Article
Computer Science, Artificial Intelligence
Tullio Mancini, Hector Calvo-Pardo, Jose Olmo
Summary: This paper introduces a novel prediction model based on an ensemble of deep neural networks with extra randomness for improved accuracy and uncertainty computation. The method performs well in terms of mean square prediction error and accuracy of prediction intervals, outperforming state-of-the-art methods in experimental settings.
Article
Engineering, Electrical & Electronic
Honglin Wen, Jie Gu, Jinghuan Ma, Lyuzerui Yuan, Zhijian Jin
Summary: A deep-learning forecasting model based on neural basis expansion analysis is proposed to improve short term load forecasting by narrowing the prediction interval. The model uses a doubly residual stacking strategy to decompose the forecasting task into sub-problems and applies conformal quantile regression for better theoretical coverage guarantee. Experiments show that the proposed model demonstrates improved performance in capturing load characteristics and providing narrow prediction intervals with nearly nominal coverage.
IEEE TRANSACTIONS ON SMART GRID
(2021)
Article
Construction & Building Technology
Yangze Zhou, Xiangning Tian, Chaobo Zhang, Yang Zhao, Tingting Li
Summary: This paper proposes an elastic weight consolidation-based sliding window fine-tuning method for dynamic building energy load prediction. The method achieves high accuracy, low computational costs, and low data storage costs. Experimental results demonstrate significant reduction in mean absolute error and better stability compared to other methods.
ENERGY AND BUILDINGS
(2022)
Article
Engineering, Civil
Afaq Ahmad, Aiman Aljuhni, Usman Arshid, Mohamed Elchalakani, Farid Abed
Summary: This study compares the performance of conventional models, proposed equations, and artificial neural networks in estimating the ultimate response of concrete columns reinforced with glass fiber reinforced polymers (GFRPs). The results show that the predictions from the artificial neural network model are closer to the experimental values and are validated through finite element analysis.
Article
Computer Science, Artificial Intelligence
Ali Hosseini Salari, Hossein Mirzaeinejad, Majid Fooladi Mahani
Summary: Tire normal forces are important for vehicle dynamic control systems, and accurate estimation of them can improve vehicle handling and safety. These forces change based on static parameters, road grade, and dynamic states of the vehicle. The proposed algorithm consists of two parts: an ANN-based estimation algorithm for vehicle mass and CG position, and a deep-learning-based algorithm and integrated hardware-software method for estimating dynamic normal forces.
APPLIED SOFT COMPUTING
(2023)
Article
Computer Science, Interdisciplinary Applications
Elnaz Sharghi, Nardin Jabbarian Paknezhad, Hessam Najafi
Summary: This paper introduces the construction of prediction intervals for Suspended Sediment Load modeling using Emotional Artificial Neural Network (EANN) and classic Neural Network models, comparing their reliability. Results show that EANN has higher reliability with Genetic Algorithm constructed PIs, reducing uncertainty levels. Additionally, the LUBE method outperforms the Bootstrap method in terms of reliability, with EANN showing lower uncertainty levels in Upper Rio Grande River modeling compared to FFNN.
EARTH SCIENCE INFORMATICS
(2021)
Review
Green & Sustainable Science & Technology
Vahid Nourani, Nardin Jabbarian Paknezhad, Hitoshi Tanaka
Summary: Despite the wide applications of artificial neural networks in modeling hydro-climatic processes, quantification of the ANNs' performance is a significant matter. Uncertainty analysis in ANN modeling has attracted noticeable attention, with prediction intervals being prevalent tools.
Article
Engineering, Electrical & Electronic
Yufan Zhang, Honglin Wen, Qiuwei Wu, Qian Ai
Summary: Prediction intervals (PIs) are effective tool for quantifying uncertainty in distribution systems. Traditional central PIs are not suitable for skewed distributions and their offline training is vulnerable to unforeseen changes. We propose an optimal online estimation approach that adapts to different data distributions by adaptively determining probability proportion pairs for quantiles. The approach uses reinforcement learning to integrate adaptive selection and quantile predictions, improving PIs' quality. Case studies show that the proposed method outperforms traditional methods in adapting to data distribution and is more robust against concept drift.
IEEE TRANSACTIONS ON SMART GRID
(2023)
Article
Engineering, Biomedical
Shuangyue Yu, Jianfu Yang, Tzu-Hao Huang, Junxi Zhu, Christopher J. Visco, Farah Hameed, Joel Stein, Xianlian Zhou, Hao Su
Summary: This paper presents a high-accuracy gait phase estimation and prediction algorithm based on a two-stage artificial neural network. The algorithm uses a portable controller with only two IMU sensors to estimate and predict the gait cycle in real time. It can detect and classify gait phases in unrhythmic conditions, and also predict future intra-and inter-stride gait phases.
