Article
Economics
Luke Marshall, Anna Bruce, Iain MacGill
Summary: Understanding competition in electricity markets is crucial for developing effective regulatory, market design, and policy frameworks. The lack of literature on the competitiveness of the Australian National Electricity Market (NEM) may hinder policymakers and researchers in addressing the policy needs of the Australian electricity sector. This study introduces a network-extended residual supply index (NERSI) as a potential solution for more accurate prediction of market power.
Article
Green & Sustainable Science & Technology
Salahuddin Khan
Summary: Short-term load forecasting is crucial for the efficient management of electric systems and the development of reliable energy infrastructure. A novel integrated model combining wavelet transform decomposition, radial basis function network, and thermal exchange optimization algorithm was developed. The performance of this model was evaluated in two deregulated power markets and compared with various standard forecasting models.
Article
Chemistry, Multidisciplinary
Stefan Ungureanu, Vasile Topa, Andrei Cristinel Cziker
Summary: In the current trend of consumption, electricity consumption will become a high cost for end-users. This study proposes a deep learning method for accurately forecasting industrial electric usage, automate the prediction process, and optimize the operation of power systems.
APPLIED SCIENCES-BASEL
(2021)
Article
Computer Science, Information Systems
Ernesto Aguilar Madrid, Nuno Antonio
Summary: This study introduces a set of machine learning models to enhance the accuracy of short-term load forecasting, with the Extreme Gradient Boosting Regressor algorithm showing the best results among them, outperforming historical predictions based on neural networks.
Article
Automation & Control Systems
Vahid Aryai, Mark Goldsworthy
Summary: Accurate forecasting of regional electrical grid carbon emissions is crucial for emissions reduction programs. This study proposes a Particle Swarm Optimised-extremely randomised trees (PSO-ERT) model for day ahead forecasting of emissions intensity in the Australian National Electricity Market (NEM). The PSO-ERT model outperforms LSTM and ELM models, as well as classic machine learning algorithms MLP and DT. Weather forecasts could further enhance the performance of the models.
ENGINEERING APPLICATIONS OF ARTIFICIAL INTELLIGENCE
(2023)
Article
Green & Sustainable Science & Technology
Abhijith Prakash, Anna Bruce, Iain MacGill
Summary: This paper provides an overview and assessment of the frequency control arrangements in Australia's National Electricity Market during energy transition. The paper discusses the challenges in maintaining secure frequency control with the growing penetration of renewable energy. Four key insights are provided on designing frequency control arrangements, including understanding control action interactions, implementing efficient price formation and cost allocation mechanisms, monitoring and assessing service provision, and considering regulatory and market mechanisms and their consequences and interactions. The paper argues for more robust and forward-looking frequency control arrangements during energy transition.
RENEWABLE & SUSTAINABLE ENERGY REVIEWS
(2022)
Article
Thermodynamics
Juyong Lee, Youngsang Cho
Summary: This study compares the performance of time series, machine learning, and hybrid models for peak load forecasting in Korea, showing that the hybrid models outperform the SARIMAX model. Among the machine learning models, those based on LSTM performed the best, with no significant performance difference observed between single and hybrid LSTM models. The LSTM, SARIMAX-SVR, and SARIMAX-LSTM models have shown better performance than the current time series-based forecasting model in Korea, suggesting that incorporating machine learning or hybrid models could improve peak load forecasting in Korea.
Article
Mathematics, Interdisciplinary Applications
Matheus Henrique Dal Molin Ribeiro, Ramon Gomes da Silva, Gabriel Trierweiler Ribeiro, Viviana Cocco Mariani, Leandro dos Santos Coelho
Summary: Efficient models for short-term load forecasting in electricity distribution and generation systems are crucial for companies' energetic planning. In this study, an ensemble learning model based on dual decomposition approach, machine learning models and hyperparameters optimization is proposed. The model successfully decomposes the time series and handles the non-linearities, and achieves accurate load forecasting results with reduced errors.
CHAOS SOLITONS & FRACTALS
(2023)
Article
Engineering, Multidisciplinary
Chixin Xiao, Danny Sutanto, Kashem M. Muttaqi, Minjie Zhang, Ke Meng, Zhao Yang Dong
Summary: This article introduces a novel online learning forecast approach to improve predispatch price forecast using the OS-ELM algorithm. The approach includes a unique data structure and modules that can continuously perceive changes in nonlinear patterns and showed promising results in simulation studies based on Australian electricity market data.
IEEE TRANSACTIONS ON INDUSTRY APPLICATIONS
(2021)
Article
Computer Science, Artificial Intelligence
Ceyhun Yildiz
Summary: This paper proposes a computationally efficient and powerful three-stage model to accurately forecast short-term electricity load. The model utilizes variational mode decomposition (VMD) for feature extraction, stacked kernel extreme learning machine (KELM)-based auto-encoders for unsupervised feature learning, and a KELM-based regression model for load forecasting. The proposed model outperforms state-of-the-art architectures and the TSO model.
