Article
Energy & Fuels
Eike Cramer, Dirk Witthaut, Alexander Mitsos, Manuel Dahmen
Summary: A probabilistic modeling approach is proposed to predict the intraday electricity price difference based on the hourly pattern of the day-ahead market prices. The study also analyzes the influence of external factors using explainable artificial intelligence (XAI). Among various models, the normalizing flow shows the highest accuracy and narrowest prediction intervals.
Article
Computer Science, Information Systems
Nadeela Bibi, Ismail Shah, Abdelaziz Alsubie, Sajid Ali, Showkat Ahmad Lone
Summary: Efficient modeling and forecasting of electricity prices are crucial in competitive electricity markets. This study examines the performance of an ensemble-based technique for short-term electricity spot price forecasting in the Italian electricity market. The results show that the ensemble-based model outperforms the others, while random forest and ARMA models are highly competitive.
Article
Construction & Building Technology
Aleksey Kychkin, Georgios C. Chasparis
Summary: This paper examines short-term forecasting of electricity-load consumption in residential buildings, introducing three new forecasting models and exploring ensemble models and adaptive model switching strategies. Simulation results on validated data demonstrate the superiority of the SPR forecasting model in reducing forecast errors compared to standard techniques.
ENERGY AND BUILDINGS
(2021)
Article
Energy & Fuels
Emma Viviani, Luca Di Persio, Matthias Ehrhardt
Summary: In this study, a probabilistic method for electricity price forecasting that overcomes traditional methods was investigated. Statistical methods were initially considered for point forecast, and compared in terms of efficiency, accuracy, and reliability. A hybrid model combining Neural Network approaches was developed for probabilistic forecasting, showing high standards of efficiency and precision when tested on German electricity price data.
Article
Construction & Building Technology
Hong Yu, Hongping Zhu, Shun Weng, Wangqing Wen, Aiguo Yan, Xingsheng Yu
Summary: This paper proposes a substructural time series model for locating and quantifying the damage in complex systems. A substructural ARMAX model is established to extract the frequencies and mode shapes of substructures as indicators for damage detection. The inverse problem of substructural damage identification is efficiently solved via sparse regularization, and structural damage can be located and quantified through the nonzero terms in the solution vector.
ADVANCES IN STRUCTURAL ENGINEERING
(2023)
Article
Economics
Kadir Ozen, Dilem Yildirim
Summary: Electricity price forecasting is a challenging task that requires consideration of multiple potential factors to improve accuracy and extract more information. This study introduces Bootstrap Aggregation method and shows substantial forecast improvements in electricity price prediction across multiple markets compared to the popular LASSO estimation method.
Article
Energy & Fuels
Sumeyra Demir, Krystof Mincev, Koen Kok, Nikolaos G. Paterakis
Summary: A model's expected generalization error is inversely proportional to its training set size, and artificially expanding the training set size can increase prediction accuracy. This study proposes using autoencoders, variational autoencoders, and Wasserstein generative adversarial networks for time series augmentation, which significantly improves regression accuracies.
Article
Thermodynamics
Luyao Liu, Feifei Bai, Chenyu Su, Cuiping Ma, Ruifeng Yan, Hailong Li, Qie Sun, Ronald Wennersten
Summary: This paper aims to accurately forecast the occurrence probability of extreme low and high electricity prices and analyze the relative importance of different influencing variables. The study proposes a Multivariate Logistic Regression (MLgR) model based on data from the Australian National Electricity Market (NEM) and compares its performance with two other models. The analysis of relative importance provides valuable insights into electricity price forecast and understanding of extreme price dynamics. The findings have significant implications for the management and establishment of a robust energy market.
Article
Economics
Kin G. Olivares, Cristian Challu, Grzegorz Marcjasz, Rafal Weron, Artur Dubrawski
Summary: We introduce NBEATSx, an enhanced version of the NBEATS model that incorporates exogenous variables to improve forecast accuracy in electricity price prediction. The NBEATSx model outperforms the original NBEATS model and other established statistical and machine learning methods by achieving a nearly 20% increase in forecast accuracy. Additionally, the NBEATSx model provides interpretability through its ability to decompose time series and visualize the impacts of trend, seasonality, and exogenous factors.
