AI-Based Virtual Assistant for Solar Radiation Prediction and Improvement of Sustainable Energy Systems

Chapter 12 - Hybrid artificial intelligence models for predicting daily runoff

The growing reliance on renewable energy, particularly solar power, presents significant challenges to grid stability due to the variability and intermittency of energy generation. Hybrid Artificial Intelligence (AI) models have emerged as a transformative solution, integrating multiple AI techniques to address these challenges effectively. This study systematically reviewed 130 peer-reviewed articles, adhering to the Preferred Reporting Items for Systematic Reviews and Meta-Analyses (PRISMA) guidelines to ensure a transparent and rigorous process. The review highlights how hybrid AI models, by combining neural networks, reinforcement learning, evolutionary algorithms, and advanced architectures like transformers, achieve superior performance in predictive modeling, fault detection, load balancing, and scalability. Predictive modeling for solar power variability saw up to a 30% improvement in mean absolute error (MAE), while fault detection systems exhibited enhanced precision and recall, diagnosing multiple simultaneous faults in real time. Hybrid AI models also excel in load balancing and demand forecasting, reducing energy wastage by 20% and enabling dynamic resource allocation. Furthermore, their scalability and adaptability make them ideal for large-scale, distributed energy networks, addressing the complexities of modern smart grids. Despite these advancements, persistent challenges such as real-time data integration, standardization, and deployment barriers remain. The study underscores the need for future research to address these limitations through innovations like federated learning and decentralized AI frameworks. This comprehensive review demonstrates the critical role of hybrid AI in advancing grid stability, optimizing renewable energy integration, and paving the way for sustainable, resilient energy infrastructures.

Diagnosing rare diseases remains a significant challenge in healthcare due to their complex nature, low prevalence, and limited clinical data. Traditional diagnostic methods often struggle to detect these conditions in a timely and accurate manner, leading to delayed treatments and poor patient outcomes. In recent years, hybrid artificial intelligence (AI) models have emerged as a promising solution, integrating multiple AI techniques such as machine learning, deep learning, natural language processing, and expert systems to improve the diagnostic process. These hybrid models offer the potential to analyze diverse data sources, including genetic, clinical, and imaging data, to identify rare diseases with greater precision. This paper explores the concept of hybrid AI models and their applications in rare disease diagnosis, highlighting their ability to improve diagnostic accuracy, reduce delays, and enhance personalized treatment. We also discuss the challenges and limitations of hybrid AI, including data scarcity, model interpretability, and ethical concerns, as well as regulatory hurdles for clinical adoption. Additionally, we examine the role of data sources like electronic health records, genomic data, and medical imaging in training these models, along with ethical considerations surrounding privacy, bias, and transparency. Finally, the paper looks toward future directions for hybrid AI in rare disease diagnosis, focusing on emerging technologies such as explainable AI, federated learning, and multi-modal data integration. By addressing these challenges and innovations, hybrid AI models have the potential to revolutionize the diagnosis and treatment of rare diseases, leading to better patient outcomes and more efficient healthcare systems.

Prediction of sediment heavy metal at the Australian Bays using newly developed hybrid artificial intelligence models.

The coronavirus disease 2019 (COVID-19) breaking out in late December 2019 is gradually being controlled in China, but it is still spreading rapidly in many other countries and regions worldwide. It is urgent to conduct prediction research on the development and spread of the epidemic. In this article, a hybrid artificial-intelligence (AI) model is proposed for COVID-19 prediction. First, as traditional epidemic models treat all individuals with coronavirus as having the same infection rate, an improved susceptible-infected (ISI) model is proposed to estimate the variety of the infection rates for analyzing the transmission laws and development trend. Second, considering the effects of prevention and control measures and the increase of the public's prevention awareness, the natural language processing (NLP) module and the long short-term memory (LSTM) network are embedded into the ISI model to build the hybrid AI model for COVID-19 prediction. The experimental results on the epidemic data of several typical provinces and cities in China show that individuals with coronavirus have a higher infection rate within the third to eighth days after they were infected, which is more in line with the actual transmission laws of the epidemic. Moreover, compared with the traditional epidemic models, the proposed hybrid AI model can significantly reduce the errors of the prediction results and obtain the mean absolute percentage errors (MAPEs) with 0.52%, 0.38%, 0.05%, and 0.86% for the next six days in Wuhan, Beijing, Shanghai, and countrywide, respectively.

