Yunbin Xu
ABSTRACT
In this paper, in order to develop a gas compressor (GC) efficiency degradation prediction model for the predictive maintenance purpose, a novel artificial neural network (ANN) model which can take both past operating data and future operation schedule as input is developed based on Direct Strategy and DLinear model. The detailed development process of the model, including data preprocessing and feature engineering, as well as the selection of model prediction strategies and the intricate construction of the model, are all presented. The test result on the test dataset shows that it can accurately forecast the GC efficiency trend over the next seven days based on the past 24 hours of operation data and the forthcoming week's operation plan, especially the efficiency degradation trend after blade washing. This model can serve as a robust tool for predictive maintenance of M251S gas turbine's GC.
- TOYOAKI Komori, NOBUYUKI Yamagami, and YUTAKA Shimamura. 2011. Development of Leading Technology for a Low-BTU Gas-firing Gas-turbine Combined-cycle Plant at a Steelworks. MHI Tech. Rev 48, 24–28..Google Scholar
- Si-yuan Song and Rui Yang. 2009. 251-type gas turbine engineering calculations and economic analysis [J]. Mechanical & Electrical Engineering Magazine, 26, 08 (2009), 9–12.Google Scholar
- Yu-han Shi. 2008. Technical and Construction Features for M251S Gas Trubine [J]. THERMAL TURBINE 03, 174–178. DOI:10.13707/j.cnki.31-1922/th.2008.03.007.Google ScholarCross Ref
- Aurora Esteban, Amelia Zafra, and Sebastián Ventura. 2022. Data mining in predictive maintenance systems: A taxonomy and systematic review. WIREs Data Mining and Knowledge Discovery 12, 5 (June 2022). DOI:https://doi.org/10.1002/widm.1471.Google ScholarCross Ref
- Zuming Liu and Iftekhar A. Karimi. 2020. Gas turbine performance prediction via machine learning. Energy 192, 116627. DOI:https://doi.org/10.1016/j.energy.2019.116627.Google ScholarCross Ref
- Ferozkhan Safiyullah, Shaharin Anwar Sulaiman, Misbah Naz, Mohd Shahrizal Jasmani, and Syed Muhammad Afdhal Ghazali. 2018. Prediction on performance degradation and maintenance of centrifugal gas compressors using genetic programming. Energy 158, 485–494. DOI:https://doi.org/10.1016/j.energy.2018.06.051.Google ScholarCross Ref
- Huanhuan Bao, Jianqin Fu, Xilei Sun, Cheng Sun, Jinbo Kuang, Xun Wang and Jingping Liu. 2023. Performance prediction of the centrifugal air compressor for fuel cells considering degradation characteristics based on the hierarchical evolutionary model[J]. Thermal Science and Engineering Progress 46, 102212.Google ScholarCross Ref
- Xiaochuan Li, Fang Duan, Panagiotis Loukopoulos, I.J. Bennett, and Éric David. 2018. Canonical variable analysis and long short-term memory for fault diagnosis and performance estimation of a centrifugal compressor. Control Engineering Practice 72, 177–191. DOI:https://doi.org/10.1016/j.conengprac.2017.12.006.Google ScholarCross Ref
- Joseph B. Kruskall. 1983. The Symmetric Time-Warping Problem: From Continuous to Discrete. In Time Warps, String Edits and Macromolecules: The Theory and Practice of Sequence Comparison. Addison-Wesley Publishing Co., Reading, Massachusetts, 125–161.Google Scholar
- Stan Salvador and Philip K. Chan. 2007. FastDTW: Toward Accurate Dynamic Time Warping in Linear Time and Space. Intelligent Data Analysis 11, 5 (2007), 561–580. Retrieved from http://cs.fit.edu/∼pkc/papers/tdm04.pdf.Google ScholarCross Ref
- Souhaib Ben Taieb, Gianluca Bontempi, Amir F. Atiya, and Antti Sorjamaa. 2012. A review and comparison of strategies for multi-step ahead time series forecasting based on the NN5 forecasting competition. Expert Systems With Applications 39, 8 (June 2012), 7067–7083. DOI:https://doi.org/10.1016/j.eswa.2012.01.039.Google ScholarDigital Library