INTERNATIONAL JOURNAL OF FORECASTING
(2023)
Article
Green & Sustainable Science & Technology
SeyedAli Ghahari, Cesar Queiroz, Samuel Labi, Sue McNeil
Summary: This paper forecasts corruption levels in countries using artificial neural network modeling and time series analysis, at both global and cluster levels. Results suggest that the NARX technique yields reliable predictions of corruption levels, assisting policymakers and organizations in assessing the expected efficacies of corruption control policies.
Article
Computer Science, Artificial Intelligence
Yueyang Wang, Ziheng Duan, Yida Huang, Haoyan Xu, Jie Feng, Anni Ren
Summary: Multivariate time series forecasting is essential for decision-making, and modeling complex relations is a challenging task. The proposed MTHetGNN, utilizing deep learning, achieves state-of-the-art results in predicting multivariate time series.
PATTERN RECOGNITION LETTERS
(2022)
Article
Economics
Jie Cheng
Summary: In this paper, the co-dependence and portfolio value-at-risk of cryptocurrencies are investigated using the generalized autoregressive score (GAS) model. The study findings reveal strong and dynamic structure dependence among virtual currencies. The GAS model effectively handles volatility and correlation changes, outperforming the classic DCC GARCH model in out-of-sample probabilistic forecasts and backtests, providing new insights into multivariate risk measures.
EMPIRICAL ECONOMICS
(2023)
Article
Environmental Sciences
Yihong Zheng, Wanjuan Zhang, Jingjing Xie, Qiao Liu
Summary: This paper proposes a nonlinear autoregressive model with an exogenous input neural network model based on rough set theory for predicting water consumption. The experimental results show that the proposed model performs better in terms of prediction accuracy.
Article
Energy & Fuels
Carolina Deina, Joao Lucas Ferreira dos Santos, Lucas Henrique Biuk, Mauro Lizot, Attilio Converti, Hugo Valadares Siqueira, Flavio Trojan
Summary: Efficient policies for forecasting electricity demand are crucial for ensuring continuous energy supply. However, the selection of independent variables remains an unresolved issue. This study proposes a model that integrates a multi-criteria approach and artificial neural networks to forecast electricity demand in different countries, considering the specificities of each application. The results show that including variables selected by the multi-criteria approach improves the forecasting performance of artificial neural networks compared to traditional linear models, with Radial Basis Function Networks and Extreme Learning Machines being potential techniques to enhance the forecasting.
Article
Economics
Oliver Grothe, Fabian Kaechele, Fabian Krueger
Summary: Modeling price risks in energy markets is crucial for economic decision making. This study proposes a generic and easy-to-implement method for generating multivariate probabilistic forecasts based on univariate point forecasts of day-ahead electricity prices. The method models dependencies across hours using copula techniques and an optional time series component. An example is demonstrated to construct realistic prediction intervals for pricing individual load profiles.
Article
Engineering, Biomedical
Dario Farina, Ivan Vujaklija, Massimo Sartori, Tamas Kapelner, Francesco Negro, Ning Jiang, Konstantin Bergmeister, Arash Andalib, Jose Principe, Oskar C. Aszmann
NATURE BIOMEDICAL ENGINEERING
(2017)
Article
Engineering, Electrical & Electronic
Jianyi Liu, Jose C. Principe, Arash Andalib
DIGITAL SIGNAL PROCESSING
(2020)
Meeting Abstract
Cardiac & Cardiovascular Systems
Diego Pava, Arash Andalib, Kan Li, Kaustubh Kale
JOURNAL OF CARDIAC FAILURE
(2022)
Meeting Abstract
Cardiac & Cardiovascular Systems
Kan Li, Diego Pava, Arash Andalib, Kaustubh Kale
JOURNAL OF CARDIAC FAILURE
(2022)
Proceedings Paper
Computer Science, Artificial Intelligence
Arash Andalib
2007 IEEE INTERNATIONAL JOINT CONFERENCE ON NEURAL NETWORKS, VOLS 1-6
(2007)
Article
Thermodynamics
Pengcheng Zhao, Jingang Wang, Liming Sun, Yun Li, Haiting Xia, Wei He
Summary: The production of green hydrogen through water electrolysis is crucial for renewable energy utilization and decarbonization. This research explores the optimal electrode configuration and system design of compactly-assembled industrial electrolyzer. The findings provide valuable insights for industrial application of water electrolysis equipment.
