Recent advances and applications of machine learning in the variable renewable energy sector

Energy Reports, Год журнала: 2024, Номер 12, С. 5044 - 5065

Опубликована: Ноя. 8, 2024

Язык: Английский

---

Global optimization strategy of prosumer data center system operation based on multi-agent deep reinforcement learning

Journal of Building Engineering, Год журнала: 2024, Номер 91, С. 109519 - 109519

Опубликована: Май 13, 2024

Язык: Английский

---

Exponential and Logarithmic Regression Models to Improve Cloud Performance Using Reinforcement Learning

Lecture notes in electrical engineering, Год журнала: 2025, Номер unknown, С. 501 - 509

Опубликована: Янв. 1, 2025

Язык: Английский

---

Reinforcement learning for data center energy efficiency optimization: A systematic literature review and research roadmap

Applied Energy, Год журнала: 2025, Номер 389, С. 125734 - 125734

Опубликована: Март 26, 2025

Язык: Английский

---

A novel 4-level joint optimal dispatch for demand response of data centers with district autonomy realization

Applied Energy, Год журнала: 2024, Номер 358, С. 122590 - 122590

Опубликована: Янв. 8, 2024

Язык: Английский

---

Convergence of AI and MEC for Autonomous IoT Service Provisioning and Assurance in B5G

IEEE Open Journal of the Communications Society, Год журнала: 2023, Номер 4, С. 2913 - 2929

Опубликована: Янв. 1, 2023

With the exponential growth of Internet Things (IoT) devices, IoT has become a transformative technology with applications spanning various domains. It encompasses wide range public and industrial vertical services that come diverse stringent Quality Service (QoS) requirements. Traditional networks often struggle to meet demands these services. As result, introduction 5G Beyond (B5G) holds promise in accommodating through network slicing technology. Network involves partitioning single physical infrastructure into multiple logically isolated ensures dedicated resources each service as per QoS Additionally, Multi-Access Edge Computing (MEC) B5G presents an innovative solution facilitate low-latency communication for However, automatic provisioning management end-to-end (e2e) across multi-domain infrastructures pose significant challenges, including manual error-prone resource configuration, slice template preparation, human intervention. This paper proposes automated Artificial Intelligence (AI) MEC-enabled managing domains specifically tailored Our provides abstraction layer generates templates domain automates deployment based on specified configuration process, reduces intervention, manages complete lifecycle slices. We have conducted several tests our system, creating slices, observed stable performance design, provisioning, isolation, management.

Язык: Английский

---

Machine learning approaches for efficient energy utilization in cloud data centers

Procedia Computer Science, Год журнала: 2024, Номер 235, С. 1782 - 1792

Опубликована: Янв. 1, 2024

Cloud computing technology provides access on demand to virtualized resources, services, and applications via a distributed network. In cloud data centers effective energy utilization is critical concern in today's technology-driven world. (CDC) are massive facilities that host manage an enormous amount of resources. This article addresses the growing significance energy-intensive nature centers. Due rapid growth computing, it offers on-demand resources globally leads substantial power consumption carbon impact environment. They consume amounts energy, optimizing their essential for reducing operational costs, minimizing environmental impact, ensuring sustainable growth. To combat this, efficient energy-saving approaches using machine learning methods have been researched. ML hold great potential enhancing efficiency CDCs by analysing data, detecting patterns, resource usage. The focus areas include CPU usage prediction, overload finding, underload estimation, selection, migration, relocation VMs attain improve utilization. paper compares results achieved different techniques minimize meet service level agreements (SLA). reduce from 1.6% 88.5% compared benchmark approach mentioned, considering various settings parameters.

Язык: Английский

---

Deep-EERA: DRL-Based Energy-Efficient Resource Allocation in UAV-Empowered Beyond 5G Networks

Tsinghua Science & Technology, Год журнала: 2024, Номер 30(1), С. 418 - 432

Опубликована: Сен. 11, 2024

Язык: Английский

---

Taxonomy of optimization algorithms combined with CNN for optimal placement of virtual machines within physical machines in data centers

Energy Informatics, Год журнала: 2024, Номер 7(1)

Опубликована: Окт. 19, 2024

Energy management in datacenters is a major challenge today due to the environmental and economic impact of increasing energy consumption. Efficient placement virtual machines physical within modern crucial for their effective management. In this context, five algorithms named CNN-GA, CNN-greedy, CNN-ABC, CNN-ACO CNN-PSO, have been developed minimize hosts' power consumption ensure service quality with relatively low response times. We propose comparative approach between other existing methods machine placement. The use optimization combined Convolutional Neural Networks build predictive models were evaluated based on accuracy complexity select optimal solution. necessary data collected using CloudSim Plus simulator, prediction results used allocate according predictions models. main objective research optimize Information Technology resources datacenters. This achieved by seeking policy that minimizes ensures an appropriate level users' needs. It considers imperatives sustainability, performance, availability reducing studied six scenarios under specific constraints determine best model machines' aims address current challenges operational efficiency.

Язык: Английский

---

Development and Assessment of Energy-Efficient Approaches for AI-Based Green Computing

Lecture notes in networks and systems, Год журнала: 2024, Номер unknown, С. 179 - 187

Опубликована: Янв. 1, 2024

Язык: Английский

---