In this paper, we present a new method for forecasting power consumption. Household consumption prediction is essential to manage and plan energy utilization. This study proposes technique using machine learning models based on the stationary wavelet transform (SWT) transformers forecast household in different resolutions. approach works by leveraging self-attention mechanisms learn complex patterns dynamics from data. The SWT its inverse are used decompose reconstruct actual forecasted...

The underwater environment presents unique challenges (color distortions, reduced contrast, blurriness) hindering accurate analysis. This work introduces MuLA-GAN, a novel approach leveraging Generative Adversarial Networks (GANs) and specifically adapted Multi-Level Attention for comprehensive image enhancement. MuLA-GAN integrates within the GAN architecture to prioritize learning discriminative features crucial precise restoration. These relevant encompass information on local details...

Semantic segmentation necessitates approaches that learn high-level characteristics while dealing with enormous amounts of data. Convolutional neural networks (CNNs) can unique and adaptive features to achieve this aim. However, due the large size high spatial resolution remote sensing images, these cannot analyze an entire scene efficiently. Recently, deep transformers have proven their capability record global interactions between different objects in image. In paper, we propose a new...

Early detection and accurate diagnosis of brain morphological abnormalities are essential for the effective management treatment Alzheimer's disease (AD) mild cognitive impairment (MCI). Structural magnetic resonance imaging (MRI) is a powerful support tool to aid in prediction. In this research study, we present an innovative approach predict (MCI) using MRI data, which integrates regional interest (ROI)-based methodology deep learning within comprehensible framework. The proposed method...

In this paper, a metacognitive octonion-valued neural network (Mc-OVNN) learning algorithm and its application to diverse time series prediction are presented. The Mc-OVNN is comprised of two components: the that represents cognitive component serves self-regulate algorithm. At each epoch, decides if, how, when occurs. deletes unneeded samples only stores those will be used. This decision determined by octonion magnitude seven phases. To evaluate algorithm's performance, it applied five...

The underwater domain presents unique challenges and opportunities for scientific exploration, resource extraction, environmental monitoring. Autonomous vehicles (AUVs) rely on simultaneous localization mapping (SLAM) real-time navigation in these complex environments. However, traditional SLAM techniques face significant obstacles, including poor visibility, dynamic lighting conditions, sensor noise, water-induced distortions, all of which degrade the accuracy robustness systems. Recent...

Accurate wind speed forecasting is a fundamental requirement for advanced and economically viable large-scale power integration. The hybridization of the quaternion-valued neural networks stationary wavelet transform has not been proposed before. In this paper, we propose novel wind-speed model that combines with networks. represents subbands in quaternion vectors, which avoid separating naturally correlated subbands. consists three main steps. first step, signal decomposed using into...

Rainfall–runoff modelling is at the core of any hydrological forecasting system. The high spatio-temporal variability precipitation patterns, complexity physical processes, and large quantity parameters required to characterize a watershed make prediction runoff rates quite difficult. In this study, hyper-complex artificial neural network in form an octonion-valued (OVNN) proposed estimate rates. Evaluation model performed using rainfall time series from raingauge near Canadian watershed....

In this article, a metacognitive sedenion-valued neural network (Mc-SVNN) and its learning algorithm are proposed. Its application to diverse time-series prediction problems is presented. The Mc-SVNN contains two components: that represents the cognitive component, which serves self-regulate algorithm. At each epoch, component decides what, how, when occurs. deletes unnecessary samples stores only those used. This decision determined by sedenion magnitude 15 phases. applied four real-world...

In the realm of exoskeleton control, achieving precise control poses challenges due to mechanical delay exoskeletons. To address this, incorporating future gait trajectories as feed-forward input has been proposed. However, existing deep learning models for prediction mainly focus on short-term predictions, leaving long-term performance these relatively unexplored. this study, we present TempoNet, a novel model specifically designed knee joint angle prediction. By harnessing dynamic temporal...

<title>Abstract</title> Early detection and accurate diagnosis of brain morphological abnormalities are essential for the effective management treatment Alzheimer's disease (AD) mild cognitive impairment (MCI). Structural magnetic resonance imaging (MRI) is a powerful support tool to aid in prediction. In this research study, we present an innovative approach predict (MCI) using MRI data, which integrates regional interest (ROI)-based methodology deep learning within comprehensible...

Efficient use of solar energy requires reliable forecasting values. In this paper, we propose a system combining the stationary wavelet transform and quaternion-valued neural networks (SWT-QVNN) it to forecast irradiance. addition, quaternion variant softplus AMSGrad learning algorithm is developed used in optimizing network. The proposed was tested using irradiance data from two cities Abu Dhabi, UAE, Tamanrasset, Algeria. It did reduce Root Mean Squared Error (RMSE) by more than 60%...

In this paper, a forecasting of the global solar irradiation in complex-valued domain is proposed. A method to transform meteorological data into complex values developed and Complex Valued Neural Network (CVNN) used model forecast daily hourly irradiation. The measured Tamanrasset city, Algeria (altitude: 1362 m; latitude: 22°48 N; longitude: 05°26 E) validate model. case, 24 h ahead will be forecasted using combination past dataset. Several models are presented test feasibility performance...

In this paper, a new architecture combining dynamic neural units and fuzzy logic approaches is proposed for complex chemical process modeling.Such processes need particular care where the designer constructs network, network models which are very useful in black box modeling.The specified to pH reactor due its large existence real industrial life it realistic nonlinear system demonstrate feasibility performance of founding results using units.A comparison was made between four strategies,...

In this study, the identification of bioprocesses using Random Search (RS) and Simulated Annealing (SA) algorithms is considered. The high nonlinearity large application give them opportunity to be challenging systems for several conventional techniques that cannot used process. derivative free algorithms, in which gradient problem optimized not required, could parameters. Two have been chosen RS SA are fed batch fermentor bioreactor. Simulation results presented above cases found...