With the increasing urgency for sustainable development and energy transition, decision-makers face complex challenges in evaluating and prioritizing viable alternatives. Traditional decision-making techniques often struggle to capture the inherent uncertainty and imprecision associated with the latest sustainable energy transition issues. This paper presents a research framework based on fuzzy set theory and the technique for order of preference by similarity to ideal solution (TOPSIS) method to address these complexities and uncertainties. Our proposed approach offers a comprehensive evaluation and ranking of alternatives for sustainable energy transition. To demonstrate the effectiveness and applicability of this system, we employ a case study in the Kingdom of Saudi Arabia (KSA). As a global leader in fossil fuel production and export, particularly oil, the KSA has recognized the need to address climate change and diversify its energy sector. By leveraging the fuzzy TOPSIS-based framework, we provide decision-makers with a powerful tool to navigate the challenges and uncertainties involved in the energy transition process. This research yields promising results, demonstrating the superior capabilities of the proposed fuzzy TOPSIS-based framework compared to traditional decision-making techniques. The case study in the KSA highlights how our approach effectively captures and addresses the uncertainties and complexities involved in sustainable energy transition decision making. Through comprehensive evaluations and rankings, decision-makers gain valuable insights into alternative solutions, facilitating informed and strategic decision-making processes. Our research contributes to sustainable energy transitions by introducing a robust decision-making framework that integrates fuzzy set theory and the TOPSIS method. Based on the fuzzy TOPSIS-based evaluation, the research findings indicate that solar energy (EA1) ranked as the most favourable alternative among the evaluated options for the sustainable energy transition in the KSA. Using our framework, stakeholders in the KSA and similar contexts can make informed decisions to accelerate their energy transition efforts and achieve sustainable development goals.

Hybridized artificial intelligence models with nature-inspired algorithms for river flow modeling: A comprehensive review, assessment, and possible future research directions

Study on prediction of ocean effective wave height based on hybrid artificial intelligence model

Predicting uniaxial compressive strength of oil palm shell concrete using a hybrid artificial intelligence model

This bibliometric study introduces a novel approach to assessing the application of artificial intelligence (AI) in dentistry. It analyzes trends in AI utilization across dental disciplines, treatment stages, data modalities, subsets, models, and tasksand proposes a comprehensive classification framework for AI applications in dentistry. A systematic search in the Web of Science Core Collection on December 1, 2024, using AI- and dentistry-related keywordsidentified original and review articles employing true AI. Data on publication details, study types, dental disciplines, treatment stages, AI subsets, models, data modalities, and tasks were extracted and analyzed using VOSviewer (Leiden University,Leiden, Netherlands) and Microsoft Excel (Microsoft Corp., Redmond, WA). Trend analysis and forecasting methods were applied to identify future research directions. Of 2,810 records, 1,368 studies met the inclusion criteria, revealing a continuous rise in AI-related dental research. While most studies focused on diagnostic applications and the orthodontics discipline, the highest recent growth was seen in treatment planning and research and education applications. Hybrid AI models and natural language processing (NLP) experienced significant increases in adoption. The most common AI tasks were classification, detection, and segmentation, although notable growth occurred in generation, data integration, and decision support. The classification framework for AI in dentistry is presented. Text-based data have shown the greatest growth among data modalities, alongside an increased use of sensor and signal data. Future research should prioritize developing NLP and hybrid AI models, conducting original studies in research and educationand treatment planning, and undertaking systematic reviews focused on the diagnosis stage of prosthodontics and endodontics.

PurposeTo develop, and evaluate the performance of, a deep learning‐based three‐dimensional (3D) convolutional neural network (CNN) artificial intelligence (AI) algorithm aimed at finding needles in ultrasound images used in prostate brachytherapy.MethodsTransrectal ultrasound (TRUS) image volumes from 1102 treatments were used to create a clinical ground truth (CGT) including 24422 individual needles that had been manually digitized by medical physicists during brachytherapy procedures. A 3D CNN U‐net with 128 × 128 × 128 TRUS image volumes as input was trained using 17215 needle examples. Predictions of voxels constituting a needle were combined to yield a 3D linear function describing the localization of each needle in a TRUS volume. Manual and AI digitizations were compared in terms of the root‐mean‐square distance (RMSD) along each needle, expressed as median and interquartile range (IQR). The method was evaluated on a data set including 7207 needle examples. A subgroup of the evaluation data set (n = 188) was created, where the needles were digitized once more by a medical physicist (G1) trained in brachytherapy. The digitization procedure was timed.ResultsThe RMSD between the AI and CGT was 0.55 (IQR: 0.35–0.86) mm. In the smaller subset, the RMSD between AI and CGT was similar (0.52 [IQR: 0.33–0.79] mm) but significantly smaller (P < 0.001) than the difference of 0.75 (IQR: 0.49–1.20) mm between AI and G1. The difference between CGT and G1 was 0.80 (IQR: 0.48–1.18) mm, implying that the AI performed as well as the CGT in relation to G1. The mean time needed for human digitization was 10 min 11 sec, while the time needed for the AI was negligible.ConclusionsA 3D CNN can be trained to identify needles in TRUS images. The performance of the network was similar to that of a medical physicist trained in brachytherapy. Incorporating a CNN for needle identification can shorten brachytherapy treatment procedures substantially.