NEURAL COMPUTING & APPLICATIONS
(2023)
Article
Chemistry, Multidisciplinary
Pyae-Pyae Phyo, Chawalit Jeenanunta
Summary: Short-term load forecasting is crucial in the electricity industry, and this study proposes a bagging ensemble model combining linear regression and support vector regression. It outperforms deep learning models in terms of accuracy.
APPLIED SCIENCES-BASEL
(2022)
Article
Chemistry, Multidisciplinary
George Stamatellos, Tassos Stamatelos
Summary: Although there have been significant developments in machine learning methods for short-term electrical load forecasting at a Country level, the complexity and diversity of the problem indicate the need for more research effort in selecting representative input datasets for training. The example of the Greek electricity system demonstrates the importance of carefully selecting and ensuring the quality of input data to achieve acceptable levels of prediction accuracy. Short-term load forecasting plays a crucial role in power system planning, operation, and control.
APPLIED SCIENCES-BASEL
(2023)
Article
Chemistry, Multidisciplinary
Manuel Lopez-Martin, Antonio Sanchez-Esguevillas, Luis Hernandez-Callejo, Juan Ignacio Arribas, Belen Carro
Summary: This study compares various traditional machine learning and deep learning techniques, as well as new methods for dynamic model analysis and short-term load forecasting. It explores the impact of critical parameters in time series forecasting, including rolling window length, forecast length, and the number/nature of features used.
APPLIED SCIENCES-BASEL
(2021)
Article
Engineering, Electrical & Electronic
Yang Yang, Zijin Wang, Shangrui Zhao, Jinran Wu
Summary: Accurate power load forecasting is crucial in power systems, and extracting effective features from raw data and having a large amount of training data is essential for high prediction accuracy. To address the issue of data sharing and privacy, the VMD-FK-SecureBoost algorithm combines variational mode decomposition (VMD), federated k-means clustering algorithm (FK), and SecureBoost. This algorithm utilizes VMD to decompose the data, FK to recombine sub-sequences, and SecureBoost for federated learning with privacy protection. The results showed that VMD-FK-SecureBoost outperformed XGBoost and SecureBoost in power load forecasting, with the lowest MAPEs in the Texas and Newcastle CBD areas.
ELECTRIC POWER SYSTEMS RESEARCH
(2023)
Article
Computer Science, Artificial Intelligence
Tiago Pinto, Isabel Praca, Zita Vale, Jose Silva
Summary: This paper presents three ensemble learning models for short term load forecasting, with results showing that the adapted Adaboost model outperforms the reference models for hour-ahead load forecasting.
Article
Energy & Fuels
Yunpeng Jiang, Zhouyang Ren, Xin Yang, Qiuyan Li, Yan Xu
Summary: A new method is proposed in this paper to improve energy flow analysis for integrated natural gas and power systems by considering temperature distribution in natural gas systems, significantly enhancing computational efficiency and convergence performance. Verified with test systems, the method accurately describes and estimates temperature distribution and pressure of natural gas networks, ensuring secure system operation.
Article
Energy & Fuels
Jizhe Liu, Yuchen Zhang, Ke Meng, Zhao Yang Dong, Yan Xu, Siming Han
Summary: This paper proposes a risk-averse deep learning method for real-time emergency load shedding. By training deep neural networks, this method is more inclined to avoid load undercutting events, thereby reducing the significant costs incurred by control failure.
Article
Engineering, Electrical & Electronic
Liudong Chen, Nian Liu, Songnan Yu, Yan Xu
Summary: This paper proposes a stochastic game approach for distributed voltage regulation based on autonomous PV prosumers. It introduces an economic incentive-based voltage regulation model to deal with PV uncertainties and solves the problem using a Markov decision process and probability distribution.
IEEE TRANSACTIONS ON POWER SYSTEMS
(2022)
Article
Engineering, Electrical & Electronic
Weitao Yao, Yu Wang, Yan Xu, Chao Deng, Qiuwei Wu
Summary: This paper studies the issue of communication time-delay in islanded microgrids with distributed secondary control architecture. A new weight-average-prediction (WAP) controller is proposed to compensate for delayed system states. Novel methods for evaluating stability with fixed and time-varying delays are also introduced. Furthermore, nonlinear WAP control methods are discussed to guide parameter tuning.
IEEE TRANSACTIONS ON POWER SYSTEMS
(2022)
Article
Energy & Fuels
Guangzhong Dong, Yan Xu, Zhongbao Wei
Summary: In this paper, a novel two-layer hierarchical approach for online SOC estimation and RUL prediction of lithium-ion batteries is proposed. The approach utilizes a robust observer and GPR to accurately estimate SOC and predict RUL. Experimental results demonstrate the effectiveness of the proposed methods.
IEEE TRANSACTIONS ON ENERGY CONVERSION
(2022)
Article
Engineering, Electrical & Electronic
Heling Yuan, Yan Xu, Cuo Zhang
Summary: This paper proposes a robust optimization method to address the impact of wind power generators on transient stability of a power system. By considering uncertain wind power output, coordinating generation dispatch and emergency load shedding, this method offers an effective solution.