ENERGY CONVERSION AND MANAGEMENT
(2024)
Article
Thermodynamics
V. Baiju, P. Abhishek, S. Harikrishnan
Summary: Thermally driven adsorption desalination systems (ADS) have gained attention as an eco-friendly solution for water scarcity. However, they face challenges related to low water productivity and scalability. To overcome these challenges, integrating ADS with other desalination technologies can create a small-scale hybrid system. This study proposes integrating ADS with a Thermo Electric Dehumidification (TED) unit to enhance its performance.
ENERGY CONVERSION AND MANAGEMENT
(2024)
Article
Thermodynamics
C. X. He, Y. H. Liu, X. Y. Huang, S. B. Wan, Q. Chen, J. Sun, T. S. Zhao
Summary: A decentralized centroid multi-path RC network model is constructed to improve the temperature prediction accuracy compared to traditional RC models. By incorporating multiple heat flow paths and decentralizing thermal capacity, a more accurate prediction is achieved.
ENERGY CONVERSION AND MANAGEMENT
(2024)
Article
Thermodynamics
Chaoying Li, Meng Wang, Nana Li, Di Gu, Chao Yan, Dandan Yuan, Hong Jiang, Baohui Wang, Xirui Wang
Summary: There is an urgent need to shift away from heavy dependence on fossil fuels and embrace renewable energy sources, particularly in the energy-intensive oil refining process. This study presents an innovative concept called the Solar Oil Refinery, which applies solar energy in oil refining. A solar multi-energies-driven hybrid chemical oil refining system that utilizes solar pyrolysis and electrolysis has been developed, significantly improving solar utilization efficiency, cracking rate, and hydrogen yield.
ENERGY CONVERSION AND MANAGEMENT
(2024)
Article
Thermodynamics
Chao Ma, Guanghui Wang, Dingbiao Wang, Xu Peng, Yushen Yang, Xinxin Liu, Chongrui Yang, Jiaheng Chen
Summary: This study proposes a bio-inspired fish-tail wind rotor to improve the wind power efficiency of the traditional Savonius rotor. Through transient simulations and orthogonal experiments, the key factors affecting the performance are identified. A response surface model is constructed to optimize the power coefficient, resulting in an improvement of 9.4% and 6.6% compared to the Savonius rotor.
ENERGY CONVERSION AND MANAGEMENT
(2024)
Article
Thermodynamics
Sina Bahmanziari, Abbas-Ali Zamani
Summary: This paper proposes a new framework for improving electrical energy harvesting from piezoelectric smart tiles through a combination of magnetic plucking, mechanical impact, and mechanical vibration force mechanisms. Experimental results demonstrate a significant increase in energy yield and average energy harvesting time compared to other mechanisms.
ENERGY CONVERSION AND MANAGEMENT
(2024)
Article
Thermodynamics
Nanjiang Dong, Tao Zhang, Rui Wang
Summary: This study establishes a multiobjective mixed-variable configuration optimization model for a comprehensive combined cooling, heating, and power energy system, and proposes an efficient generating operator to optimize this model. The experimental results show that the proposed algorithm performs better than other state-of-the-art algorithms.
ENERGY CONVERSION AND MANAGEMENT
(2024)
Article
Thermodynamics
Ahmed E. Mansy, Eman A. El Desouky, Tarek H. Taha, M. A. Abu-Saied, Hamada El-Gendi, Ranya A. Amer, Zhen-Yu Tian
Summary: This study aims to convert office paper waste into bioethanol through a sustainable pathway. The results show that physiochemical and enzymatic hydrolysis of the waste can yield a high glucose concentration. The optimal conditions were determined using the Box-Behnken design, and a blended membrane was used for ethanol purification.