ANNALS OF BIOMEDICAL ENGINEERING
(2023)
Review
Computer Science, Information Systems
Oscar Cartagena, Sebastian Parra, Diego Munoz-Carpintero, Luis G. Marin, Doris Saez
Summary: Researchers are paying more attention to uncertainty quantification, and prediction intervals are widely used to represent the effect of uncertainty on future process behavior. Methods for prediction interval modeling based on fuzzy systems and neural networks are being developed, with some recommendations provided for method selection. Demonstrating the advantages of these methodologies, a comparative analysis of different methods using data from solar power generation is presented.
Article
Multidisciplinary Sciences
Zhengcai Li, Xinmin Hu, Chun Chen, Chenyang Liu, Yalu Han, Yuanfeng Yu, Lizhi Du
Summary: This paper investigates the optimization algorithms based on machine learning for settlement prediction. By comparing the performance of different algorithms, the study finds that Sparrow Search Algorithm (SSA) significantly improves the optimization effect of the gradient descent model and enhances its stability to a certain degree.
SCIENTIFIC REPORTS
(2022)
Article
Engineering, Mechanical
David Betancourt, Rafi L. Muhanna
Summary: Solving problems in complex engineering systems often involves employing a cascade of computational models, and considering the compatibility and processing of uncertainty is crucial in obtaining reliable predictions. However, the second consideration of uncertainty propagation is often less understood and implemented, highlighting the need for models capable of handling data uncertainty.
PROBABILISTIC ENGINEERING MECHANICS
(2022)
Article
Computer Science, Artificial Intelligence
Theodoros Tsiligkaridis
Summary: This paper introduces the Information Aware Dirichlet networks method to improve uncertainty estimation in neural network predictions by learning a Dirichlet prior distribution. Experimental results demonstrate that this method significantly outperforms existing neural network techniques in accuracy and detecting adversarial examples.
Article
Engineering, Industrial
Tony Tohme, Kevin Vanslette, Kamal Youcef-Toumi
Summary: This study proposes a loss function for estimating the predictive uncertainty of deep neural networks based on the Bayesian Validation Metric (BVM) framework and ensemble learning. Experimental results demonstrate that the proposed method is competitive on in-distribution data and robust to statistical change with superior predictive capability on out-of-distribution data.
RELIABILITY ENGINEERING & SYSTEM SAFETY
(2023)
Article
Thermodynamics
Ziwei Xiao, Cheng Fan, Jiaqi Yuan, Xinhua Xu, Wenjie Gang
Summary: In this study, non-intrusive load monitoring (NILM) methods based on artificial neural network (ANN) and random forest (RF) were proposed and compared for disaggregating cooling loads. Results indicated that both methods could accurately achieve load disaggregation, with the RF-based method outperforming the ANN-based one. The equipment load could be disaggregated with the highest accuracy among the four sub-loads, and the detailed sub-loads obtained can guide building renovation and optimization design of building energy systems.
CASE STUDIES IN THERMAL ENGINEERING
(2021)
Article
Energy & Fuels
Cheng Fan, Fu Xiao, Chengchu Yan, Chengliang Liu, Zhengdao Li, Jiayuan Wang
Article
Energy & Fuels
Borui Cui, Cheng Fan, Jeffrey Munk, Ning Mao, Fu Xiao, Jin Dong, Teja Kuruganti
Article
Energy & Fuels
Cheng Fan, Jiayuan Wang, Wenjie Gang, Shenghan Li
Article
Energy & Fuels
Cheng Fan, Fu Xiao, Mengjie Song, Jiayuan Wang
Article
Energy & Fuels
Cheng Fan, Yongjun Sun, Fu Xiao, Jie Ma, Dasheng Lee, Jiayuan Wang, Yen Chieh Tseng
Article
Energy & Fuels
Zhikun Ding, Rongsheng Liu, Zongjie Li, Cheng Fan
Article
Construction & Building Technology
Chaobo Zhang, Junyang Li, Yang Zhao, Tingting Li, Qi Chen, Xuejun Zhang
ENERGY AND BUILDINGS
(2020)
Article
Thermodynamics
Ao Li, Fu Xiao, Cheng Fan, Maomao Hu
Summary: This study aims to develop a data-driven building energy prediction model through transfer learning with limited training data. The research findings suggest that building usage and industry are crucial factors influencing the effectiveness of transfer learning. Transfer learning can effectively enhance the accuracy of BPNN-based building energy models for information-poor buildings.
