The heavy-duty gas turbine is playing an increasingly significant role on power generation due to its lower-emission, higher flexibility and thermo-efficiency. Main subsystems of the gas turbine like compressor, combustor and turbine degrade over the operating time under the harsh environmental conditions, which largely impacts the efficiency and productivity of the system. Therefore, it is critical to develop effective approaches to monitor performance degradation of a heavy-duty gas turbine for system predictive maintenance thus improving the efficiency and productivity of the machine. This paper presents a new physics informed machine learning methodology to predict the degradation of gas turbine by seamlessly integrating thermodynamic heat balancing mechanism, component characteristics, multi-source data and artificial neural network model. The mechanism-based thermodynamic model is established for multiple subsystems considering the balance of flow, mass and energy, and then integrated to a system level for performance simulation of the gas turbine under different conditions. The system model is able to effectively simulate values for those parameters that are not measurable (e.g. GT exhaust flow) or inaccurately measured (e.g. fuel flow). Machine learning based data cleaning approach is employed to preprocess the multivariate raw data of the gas turbine. The difference between design performance data and corrected value obtained from the physics-informed model under ISO conditions is utilized to assess the performance degradation. A Long Short-Term Memory (LSTM) model is established from the fusion of the actual and simulation data to predict the performance degradation of the gas turbine. A comparison study with the classical Nonlinear Autoregressive Network with External Input (NARX) neural network is conducted to demonstrate the advantage of the proposed method. Key Word: Gas Turbine, Thermodynamic Balance, Performance Degradation Predict, Machine Learning, LSTM
Gas Turbine, Thermodynamic Balance, Performance Degradation Predict, Machine Learning, LSTM
I. S. Diakunchak. (1992). Performance Deterioration in Industrial Gas Turbines. Journal of engineering for gas turbines and power, vol. 114 (2), pp. 161-168.
Rowen, W. I. (1983). Simplified mathematical representations of heavy-duty gas turbines, Trans. ASME, J. Eng. Power, vol. 105, pp. 865-869.
DeMello, F.P. (1994). Dynamic models for combined cycle plants in power system studies. IEEE Working Group on Prime mover and Energy supply models for system dynamic performance studies, IEEE Trans. Power Syst., vol. 9(3), pp. 1698-1708.
Camporeale, S.M., Fotunato, B., & Mastrovito, M. (2006). A modular code for real time dynamic simulation of gas turbines in simulink. Trans. ASME, vol. 128, pp. 506516.
Tsoutsanis, E. , & Meskin, N. (2019). Dynamic performance simulation and control of gas turbines used for hybrid gas/wind energy applications. Applied Thermal Engineering. vol. 147, pp.122-142.
Kurosaki, M., Sasamoto, M., & Asaka, K., et al. (2018). An efficient transient simulation method for a volume dynamics model. ASME Turbo Expo: Turbomachinery Technical Conference and Exposition. V006T05A008.
Bahlawan, H., & Morini, M., et al. (2018). Development of Reliable NARX Models of Gas Turbine Cold, Warm, and Hot Start-Up. Journal of Engineering for Gas Turbines and Power. vol. 140, pp. 071202-1-13.
Liu, Z. M., & Karimi, I. A. (2020). Gas turbine performance prediction via machine learning. Energy, vol. 192, pp. 110.
Asgari, H., & Chen, X. Q. (2018). Gas Turbines Modeling, Simulation, and Control Using Artificial Neural Networks.
Kumar, V., Goswami, S., & Smith, D., et al. (2023). Realtime prediction of multiple output states in diesel engines using a deep neural operator framework.
Taniquchi H, Miyamae S. (2000). Power generation analysis for high temperature gas turbine in thermodynamic process. J. Propul. Power. 16: 557-561.
M. Raissi, P. Perdikaris, G. E. Karniadakis. (2019). Physicsinformed neural networks: A deep learning framework for solving forward and inverse problems involving nonlinear partial differential equations. Journal of Computational Physics. vol. 378, pp.686-707.
Goswami, S., Yin, M., Yu, Y., & Karniadakis, G. E. (2022). A physics-informed variational DeepONet for predicting crack path in quasi-brittle materials. Computer Methods in Applied Mechanics and Engineering. vol. 391, pp.114587.
L. Lu, P. Jin, G. Pang, Z. Zhang, G. E. Karniadakis. (2021). Learning nonlinear operators via DeepONet based on the universal approximation theorem of operators, Nature Machine Intelligence. vol. 3 (3), pp. 218–229.
C. Lin, Z. Li, L. Lu, S. Cai, M. Maxey, G. E. Karniadakis. (2021). Operator learning for predicting multiscale bubble growth dynamics, The Journal of Chemical Physics. vol. 154 (10), pp. 104118.
Y.Y. Liu & X.M. Jiang. (2022). Towards Predictive Maintenance of a Heavy-Duty Gas Turbine A New Hybrid Intelligent Methodology for Performance Simulation, Prognostic and Health Management PHM 2022. DOI: 10.36001/phmconf.2022.v14i1.3148
R, Yang & M.Y. Zhong. (2022). Machine Learning-Based Fault Diagnosis for Industrial Engineering Systems. CRC Press.
M. de Castro-Cros, M. Velasco & C. Angulo. (2021). Machine-Learning-Based Condition Assessment of Gas Turbines—A Review. Energies, vol. 14, pp. 8468. https://doi.org/10.3390/en14248468
Z. M. Liu & I. A. Karimi. (2020). Gas Turbine Performance Prediction via Machine Learning. Energy, vol.192, pp.110.
S. Hochreiter & J. Schmidhuber. (1997). Long short-term memory. Neural Comput., vol. 9(8), pp.1735–1780.
P. J. Werbos. (1990). Backpropagation through time: What it does and how to do it. Proc. IEEE, vol. 78(10), pp. 1550– 1560.
This work is licensed under a Creative Commons Attribution 3.0 Unported License.