A5006554957- Energy Load and Power Forecasting
- Wind Energy Research and Development
- Machine Fault Diagnosis Techniques
- Advanced Control Systems Optimization
- Fault Detection and Control Systems
- Electric Power System Optimization
- Neural Networks and Applications
- Wind Turbine Control Systems
- Solar Radiation and Photovoltaics
- UAV Applications and Optimization
- Anomaly Detection Techniques and Applications
- Advanced Wireless Communication Technologies
- Refrigeration and Air Conditioning Technologies
- Satellite Communication Systems
- Advanced Multi-Objective Optimization Algorithms
- Time Series Analysis and Forecasting
- Product Development and Customization
- Structural Health Monitoring Techniques
- Wind and Air Flow Studies
- Cloud Computing and Resource Management
- Smart Grid Energy Management
- Process Optimization and Integration
- Millimeter-Wave Propagation and Modeling
- Thermodynamic and Exergetic Analyses of Power and Cooling Systems
- Laser-Matter Interactions and Applications
Beijing Institute of Technology
2011-2025
Nanjing University
2014-2024
Shanghai Electric (China)
2024
Liaoning Normal University
2024
University of Iowa
2006-2022
Xi'an University of Technology
2020-2022
Engineering Arts (United States)
2021-2022
University College London
2021
Shanghai Electric Cable Research Institute
2021
Harbin Engineering University
2020
This paper examines time series models for predicting the power of a wind farm at different scales, i.e., 10-min and hour-long intervals. The are built with data mining algorithms. Five algorithms have been tested on various datasets. Two five performed particularly well. support vector machine regression algorithm provides accurate predictions speed intervals up to 1 h into future, while multilayer perceptron is in over 4 ahead. Wind can be predicted fairly accurately based its historical...
Abstract In this paper, models for short‐ and long‐term prediction of wind farm power are discussed. The built using weather forecasting data generated at different time scales horizons. maximum forecast length the short‐term model is 12 h, 84 h. with five mining algorithms. accuracy analysed. by a neural network outperforms all other both prediction. Two basic methods presented: direct model, whereby directly from data, integrated speed then predicted speed. offers better performance than...
Ground subsidence is a common geological hazard in urban areas that endangers the safety of infrastructure, such as subways. In this study, ground risk assessment method considering both intensity and susceptibility proposed applied to assess Shanghai Metro network. Initially, PS-InSAR used for survey area. Subsequently, ten causal factors are collected, LightGBM machine learning algorithm employed conduct analysis. Then, matrix introduced define by combining susceptibility. Finally, map...
This paper investigates the wind turbine power generation performance monitoring based on supervisory control and data acquisition (SCADA) data. The proposed approach identifies turbines with weakened through assessing curve profiles. Profiles that statistically summarize curvatures shapes of a over consecutive time intervals are constructed by fitting models into SCADA sets least square method. To monitor variations profiles time, multivariate residual approaches introduced applied. Two...
Customers benefit from the ability to select their desired options configure final products. Manufacturing companies, however, struggle with dilemma of product diversity and manufacturing complexity. It is important, therefore, for them capture correlations among provided customers. In this paper, a data mining approach applied manage Rules are extracted historical sales used form sub-assemblies as well configurations. Methods discovering frequently ordered configurations 'if-then' rules...
In this paper, a data-mining approach is applied to optimize combustion efficiency of coal-fired boiler. The process complex, nonlinear, and nonstationary. A virtual testing procedure developed validate the results produced by optimization methods. quantifies improvements in without performing live testing, which expensive time consuming. ideas introduced paper are illustrated with an industrial case study.
<para xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink"> In this paper, a data-mining approach is used to develop model for optimizing the efficiency of an electric-utility boiler subject operating constraints. Selection process variables optimize combustion discussed. The selected are critical control coal-fired in presence Two schemes generating settings and updating evaluated. One scheme based on controllable noncontrollable variables. second one...
The concept of anticipatory control applied to wind turbines is presented. Anticipatory based on the model predictive (MPC) approach. Unlike MPC method, noncontrollable variables (such as speed) are directly considered in dynamic equations presented paper predict response variables, e.g., rotor speed and turbine power output. To determine future states drive with equations, a time series was built for speed. fused over certain prediction horizon. Based these predictions, an optimization...
A Bayesian inference based Markov regime switching model is introduced to predict the intraday solar radiance. The proposed utilizes a process describe evolution of radiance time series. optimal number regimes and regime-specific parameters are determined by inference. provides both point interval prediction on posterior distribution derived from historical data Four forecasting models, persistence model, autoregressive (AR) Gaussian regression (GPR) neural network (NN) considered as...
With the rapid development of wind farm worldwide, monitoring status numerous turbines becomes essential work. Abnormal data in power curve (WPC) are quite important for operations and maintenances because they usually reveal turbine failures or some extreme conditions. This paper proposes a new algorithm WPC abnormal detection cleaning by image thresholding based on minimization dissimilarity-and-uncertainty-based energy (MDUE). The basic idea is to transform scattered into digital problem...
The motivation for this work is the necessity to be able select an appropriate Cloud service provider offering migration of existing applications, based on cost minimization. While providers offer pricing information publicly, and online tools allow calculation various offerings, selection which fits better application requirements left developers. For purpose, proposes a decision support system that incorporates both matching calculation, combining features from approaches in State Art....