities of polysaccharides from a number of important natural sources, such as wheat bran, orange peel, barely, fungi, algae, lichen, etc. This review also focused on biomedical applications of polysaccharides. The contents presented in this review will be useful as a reference for future research as well as for the extraction and application of these bioactive polysaccharides as a therapeutic agent.

Task scheduling and resource allocation are the key issues for computational grids. Distributed resources usually work at different autonomous domains with their own access and security policies that impact successful job executions across the domain boundaries. In this paper we present a security-aware grid simulator Secure-Sim-G, which facilitates the evaluation of the different scheduling heuristics under various scheduling criteria in several grid scenarios defined by the security conditions, grid size and system dynamics. The simulator allows the flexible activation or inactivation of all of the scheduling criteria and modules, which makes the application well adapted to the proper illustration of the different realistic scenarios and avoids the possible restriction to the specific scheduling resolution methods. The simulation results and traces may be graphically represented and stored at the server and can retrieved in different formats such as spreadsheets or pdf files.

Multicell cooperation has been identified as one of the underlying principles for future wireless communication systems. This paper studies the benefits of multicell cooperation in broadcast TV network from an information theoretical perspective. We define outage capacity as the figure of merit and derive the broadcast coverage area to evaluate such system. Specifically, we calculate the broadcast coverage area with given common information rate and outage probabilities when multiple base stations collaboratively transmit the broadcast signals. For the general MIMO case where receivers have multiple antennas, we provide simulation results to illustrate the expanded coverage area. In all cases, our results show that the coverage of a TV broadcast network can be significantly improved by multicell cooperation.

Objective To compare the frequency of nail changes in Chronic Kidney Disease patients undergoing haemodialysis with patients on maintenance therapy. Methods This analytical cross-sectional study was carried out at Dermatology and Nephrology Department, Sahiwal Teaching Hospitals, Sahiwal, after approval from Institutional Ethical Review Committee. A total of 158 patients with chronic kidney disease of either sex and age were divided into two groups: Group A (patients under regular haemodialysis) Group B (patients on oral therapy). Patient's demographic, clinical and laboratory data were recorded in predesigned proforma. Data were analyzed by using SPSS version 28. Results Out of total 158 patients, male female ratio was 1.1:1 and majority of patients (63.3%) were above age of 40 years. Most commonly reported nail changes in both groups were absence of lunula (65%, 57%), longitudinal ridging (54%, 58%) and onycholysis (45%, 35%) but variations were observed in their frequencies. There was no statistically significant difference in nail changes of both groups. Conclusion Patients with CKD undergoing HD therapy have slight higher frequency of nail changes compared to patients on maintenance therapy. Further studies with large sample size may elaborate Results. [ABSTRACT FROM AUTHOR]

The growing demand for high-quality industrial products has led to a significant emphasis on image anomaly detection (AD). AD in industrial goods presents a formidable research challenge that demands the application of sophisticated techniques to identify and address deviations from the expected norm accurately. Manufacturers increasingly recognize the significance of employing intelligent systems to detect flaws and defects in product parts. However, industrial settings pose several challenges: diverse categories, limited abnormal samples and vagueness. Hence, there is a growing demand for advanced image AD techniques within industrial product manufacturing. In this paper, an intelligent industrial defective chips detection framework is proposed which mainly consists of three core components. First, the convolutional features of the efficient backbone model is effectively utilized to balance the computational complexity and performance of industrial resource-constrained devices. Secondly, a novel inverse feature matching followed by masking method is proposed to enhance the explanability that localizes the abnormal regions of the abnormal chips. Finally, to evaluate our proposed method a comprehensive ablation study is conducted, where different machine learning and deep learning algorithms are analysed to claim the superiority of our method. Furthermore, to help the research community, a benchmark dataset is collected from real-world industry manufacturing for defective chip detection. The empirical results from the dataset demonstrate the strength and effectiveness of the proposed model compared to the other models. [ABSTRACT FROM AUTHOR]

Energy management systems for residential and commercial buildingsmust use an appropriate and efficientmodel to predict energy consumption accurately. To deal with the challenges in powermanagement, the short-termPower Consumption (PC) prediction for household appliances plays a vital role in improving domestic and commercial energy efficiency. Big data applications and analytics have shown that data-driven load forecasting approaches can forecast PC in commercial and residential sectors and recognize patterns of electric usage in complex conditions. However, traditional Machine Learning (ML) algorithms and their features engineering procedure emphasize the practice of inefficient and ineffective techniques resulting in poor generalization. Additionally, different appliances in a home behave contrarily under distinct circumstances, making PC forecasting more challenging. To address these challenges, in this paper a hybrid architecture using an unsupervised learning strategy is investigated. The architecture integrates a one-dimensional Convolutional Neural Network (CNN) based Autoencoder (AE) and online sequential Extreme Learning Machine (ELM) for commercial and residential short-term PC forecasting. First, the load data of different buildings are collected and cleaned from various abnormalities. A subsequent step involves AE for learning a compressed representation of spatial features and sending them to the online sequential ELM to learn nonlinear relations and forecast the final load. Finally, the proposed network is demonstrated to achieve State-of-the-Art (SOTA) error metrics based on two benchmark PC datasets for residential and commercial buildings. The Mean Square Error (MSE) values obtained by the proposed method are 0.0147 and 0.0121 for residential and commercial buildings datasets, respectively. The obtained results prove that our model is suitable for the PC prediction of different types of buildings. [ABSTRACT FROM AUTHOR]

Photovoltaic (PV) systems are environmentally friendly, generate green energy, and receive support from policies and organizations. However, weather fluctuations make large-scale PV power integration and management challenging despite the economic benefits. Existing PV forecasting techniques (sequential and convolutional neural networks (CNN)) are sensitive to environmental conditions, reducing energy distribution system performance. To handle these issues, this article proposes an efficient, weather-resilient convolutional-transformer-based network (CT-NET) for accurate and efficient PV power forecasting. The network consists of three main modules. First, the acquired PV generation data are forwarded to the pre-processing module for data refinement. Next, to carry out data encoding, a CNNbased multi-head attention (MHA) module is developed in which a single MHA is used to decode the encoded data. The encoder module is mainly composed of 1D convolutional and MHA layers, which extract local as well as contextual features, while the decoder part includesMHA and feedforward layers to generate the final prediction. Finally, the performance of the proposed network is evaluated using standard error metrics, including the mean squared error (MSE), root mean squared error (RMSE), and mean absolute percentage error (MAPE). An ablation study and comparative analysis with several competitive state-of-the-art approaches revealed a lower error rate in terms of MSE (0.0471), RMSE (0.2167), and MAPE (0.6135) over publicly available benchmark data. In addition, it is demonstrated that our proposed model is less complex, with the lowest number of parameters (0.0135 M), size (0.106 MB), and inference time (2 ms/step), suggesting that it is easy to integrate into the smart grid. [ABSTRACT FROM AUTHOR]