BUILDING SIMULATION
(2021)
Editorial Material
Thermodynamics
Cheng Fan, Fu Xiao, Da Yan
BUILDING SIMULATION
(2021)
Article
Construction & Building Technology
Guannan Li, Xiaowei Zhao, Cheng Fan, Xi Fang, Fan Li, Yubei Wu
Summary: Comparative evaluations were conducted on various improved LSTM models, showing that LSTM models exhibited improved prediction accuracy of 6.2%-29.2% after parameter optimization. When using one-year data for modeling, CNN-LSTM reduced the average RMSE of LSTM by up to 2.9%. LSTM-ATT demonstrated more stable prediction performance than other models when using two-year data for modeling and decreased the average RMSE of LSTM by up to 5.6%.
JOURNAL OF BUILDING ENGINEERING
(2021)
Review
Thermodynamics
Cheng Fan, Da Yan, Fu Xiao, Ao Li, Jingjing An, Xuyuan Kang
Summary: Buildings play a significant role in global sustainability, and data-driven research methods have greatly enriched knowledge in building energy modeling and improved building performance. With the ongoing development of smart buildings and IoT-driven smart cities, big data-driven research paradigm is becoming an essential complement to existing scientific research methods in the building sector.
BUILDING SIMULATION
(2021)
Proceedings Paper
Energy & Fuels
Song Mengjie, Fan Cheng, Mao Ning, Wang Zhihua, Xia Yudong
INNOVATIVE SOLUTIONS FOR ENERGY TRANSITIONS
(2019)
Proceedings Paper
Energy & Fuels
Cheng Fan, Mengjie Song, Fu Xiao, Xue Xue
INNOVATIVE SOLUTIONS FOR ENERGY TRANSITIONS
(2019)
Article
Thermodynamics
Hai Zhao, Puzhen Gao, Xiaochang Li, Ruifeng Tian, Hongyang Wei, Sichao Tan
Summary: This study numerically investigates the interaction between flow-induced vibration and forced convection heat transfer in a tube bundle. The results show that the impact of flow-induced vibration on heat transfer varies in different flow velocity regions.
APPLIED THERMAL ENGINEERING
(2024)
Article
Thermodynamics
Rohit Chintala, Jon Winkler, Sugirdhalakshmi Ramaraj, Xin Jin
Summary: The current state of fault detection and diagnosis for residential air-conditioning systems is expensive and not suitable for widespread implementation. This paper proposes a cost-effective solution by introducing an automated fault detection algorithm as a screening step before more expensive tests can be conducted. The algorithm uses home thermostats and local weather information to identify thermodynamic parameters and detect high-impact air-conditioning faults.
APPLIED THERMAL ENGINEERING
(2024)
Article
Thermodynamics
A. Azimi, N. Basiri, M. Eslami
Summary: This paper presents a novel optimization algorithm for improving the water-film cooling system of photovoltaic panels, resulting in a significant increase in net energy generation.
APPLIED THERMAL ENGINEERING
(2024)
Article
Thermodynamics
Duc-Thuan Phung, Chin-Hsiang Cheng
Summary: In this study, a novel CFDMD model is used to analyze and investigate the behavior of thermal-lag engines (TLE). The study shows that the CFDMD model effectively captures the thermodynamic behavior of the working gas and the dynamic behavior of the engine mechanism. Additionally, the study explores the temporal evolution of engine speed and the influence of various parameters on shaft power and brake thermal efficiency. The research also reveals the existence of a thermal-lag phenomenon in TLE.
APPLIED THERMAL ENGINEERING
(2024)
Article
Thermodynamics
Haiying Yang, Yinjie Shen, Lin Li, Yichen Pan, Ping Yang
Summary: The purpose of this article is to find a measure to improve the interfacial thermal transfer of graphene/silicon heterojunction. Through molecular dynamics simulation, it is found that surface modification can significantly reduce the thermal resistance, thereby improving the thermal conductivity of the graphene/silicon interface.
APPLIED THERMAL ENGINEERING
(2024)
Article
Thermodynamics
Qiong Wu, Yancheng Wang, Haonan Zhou, Xingye Qiu, Deqing Mei
Summary: This article introduces a visible methanol steam reforming microreactor, which uses an optical crystal as an observation window and measures the reaction temperature in real-time using infrared thermography. The results show that under lower oxygen to carbon ratio conditions, the microreactor has a higher heating rate and a stable gradient in temperature distribution.
