Machine Learning Applications in Building Energy Systems: Review and Prospects

Buildings, Journal Year: 2025, Volume and Issue: 15(4), P. 648 - 648

Published: Feb. 19, 2025

Building energy systems (BESs) are essential for modern infrastructure but face significant challenges in equipment diagnosis, consumption prediction, and operational control. The complexity of BESs, coupled with the increasing integration renewable sources, presents difficulties fault detection, accurate forecasting, dynamic system optimisation. Traditional control strategies struggle low efficiency, slow response times, limited adaptability, making it difficult to ensure reliable operation optimal management. To address these issues, researchers have increasingly turned machine learning (ML) techniques, which offer promising solutions improving scheduling, real-time BESs. This review provides a comprehensive analysis ML techniques applied According results literature review, supervised methods, such as support vector machines random forest, demonstrate high classification accuracy detection require extensive labelled datasets. Unsupervised approaches, including principal component clustering algorithms, robust identification capabilities without data may complex nonlinear patterns. Deep particularly convolutional neural networks long short-term memory models, exhibit superior forecasting Reinforcement further enhances management by dynamically adjusting parameters maximise efficiency cost savings. Despite advancements, remain terms availability, computational costs, model interpretability. Future research should focus on hybrid integrating explainable AI enhancing adaptability evolving demands. also highlights transformative potential BESs outlines future directions sustainable intelligent building

Language: Английский

---

Artificial Intelligence in Offshore Infrastructure

Lecture notes in civil engineering, Journal Year: 2025, Volume and Issue: unknown, P. 13 - 20

Published: Jan. 1, 2025

Language: Английский

---

Practical inspection and exploration of assistive positioning and generative artificial intelligence empowering civil trials

Published: May 1, 2025

With the rapid development of artificial intelligence technology, application AI in judicial field is gradually expanding, especially civil trials, where auxiliary positioning and generative technology show great potential. Through intelligent case analysis, legal text retrieval, matching, document generation, can effectively improve efficiency, ensure quality judgments, reduce workload judges, promote fairness. However, still faces challenges such as technical adaptation, data privacy, ethical issues. This paper aims to explore how empower reveal its potential prospects by analyzing current status, principles, practical effects, trials. Studies have shown that introduction efficiency consistency but it also poses new system framework. In future, with continuous advancement improvement policy frameworks, will play an increasingly important role trials contribute transformation field.

Language: Английский

---

Intelligent Diagnosis of Urban Underground Drainage Network: From Detection to Evaluation

Structural Control and Health Monitoring, Journal Year: 2024, Volume and Issue: 2024(1)

Published: Jan. 1, 2024

During the process of urban development, there is large‐scale laying underground pipeline networks and coordinated operation both new old networks. The concrete drainage pipes have become a focus maintenance due to their strong concealment serious corrosion. current manual inspections for subterranean pipelines involve high workloads risks, which makes meeting diagnostic needs intricate challenging. Through advanced information technology, it has reached consensus intelligently perceive, accurately identify, precise prediction condition development detection evaluation methods pipe this study. study discusses common algorithms classifying, locating, quantifying defects by combining principles deep learning with typical application examples. intelligent progression collection methods, image processing techniques, damage models, systems systematically elaborated upon. Lastly, prospects future research diagnosis are provided.

Language: Английский

---

Integration of Building Information Modeling (BIM) and Big Data in China: Recent Application and Future Perspective

Buildings, Journal Year: 2023, Volume and Issue: 13(10), P. 2435 - 2435

Published: Sept. 25, 2023

The integrated application of building information modeling (BIM) and big data (BD) has received widespread attention, been involved in smart construction sites, project management, budgeting. Nevertheless, research on the implementation BIM BD China mainly concentrates a stage or profession, exploration technology integration mostly focuses theoretical level, distribution is scattered. As such, intention this paper to reveal history China, as well study methodologies fields for more thorough knowledge development status Chinese sector, which adopts mixed method that uses quantitative via two analytical software tools, i.e., CiteSpace version 6.1.R6 Statistical Analysis Toolkit line edition Informetrics packages, conduct macro bibliometric analysis National Knowledge Infrastructure database, provides follow-up micro qualitative with content analysis. To ensure comprehensiveness research, core articles topic web science database have sorted out analyzed fully understanding field construction, resulting identifying current hotspots trends China. results suggest popular keywords since year 2015 focused informatization, internet things, rail transportation. Three fruitful themes identified, including operation, bridge informatization.