- Hansika Hewamalage, Christoph Bergmeir, and Kasun Bandara. 2021. Recurrent Neural Networks for Time Series Forecasting: Current status and future directions. International Journal of Forecasting 37, 1 (January 2021), 388–427. DOI:https://doi.org/10.1016/j.ijforecast.2020.06.008.Google ScholarCross Ref
- Yong Yu, Xiaosheng Si, Changhua Hu, and Jianxun Zhang. 2019. A review of Recurrent Neural networks: LSTM cells and network architectures. Neural Computation 31, 7 (July 2019), 1235–1270. DOI:https://doi.org/10.1162/neco_a_01199.Google ScholarDigital Library
- Bibhuti Bhusan Sahoo, Ramakar Jha, Anshuman Singh, and Deepak Kumar. 2019. Long short-term memory (LSTM) recurrent neural network for low-flow hydrological time series forecasting. Acta Geophysica 67, 5 (July 2019), 1471–1481. DOI:https://doi.org/10.1007/s11600-019-00330-1.Google ScholarCross Ref
- Penghui Liu, Jing Liu, and Kai Wu. 2020. CNN-FCM: System modeling promotes stability of deep learning in time series prediction. Knowledge Based Systems 203, 106081. DOI:https://doi.org/10.1016/j.knosys.2020.106081.Google ScholarCross Ref
- Fabien Viton, Mahmoud Elbattah, Jean‐Luc Guérin, and Gilles Dequen. 2020. Heatmaps for Visual Explainability of CNN-Based Predictions for Multivariate Time Series with Application to Healthcare. 2020 IEEE International Conference on Healthcare Informatics (ICHI) (2020). IEEE, 1–8. DOI:https://doi.org/10.1109/ichi48887.2020.9374393.Google ScholarCross Ref
- Ashish Vaswani, Noam Shazeer, Niki Parmar, Jakob Uszkoreit, Llion Jones, Aidan N. Gomez, Łukasz Kaiser, and Illia Polosukhin. 2017. Attention is All you Need. arXiv (Cornell University) 30, (June 2017), 5998–6008. Retrieved from https://arxiv.org/pdf/1706.03762v5.Google Scholar
- Neo Wu, Bradley Green, Xue Ben, and Shawn O'Banion. 2020. Deep transformer models for time series forecasting: The influenza prevalence case. arXiv (Cornell University) (January 2020). Retrieved from https://arxiv.org/pdf/2001.08317.pdf.Google Scholar
- Ailing Zeng, Muxi Chen, Lei Zhang, and Qian Xu. 2023. Are transformers effective for time series forecasting? Proceedings of the ... AAAI Conference on Artificial Intelligence 37, 9 (June 2023), 11121–11128. DOI:https://doi.org/10.1609/aaai.v37i9.26317.Google ScholarDigital Library
Index Terms
- A Novel Gas Compressor Performance Degradation Prediction Model for M251S Gas Turbine System
Recommendations
A hybrid model of an artificial neural network with thermodynamic model for system diagnosis of electrical power plant gas turbine
In this paper, the diagnosis system of power plant gas turbine has been developed to detect the deterioration of engine performance. This system can be analyzed the gas path measurement to predict the deterioration of engine main component by using ...
Multivariable model predictive control for a gas turbine power plant
ACMOS'08: Proceedings of the 10th WSEAS International Conference on Automatic Control, Modelling & SimulationIn this brief, constrained multi variable model predictive control (MPC) strategy is investigated for a GE9001E gas turbine power plant. So the rotor speed and exhaust gas temperature are controlled manipulating the fuel command and compressor inlet ...
Constrained model predictive control implementation for a heavy-duty gas turbine power plant
In this paper, model predictive control (MPC) strategy is implemented to a GE9001E gas turbine power plant. A linear model is developed for the gas turbine using conventional mathematical models and ARX identification procedure. Also a process control ...
Comments