The accurate prediction of implant failure is critical for enhancing patient outcomes, minimizing revision surgeries, and improving the longevity of biomedical implants. Traditional failure prediction methods, including statistical models and physics-based simulations, often lack the adaptability required for patient-specific risk assessments. Recent advancements in artificial intelligence (AI) and machine learning (ML) have enabled the development of predictive models that leverage diverse data sources, including medical imaging, biomechanical simulations, and real-time sensor data. This book chapter explores the integration of advanced machine learning techniques, such as gradient boosting, deep learning architectures, and hybrid AI models, to enhance implant failure detection and optimize implant longevity. The role of hybrid AI models, which combine physics-based simulations with data-driven approaches, is particularly emphasized for improving prediction accuracy in patient-specific cases. Additionally, ensemble learning methods, including voting classifiers and hybrid CNN-RNN frameworks, are discussed for their ability to process multimodal medical data, ensuring a comprehensive failure assessment. The implementation of smart implant technologies, real-time biomechanical monitoring, and explainable AI (XAI) techniques is explored to enhance clinical decision-making and model interpretability. The challenges associated with AI-driven implant failure prediction, including data heterogeneity, model generalizability, and computational complexity, are also addressed. By integrating AI-driven predictive frameworks with clinical workflows, this chapter highlights the potential of machine learning to revolutionize implant failure detection, enabling early intervention and personalized treatment strategies for improved patient care.

Deep sequence modelling for Alzheimer's disease detection using MRI

Reducing household energy use in social housing buildings can substantially contribute to mitigating global climate change. While municipalities and social housing corporations are willing to invest in sustainable renovations and innovations, social housing residents' inclusion in the sustainable energy transition lags behind. This pilot study explored social housing residents' attitudes toward sustainability and sustainable renovation of their apartment building, as well as (factors underlying) their motivation toward two specific sustainable behaviors. Semi-structured interviews, containing both open- and close- ended questions, were conducted with 20 residents of one social housing building that was due for renovations. Results showed that respondents were concerned about climate change, including environmental justice beliefs, typically already engaged in various sustainable behaviors, and were motivated to add sustainable behaviors to their repertoire after the renovation. Yet, perceived social norms were not always supportive of behaving sustainably and respondents sometimes failed to recognize the sustainable value of these behaviors. Furthermore, while respondents were more positive than negative about the sustainable renovation, they nevertheless listed many concerns and problems regarding the renovation process, including procedural justice concerns. This small-scale study provided important insights into barriers and facilitators of the sustainable energy transition among social housing residents, who are at risk of lagging behind in the sustainable urban energy transition. Findings underline the importance of including residents in the sustainable renovation process through engagement, communication, and co-creation.

The global demand for sustainable and clean energy sources has driven significant advancements in the field of geothermal energy. This paper provides a comprehensive review of the latest technological developments and environmental impacts associated with the advancement of geothermal energy. Technological developments in geothermal energy have focused on enhancing efficiency, scalability, and cost-effectiveness. Innovations in drilling techniques, such as enhanced geothermal systems (EGS), have expanded the potential for harnessing geothermal resources in previously untapped regions. Furthermore, advancements in materials science and reservoir management techniques have contributed to increased energy extraction and prolonged reservoir life. This review also delves into the environmental impacts of geothermal energy, addressing both the positive and negative aspects. Geothermal power generation produces minimal greenhouse gas emissions compared to traditional fossil fuels, contributing to a cleaner and more sustainable energy landscape. However, concerns about induced seismicity, subsurface fluid management, and the potential release of trace gases during geothermal operations require careful consideration. The integration of geothermal energy into the broader energy mix is explored, emphasizing its role in reducing reliance on fossil fuels and mitigating climate change. Additionally, the review discusses the importance of regulatory frameworks and community engagement in ensuring responsible geothermal development. This paper highlights the dynamic landscape of geothermal energy, showcasing the progress made in technology and the environmental considerations that accompany its expansion. As the world seeks alternative energy sources to address climate change and energy security, understanding the evolving nature of geothermal energy is crucial for informed decision-making and sustainable energy transitions. Keywords: Energy, Geothermal, Environmental Impact, Development, Review.