Language: Английский

---

Estimativa da carga de ruptura em estacas a partir de ensaios de prova de carga estática utilizando machine learning

Programa de Iniciação Científica - PIC/UniCEUB - Relatórios de Pesquisa, Journal Year: 2024, Volume and Issue: unknown

Published: Feb. 28, 2024

A investigação do sistema solo-fundação é de extrema importância para a construção edificações seguras. Uma das formas geotécnica bastante utilizada o ensaio SPT (Standard Penetration Test). Este utilizado estimar resistência penetração solo, tipo presença nível d’água e dependendo equipamento, atrito lateral solo. Além SPT, realização projetos fundações mais seguros econômicos, tem-se prova carga estática, qual permite verificar desempenho fundações. um método análise direta da capacidade suporte carga, podendo ser realizada em diversos tipos estruturas. Porém, apesar ideal não com frequência, devido ao seu custo à obrigatoriedade norma se utilizar somente obras grande porte. Sendo assim, esse estudo estimou ruptura estacas carregadas axialmente, por meio machine learning, partir ensaios provas estáticas sondagens percussão existentes situadas no Distrito Federal.Para isso, foi desenvolvido algoritmo linguagem phyton que pudesse, dos dados treino, prever os resultados apresentassem apenas sondagem simples. Para desenvolvimento foram utilizados modelos aprendizado supervisionado (Random Forest). Desta forma, possível obter acurácia 78,12%. vale ressaltar existem limitações relação resultado final, influenciaram valor acurácia, tais como: número limitado amostragem (67 carga), variabilidade perfil estratigráfico associado aos pontuais dificulta representatividade condições solo local próprio Forest) quanto maior dispersão produz menor previsão.

---

A influência de informação sobre o desenvolvimento do produto na intenção de compra: fatores moderadores e mediadores na avaliação de produtos criados por IA vs Humanos

Published: Jan. 1, 2024

The advancement of Artificial Intelligence (AIs) has generated a significant impact in various areas, especially Consumer Behavior. With the ability AIs to create innovative products, question arises: how do consumers react hedonic products created by Intelligence? Thus, objective this investigation is explain consumer responses developed and Humans, testing mediation Narrative Transportation Attitude towards product, moderation Algorithm Trust, Need for Cognition, Technological Readiness. For study, experimental method among subjects was employed, where final study involved 177 participants (n=177), divided into control groups. They were exposed two identical with one having information that it an AI other human. focus on relationship between creation (Independent Variable) Purchase Intention (Dependent Variable). Theories such as Automation Bias Speciesism provided conceptual basis study. results revealed product higher when mediated moderated Trust. However, Trust lower, there lower Intention. Cognition Readiness not confirmed. find support Speciesism. this, possible conclude who trusts algorithm tends believe more reliable than Human (Automation Bias). trust reject AI, trusting (Speciesism). This work contributes academic business understanding acceptance products. research limited specific (short film) diverse audience, suggesting need future investigations types audiences. highlights importance algorithms, paving way studies can expand our interaction humans intelligent technologies

Language: Английский

---

Visual detection method of tunnel water leakage diseases based on feature enhancement learning

Tunnelling and Underground Space Technology, Journal Year: 2024, Volume and Issue: 153, P. 106009 - 106009

Published: Aug. 8, 2024

Language: Английский

---

Leveraging AI and Machine Learning for Advancing Marketing Research and Practice

Published: Dec. 10, 2024

Objective: This study examines the transformative potential of artificial intelligence (AI) and machine learning (ML) in marketing research practice, highlighting their role improving predictive accuracy, unlocking insights from complex data, supporting transparent analytics, optimizing customer journey mapping. It also how integration human with AI contributes to advancement theories practices.Methods: A comprehensive methodological framework has been designed assess interplay between AI/ML-driven models key constructs. Advanced statistical analyses were employed ensure robust validation theoretical practical implications. Variables operationalised using well-established instruments reliability construct validity.Results: The identifies trends opportunities, showing AI/ML technologies are reshaping by addressing challenges, enabling new capabilities providing actionable insights. highlights gaps current methodologies, calling for a nuanced understanding applications.Novelty: By bridging advanced techniques theory, this offers fresh perspective on integrating technological innovation human-centred addresses importance ethical frameworks interpretability models, thus paving way responsible AI-driven marketing.Implications Research: findings encourage researchers further explore intersection marketing, exploring underrepresented contexts, refining interpretative ethics. Future should aim combine advances consumer-centred theory-driven approaches.

Language: Английский

---

Feasible Applicability of Deep Learning for Solid Detection in Concrete Wastewater: An Evaluation

Applied Sciences, Journal Year: 2023, Volume and Issue: 13(15), P. 8652 - 8652

Published: July 27, 2023

Concrete wastewater from mixing stations leads to environment contamination due its high alkalinity. The can be reused if solid content is accurately and timely detected. However, investigations into the traditional methods for reuse have demonstrated that they are time consuming not efficient. Therefore, exact acquirement of in concrete becomes a necessity. Recent studies shown deep learning has been successfully applied detect concentration chemical solutions particle suspending liquid. Moreover, also used recognize accurate water level, which facilitates detection solid–liquid separation surface after sedimentation. this article feasibility challenges applying were comprehensively evaluated discussed. Finally, an experimental setup was proposed future research, it indicated transfer learning, data augmentation, hybrid approaches, multi-sensor integration techniques selected facilitate performances.

Language: Английский

---