IEEE TRANSACTIONS ON POWER SYSTEMS
(2022)
Editorial Material
Engineering, Electrical & Electronic
Yanli Liu, Lamine Mili, Yan Xu, Junbo Zhao, Innocent Kamwa, Dipti Srinivasan, Ali Mehrizi-Sani, Pablo Arboleya, Vladimir Terzija
Summary: This is the Guest Editorial of the Special Issue on Data-Analytics for Stability Analysis, Control, and Situational Awareness of Power System with High-Penetration of Renewable Energy.
INTERNATIONAL JOURNAL OF ELECTRICAL POWER & ENERGY SYSTEMS
(2022)
Article
Engineering, Electrical & Electronic
Shiwei Xie, Yan Xu, Xiaodong Zheng
Summary: This study focuses on the temporal and spatial coupling between power distribution networks and transportation networks. It proposes a dynamic network equilibrium model to capture the dynamic interactions between the two networks, and accurately describe the choices of drivers, traffic flows, queues, and electricity prices.
IEEE TRANSACTIONS ON SMART GRID
(2022)
Article
Engineering, Electrical & Electronic
Ruipeng Xu, Cuo Zhang, Yan Xu, Zhaoyang Dong, Rui Zhang
Summary: This paper proposes a multi-objective hierarchically-coordinated VVC method to maximize the benefits of inverter-based VVC. By simultaneously optimizing reactive power setpoints for central control and droop control functions for local control, the method aims to minimize average bus voltage deviation and network power loss.
IEEE TRANSACTIONS ON SMART GRID
(2022)
Article
Engineering, Electrical & Electronic
Chao Ren, Xiaoning Du, Yan Xu, Qun Song, Yang Liu, Rui Tan
Summary: This study focuses on the vulnerability of data-driven power system stability assessment models to adversarial examples and proposes an adversarial training-based mitigation strategy to enhance the robustness of the models.
IEEE TRANSACTIONS ON SMART GRID
(2022)
Article
Green & Sustainable Science & Technology
Yang Xia, Yan Xu, Yu Wang, Suman Mondal, Souvik Dasgupta, Amit K. Gupta, Gaurav M. Gupta
Summary: In this paper, a data-driven decentralized economic frequency control method is proposed for isolated networked-microgrid systems. Each agent in the multi-agent deep reinforcement learning framework is trained offline to generate optimal control actions based on local information in online application. Additionally, a safety model is designed to support each agent and can be used for online monitoring and guidance.
IEEE TRANSACTIONS ON SUSTAINABLE ENERGY
(2022)
Article
Engineering, Electrical & Electronic
Nan Zhou, Bei Han, Yan Xu, Lingen Luo, Gehao Sheng, Xiuchen Jiang
Summary: A novel method for fault locating and severity assessment in power distribution systems is proposed based on elasticity network mapping. By comparing the normalized distribution rate values, the fault location and severity can be accurately evaluated.
INTERNATIONAL JOURNAL OF ELECTRICAL POWER & ENERGY SYSTEMS
(2023)
Article
Energy & Fuels
Feilong Fan, Rui Zhang, Yan Xu, Shuyun Ren
Summary: This paper investigates an emission-free hybrid hydrogen-battery energy storage microgrid and proposes a coordinated operational strategy to minimize daily operation costs. Numerical simulations using Australian energy market data validate the effectiveness of the proposed strategy.
CSEE JOURNAL OF POWER AND ENERGY SYSTEMS
(2022)
Article
Engineering, Electrical & Electronic
Yang Xia, Yan Xu, Bin Gou, Qingli Deng
Summary: This article proposes a speed sensor fault diagnosis methodology in induction motor drive systems. The method utilizes a learning-based data-driven principle and involves signal estimation, residual evaluation, and decision-making based on outlier test. The speed estimation is achieved through a nonlinear autoregressive exogenous (NARX) learning model and a randomized learning technique called random vector functional link (RVFL) network. The proposed approach shows promising performance in offline and real-time tests, without requiring motor parameters or additional hardware.
IEEE TRANSACTIONS ON INSTRUMENTATION AND MEASUREMENT
(2022)
Article
Energy & Fuels
Chao Ren, Yan Xu, Junhua Zhao, Rui Zhang, Tong Wan
Summary: This paper presents a fully data-driven method for short-term voltage stability assessment of power systems with incomplete PMU measurements. The method uses a deep learning convolutional neural network to handle missing PMU measurements and employs incremental learning to update the model for better online performance. Unlike existing methods, this approach can fill missing data under various scenarios of PMU placement information loss and network topology changes. Simulation results demonstrate that the proposed method outperforms other methods in terms of accuracy and tolerance to missing data in STVS assessment.
CSEE JOURNAL OF POWER AND ENERGY SYSTEMS
(2022)