WoS İndeksli Yayınlar Koleksiyonu
Permanent URI for this collectionhttps://hdl.handle.net/20.500.12573/394
Browse
6 results
Search Results
Article Supervised Learning-Driven Dead Band Control of Occupant Thermostats for Energy-Efficient Residential HVAC(Elsevier, 2026-03) Savasci, Alper; Ceylan, Oguzhan; Paudyal, SumitHeating, ventilation, and air conditioning (HVAC) systems play a crucial role in demand-side management (DSM) by shaping residential electricity consumption and enabling flexible, grid-responsive operation. Thermostats in HVAC systems regulate indoor temperature as part of a closed-loop control framework, typically incorporating a fixed temperature dead band-a range around the setpoint where no action is taken-to reduce energy use and prevent frequent cycling of the HVAC system. Although essential for efficiency and equipment longevity, fixed dead bands limit adaptability, as dynamically adjusting them under varying environmental conditions remains challenging for occupants. To address this limitation, we propose a machine learning (ML)-based dead band tuning framework that optimally adjusts thermostat settings in real time. The method integrates conventional optimization with data-driven modeling: a mixed-integer linear programming (MILP) model is first used to gen erate optimal dead band values under measured outdoor temperature records (diverse seasonal weather scenarios) which are then employed to train the ML-based predictor to learn a real-time discrete dead band decision policy that approximates the MILP-optimal hysteresis-aware decisions. Among the evaluated models, Random Forest demonstrates superior predictive performance, achieving a mean squared error (MSE) of 0.0399 and a coefficient of determination (R2) of 95.75 %.Article Citation - WoS: 4Citation - Scopus: 6Traffic Aware Cyclic Sleep-Based Power Consumption Model for a Passive Optical Network(Wiley, 2022-02-28) Butt, Rizwan Aslam; Faheem, Muhammad; Anwar, Muhammad; Mohammadani, Khalid H.; Idrus, Sevia M.For a network, a power consumption model is an important tool to test the performance of a network process for different traffic loads. In a Passive optical network (PON), the optical network unit (ONU) is responsible for the major power consumption of PON. Both IEEE and ITU have standardized a cyclic sleep process (CSP) for ONU energy conservation. In next-generation PON; TWDM and XGS PON, the ONU power contribution has increased further due to higher number of ONUs and ONU being tunable. Therefore, an accurate power consumption model of the CSP process for energy efficiency studies under different traffic conditions is of prime importance. The existing CSP power consumption models do not depict the CSP process accurately especially the inactivity of the ONU in the asleep and sleep aware states are not taken into account which reduce the accuracy of the model. The proposed inactivity aware model (IAM) overcomes these gaps and very accurately models the CSP process, as evident from the results, which are better than earlier model results and quite close to earlier published simulation results. The model is also validated through a simulation-based study and the simulation results are observed to be very close to the model results with only a 5% deviation.Article Citation - WoS: 12Citation - Scopus: 12The Role of Energy Efficiency, Renewable Resources, Green Innovation, and Fiscal Decentralization in Sustainable Development: Evidence From OECD Countries(Elsevier Sci Ltd, 2025-08) Binsaeed, Rima H.; Khan, Zeeshan; Dogan, Eyup; Rahim, SyedEnergy efficiency and renewable resources for sustainable development are novel discussion areas for academics and researchers. Similarly, most developed and emerging countries are experiencing fiscal decentralization to enhance regional development. However, the importance of these sectors in sustainable development is still unclear in the literature. This research investigates the influence of energy efficiency, renewable energy, green innovation, and fiscal decentralization on sustainable development. Using the data for 18 fiscally decentralized OECD countries from 1995 to 2020, the roles of linear and nonlinear green innovation and renewable energy are also considered. This study uses novel moment quantile regression and finds that revenue decentralization, expenditure decentralization, and fiscal decentralization are significant drivers of sustainable development. Additionally, energy efficiency and value-added manufacturing significantly enhance sustainability in the region. However, green innovation and renewables are resources that exhibit a U-shaped association with sustainable development. The robustness of these results is validated via a series of parametric and nonparametric approaches. From the policy perspective, this research suggests improved research and development on renewable energy, green innovation, and energy efficiency could significantly encourage the OECD's journey towards sustainable development. Additionally, subnational governments should be given more fiscal autonomy, which may encourage regional level investments and boost the confidence of clean energy producing sectors to accelerate sustainable regional development.Article Citation - WoS: 6Citation - Scopus: 6Sleep-Aware