APPLIED THERMAL ENGINEERING
(2024)
Review
Thermodynamics
Giulia Manco, Umberto Tesio, Elisa Guelpa, Vittorio Verda
Summary: In the past decade, there has been a growing interest in studying energy systems for the combined management of power vectors. Most of the published works focus on finding the optimal design and operations of Multi Energy Systems (MES). However, for newcomers to this field, understanding how to achieve the desired optimization details while controlling computational expenses can be challenging and time-consuming. This paper presents a novel approach to analyzing the existing literature on MES, with the aim of guiding practical development of MES optimization. Through the discussion of six case studies, the authors provide a mathematical formulation as a reference for building the model and emphasize the impact of different aspects on the problem nature and solver selection. In addition, the paper also discusses the different approaches used in the literature for incorporating thermal networks and storage in the optimization of multi-energy systems.
APPLIED THERMAL ENGINEERING
(2024)
Article
Thermodynamics
Xuepeng Yuan, Caiman Yan, Yunxian Huang, Yong Tang, Shiwei Zhang, Gong Chen
Summary: In this study, a multi-scale microgroove wick (MSMGW) was developed by laser irradiation, which demonstrated superior capillary performance. The surface morphology and performance of the wick were affected by laser scan pitch, laser power, repetition frequency, and scanning speed. The MSMGW showed optimal capillary performance in alumina material and DI water as the working fluid.
APPLIED THERMAL ENGINEERING
(2024)
Article
Thermodynamics
Maofei Mei, Feng Hu, Chong Han
Summary: This paper proposes an effective local search method based on detection of droplet boundaries for understanding the dynamic process of droplet growth during dropwise condensation. The method is validated by comparing with experimental data. The present simulation provides an effective approach to more accurately predict the nucleation site density in future studies.
APPLIED THERMAL ENGINEERING
(2024)
Article
Thermodynamics
Rahul Kumar Sharma, Ashish Kumar, Dibakar Rakshit
Summary: The study explores the use of phase change materials (PCM) as a retrofit with Heating Ventilation and Air-conditioning systems (HVAC) to reduce energy consumption and improve air quality. By incorporating PCM with specific thickness and fin configurations, significant energy savings can be achieved in comparison to standard HVAC systems utilizing R134a. This research provides policymakers with energy-efficient and sustainable solutions for HVAC systems to combat climate change.
APPLIED THERMAL ENGINEERING
(2024)
Article
Thermodynamics
Zhenhua Ren, Xiangjin Meng, Xingang Qi, Hui Jin, Yunan Chen, Bin Chen, Liejin Guo
Summary: This paper investigates the heat transfer mechanism and factors influencing thermal radiation in the process of supercritical water gasification (SCWG) of coal, and proposes a comprehensive numerical model to simulate the process. Experimental validation results show that thermal radiation accounts for a significant proportion of the total heat exchange in the reactor and a large amount of radiant energy exists in the important spectral range of supercritical water. Enhancing radiative heat transfer can effectively increase the temperature of the reaction medium and the gasification rate.
APPLIED THERMAL ENGINEERING
(2024)
Article
Thermodynamics
Mauro Abela, Mauro Mameli, Sauro Filippeschi, Brent S. Taft
Summary: Pulsating Heat Pipes (PHP) are passive two-phase heat transfer devices with a simple structure and high heat transfer capabilities. The actual unpredictability of their dynamic behavior during startup and thermal crisis hinders their large-scale application. An experimental apparatus is designed to investigate these phenomena systematically. The results show that increasing the number of evaporator sections and condenser temperature improves the performance of PHP. The condenser temperature also affects the initial liquid phase distribution and startup time.
APPLIED THERMAL ENGINEERING
(2024)
Article
Thermodynamics
Ke Gan, Ruilian Li, Yi Zheng, Hui Xu, Ying Gao, Jiajie Qian, Ziming Wei, Bin Kong, Hong Zhang
Summary: A 3-dimensional enhanced heat pipe radiator has been developed to improve heat dissipation and temperature uniformity in cooling high-power electronic components. Experimental results show that the radiator has superior heat transfer performance compared to a conventional aluminum fin radiator under different heating powers and wind speed conditions.
APPLIED THERMAL ENGINEERING
(2024)
Article
Thermodynamics
Xinyi Zhang, Shuzhong Wang, Daihui Jiang, Zhiqiang Wu
Summary: This study focuses on recovering waste heat from blast furnace slag using dry centrifugal pelletizing technology. A comprehensive two-dimensional model was developed to analyze heat transfer dynamics and investigate factors influencing heat exchange efficiency. The findings have important implications for optimizing waste heat recovery and ensuring safe operations.
APPLIED THERMAL ENGINEERING
(2024)
Article
Thermodynamics
Xincheng Wu, An Zou, Qiang Zhang, Zhaoguang Wang
Summary: The boosting heat generation rate of high-performance processors is challenging traditional cooling techniques. This study proposes a combined design of active jet intermittency and passive surface modification to enhance heat transfer.
APPLIED THERMAL ENGINEERING
(2024)