ENERGY CONVERSION AND MANAGEMENT
(2024)
Article
Thermodynamics
Sven Klute, Marcus Budt, Mathias van Beek, Christian Doetsch
Summary: Heat pumps are crucial for decarbonizing heat supply, and steam generating heat pumps have the potential to decarbonize the industrial sector. This paper presents the current state, technical and economic data, and modeling principles of steam generating heat pumps.
ENERGY CONVERSION AND MANAGEMENT
(2024)
Article
Thermodynamics
Le Zhang, To-Hung Tsui, Yen Wah Tong, Pruk Aggarangsi, Ronghou Liu
Summary: This study investigates the effectiveness of a current-carrying-coil-based magnetic field in promoting anaerobic digestion of chicken manure. The results show that the applied magnetic field increases methane yield, decreases carbon dioxide production, and reduces the concentration of ammonia nitrogen. Microbial community analysis reveals the enrichment of certain methanogenic genera and enhanced metabolic pathways. Pilot-scale experiments confirm the technical effectiveness of the magnetic field assistance in enhancing anaerobic digestion of chicken manure.
ENERGY CONVERSION AND MANAGEMENT
(2024)
Article
Thermodynamics
Bo Chen, Ruiqing Ma, Yang Zhou, Rui Ma, Wentao Jiang, Fan Yang
Summary: This paper presents an advanced energy management strategy for fuel cell hybrid electric heavy-duty vehicles, focusing on speed planning and energy allocation. By utilizing predictive co-optimization control, this strategy ensures safe inter-vehicle distance and minimizes energy demand. Simulation results demonstrate the effectiveness of the proposed method in reducing fuel cell degradation cost and overall operation cost.
ENERGY CONVERSION AND MANAGEMENT
(2024)
Article
Thermodynamics
Fabio Fatigati, Roberto Cipollone
Summary: Organic Rankine Cycle-based microcogeneration systems that use solar sources to generate electricity and hot water can help reduce CO2 emissions in residential energy-intensive sectors. The adoption of a recuperative heat exchanger in these systems improves efficiency, reduces thermal power requirements, and saves on electricity costs.
ENERGY CONVERSION AND MANAGEMENT
(2024)
Article
Thermodynamics
Lipeng He, Renwen Liu, Xuejin Liu, Xiaotian Zheng, Limin Zhang, Jieqiong Lin
Summary: This research proposes a piezoelectric-electromagnetic hybrid energy harvester (PEHEH) for low-frequency wave motion and self-sensing wave environment monitoring. The PEHEH shows promising power output and the ability to self-power and self-sense the wave environment.
ENERGY CONVERSION AND MANAGEMENT
(2024)
Article
Thermodynamics
Shangling Chu, Yang Liu, Zipeng Xu, Heng Zhang, Haiping Chen, Dan Gao
Summary: This paper studies a distributed energy system integrated with solar and natural gas, analyzes the impact of different parameters on its energy utilization and emissions reduction, and obtains the optimal solution through an optimization algorithm. The results show that compared to traditional separation production systems, this integrated system achieves higher energy utilization and greater reduction in carbon emissions.
ENERGY CONVERSION AND MANAGEMENT
(2024)
Article
Thermodynamics
Qingpu Li, Yaqi Ding, Guangming Chen, Yongmei Xuan, Neng Gao, Nian Li, Xinyue Hao
Summary: This paper proposes and studies a piston-type thermally-driven pump with a structure similar to a linear compressor, aiming to eliminate the high-quality energy consumption of existing pumps and replace mechanical pumps. The coupling mechanism of working fluid flow and element dimension is analyzed based on force analysis, and experimental data analysis is used to determine the pump operation stroke. Theoretical simulation is conducted to analyze the correlation mechanism of the piston assembly. The research shows that the thermally-driven pump can greatly reduce power consumption and has potential for industrial applications.
ENERGY CONVERSION AND MANAGEMENT
(2024)