Wavelength and Bandwidth Assignment Scheme for TWDM PON(Springer, 2021-06) Butt, Rizwan Aslam; Faheem, Muhammad; Ashraf, M. Waqar; Arfeen, Asad; Memon, Kamran Ali; Khawaja, AttaullahThe energy efficiency and delay performance of PON are two inversely related phenomena. Higher sleep time of the Optical Network Units (ONUs) results in higher upstream (US) delays due to increased traffic queues during the ONU Asleep state. Although an efficient dynamic bandwidth and wavelength assignment (DWBA) scheme can decrease US delays by minimizing the bandwidth waste and improving the fairness of bandwidth distribution among the ONUs. However, the conventional DWBA schemes are not designed to work with cyclic sleep mode (CSM) and they keep on assigning bandwidth to ONUs even if the ONU is in Asleep state leading to wastage of bandwidth and degraded CSM performance. Therefore, in this work a sleep aware DWBA scheme for TWDM PON is presented to coordinate with CSM mode. It only assign bandwidth to Active ONUs during the guaranteed phase, surplus phase and excess phase allocation phases which minimizes the bandwidth waste and the bandwidth lost at the ONU end. The wavelength switching process is also improved by only considering the Active state ONUs to balance the traffic load on all the wavelengths. The simulation results support our claim as the SA-DWBA scheme on average achieves DWBA schemes due to up to 50% to 65% higher energy savings compared to other due to longer ONU Asleep times. However, the increased upstream delays of all the traffic classes in SA-DWBA scheme remain within the set delay limit of 50 ms.Article Citation - WoS: 46Citation - Scopus: 54Analyzing the Relationship Between Energy Efficiency and Environmental and Financial Variables: A Way Towards Sustainable Development(Pergamon-Elsevier Science Ltd, 2022-08) Taskin, Dilvin; Dogan, Eyup; Madaleno, MaraThe literature has mainly relied on an annual and short span of data to analyze the relationship between energy, environmental and financial indicators. This study analyzes the relationship between energy efficiency, energy research, pollution mitigation, and FinTech by applying two novel methods-the causality test in the frequency domain [11] and the causality test in the time domain (Shi et al., 2018; 2020) on the daily data from June 17, 2016 to November 16, 2021. Empirical results from the frequency domain test report that pollution mitigation temporarily causes energy efficiency only in the short run while energy efficiency Granger causes it in the short, medium, and long run. Furthermore, energy efficiency can predict FinTech in the short, medium, and long-run; on the other way, FinTech Granger causes energy efficiency in the long and medium run, suggesting a permanent causality relationship. Empirical results from the time-varying test show a bidirectional relationship between energy efficiency, and environmental and financial variables, especially with very high significant episodes around the recent pandemic collapse. Policymakers should promote the launch of financial technologies that will provide finance through green bonds for energy efficiency improvements as well as energy efficiency improvements for pollution mitigation. Further policy implications are discussed in the study.(c) 2022 Elsevier Ltd. All rights reserved.Article Citation - WoS: 23Citation - Scopus: 35A Review of On-Device Machine Learning for IoT: An Energy Perspective(Elsevier, 2024-02) Tekin, Nazli; Aris, Ahmet; Acar, Abbas; Uluagac, Selcuk; Gungor, Vehbi CagriRecently, there has been a substantial interest in on-device Machine Learning (ML) models to provide intelligence for the Internet of Things (IoT) applications such as image classification, human activity recognition, and anomaly detection. Traditionally, ML models are deployed in the cloud or centralized servers to take advantage of their abundant computational resources. However, sharing data with the cloud and third parties degrades privacy and may cause propagation delay in the network due to a large amount of transmitted data impacting the performance of real-time applications. To this end, deploying ML models on-device (i.e., on IoT devices), in which data does not need to be transmitted, becomes imperative. However, deploying and running ML models on already resource-constrained IoT devices is challenging and requires intense energy consumption. Numerous works have been proposed in the literature to address this issue. Although there are considerable works that discuss energy-aware ML approaches for on-device implementation, there remains a gap in the literature on a comprehensive review of this subject. In this paper, we provide a review of existing studies focusing on-device ML models for IoT applications in terms of energy consumption. One of the key contributions of this study is to introduce a taxonomy to define approaches for employing energy-aware on-device ML models on IoT devices in the literature. Based on our review in this paper, our key findings are provided and the open issues that can be investigated further by other researchers are discussed. We believe that this study will be a reference for practitioners and researchers who want to employ energy-aware on-device ML models for IoT applications.