Real-time power prediction approach for turbine using deep learning techniques

被引:38
作者
Sun, Lei [1 ]
Liu, Tianyuan [1 ]
Xie, Yonghui [1 ]
Zhang, Di [2 ]
Xia, Xinlei [3 ]
机构
[1] Xi An Jiao Tong Univ, State Key Lab Strength & Vibrat Mech Struct, Xian, Peoples R China
[2] Xi An Jiao Tong Univ, MOE Key Lab Thermofluid Sci & Engn, Xian, Peoples R China
[3] Shanghai Elect Power Generat Equipment Co Ltd, Shanghai, Peoples R China
关键词
Power prediction; Deep learning; Machine learning; Recurrent neural network; Convolutional neural network; Power plant; ORGANIC RANKINE-CYCLE; SOLAR COLLECTOR FIELDS; CONTROL SCHEMES; DESIGN; TEMPERATURE; PERFORMANCE; NETWORK; MODEL;
D O I
10.1016/j.energy.2021.121130
中图分类号
O414.1 [热力学];
学科分类号
摘要
Accurate power forecasting is of great importance to the turbine control and predictive maintenance. However, traditional physics models and statistical models can no longer meet the needs of precision and flexibility when thermal power plants frequently undertake more and more peak and frequency mod-ulation tasks. In this study, the recurrent neural network (RNN) and convolutional neural network (CNN) for power prediction are proposed, and are applied to predict real-time power of turbine based on DCS data (recorded for 719 days) from a power plant. In addition, the performances of two deep learning models and five typical machine learning models are compared, including prediction deviation, variance and time cost. It is found that deep learning models outperform other shallow models and RNN model performs best in balancing the accuracy-efficient trade-off for power prediction (the relative prediction error of 99.76% samples is less than 1% in all load range for test 216 days). Moreover, the influence of training size and input time-steps on the performance of RNN model is also explored. The model can achieve remarkable performance by learning only 30% samples (about 216 days) with 3 input time-steps (about 60 s). Those results of the proposed models based on deep-learning methods indicated that deep learning is of great help to improve the accuracy of turbine power prediction. It is therefore convinced that those models have a high potential for turbine control and predictable maintenance in actual in-dustrial scenarios. (c) 2021 Elsevier Ltd. All rights reserved.
引用
收藏
页数:18
相关论文
共 45 条
[1]   Long short-term memory [J].
Hochreiter, S ;
Schmidhuber, J .
NEURAL COMPUTATION, 1997, 9 (08) :1735-1780
[2]   CFD based design of a 4.3MW Francis turbine for improved performance at design and off-design conditions [J].
Aradag, Selin ;
Akin, Hasan ;
Celebioglu, Kutay .
JOURNAL OF MECHANICAL SCIENCE AND TECHNOLOGY, 2017, 31 (10) :5041-5049
[3]  
Balje O. E., 1981, Journal of Fluids Engineering, V103, P644
[4]   Deep learning framework to forecast electricity demand [J].
Bedi, Jatin ;
Toshniwal, Durga .
APPLIED ENERGY, 2019, 238 :1312-1326
[5]  
Bengio Y., ARXIV PREPRINT ARXIV
[6]   Support Vector Machines for classification and regression [J].
Brereton, Richard G. ;
Lloyd, Gavin R. .
ANALYST, 2010, 135 (02) :230-267
[7]   Operational flexibility and economics of power plants in future low-carbon power systems [J].
Brouwer, Anne Sjoerd ;
van den Broek, Machteld ;
Seebregts, Ad ;
Faaij, Andre .
APPLIED ENERGY, 2015, 156 :107-128
[8]   Day-ahead building-level load forecasts using deep learning vs. traditional time-series techniques [J].
Cai, Mengmeng ;
Pipattanasomporn, Manisa ;
Rahman, Saifur .
APPLIED ENERGY, 2019, 236 :1078-1088
[9]   Thermoeconomic analysis and off-design performance of an organic Rankine cycle powered by medium-temperature heat sources [J].
Calise, Francesco ;
Capuozzo, Claudio ;
Carotenuto, Alberto ;
Vanoli, Laura .
SOLAR ENERGY, 2014, 103 :595-609
[10]   A survey on control schemes for distributed solar collector fields. Part II: Advanced control approaches [J].
Camacho, E. F. ;
Rubio, F. R. ;
Berenguel, M. ;
Valenzuela, L. .
SOLAR ENERGY, 2007, 81 (10) :1252-1272