COMBUSTION EFFICIENCY PREDICTION IN GAS TURBINE COMBUSTOR USING SIGMOID NORMALIZATION AND FOOTBALL OPTIMIZATION BASED LSTM
DOI:
https://doi.org/10.55766/sujst6024Keywords:
Can Combustor, Football Game Optimization and Combustion Efficiency, Gas Turbine Combustor, LSTM; Sigmoid NormalizationAbstract
An essential part of gas turbines is the combustion system, which burns the fuel and air mixture to produce thrust or power. Anomalies in combustors, such as problems with fuel nozzles, excessive vibration from acoustic waves and oscillations in the release of heat, and non-compliant emissions, are frequently caused by instability in combustion and uneven fuel distribution. To overcome these issues, FBO-LSTM is developed to predict combustion efficiency and power influence. A can combustor was designed with a 150 mm diameter casting, 210 mm length, and a 170 mm long combustor model, featuring variations in casting angle and dump gap. With a flow rate of 0.473 kg/s and 0.0096 kg/s, it has a vertical fluid and air input. Six scenarios were explored, varying casting angles and dump gap to analyse temperature, static pressure, velocity, power, and combustion efficiency. To normalize the input data, these simulated data are gathered. The data are then pre-processed using Sigmoid Normalization. Next, the LSTM is used to estimate the combustion power and efficiency using the pre-processed data. The Football Optimization (FBO) approach is utilized to select the optimal learning rate and batch size of the LSTM classifier. This technique achieved 96.2% of accuracy, 93.5% of selectivity, 89.7% of MCC and 3.8% of error, demonstrating considerable improvements in the findings. The error difference between the turbulence intensity of the dump gap and casting angle is 2% and 0.4%. Thus, this FBO-LSTM approach is a better choice to predict the power and efficiency of the combustor.
References
Bai, M., Yang, X., Liu, J., Liu, J., and Yu, D. (2021). Convolutional neural network-based deep transfer learning for fault detection of gas turbine combustion chambers. Applied Energy, 302:117509. https://doi.org/10.1016/j.apenergy.2021.117509
Chang, S.H., Bak, H.S., Yu, H., and Yoo, C.S. (2020). A numerical study of combustion and NO X emission characteristics of a lean premixed model gas turbine combustor. Journal of Mechanical Science and Technology, 34:1795-1803. https://doi.org/10.1007/s12206-020-0341-y
Choi, O., Choi, J., Kim, N., and Lee, M.C. (2020). Combustion instability monitoring through deep-learning-based classification of sequential high-speed flame images. Electronics, 9(5):848. https://doi.org/10.3390/electronics9050848
Gangopadhyay, T., Ramanan, V., Akintayo, A., Boor, P.K., Sarkar, S., Chakravarthy, S.R., and Sarkar, S. (2021). 3D convolutional selective autoencoder for instability detection in combustion systems. Energy and AI, 4:100067. https://doi.org/10.1016/j.egyai.2021.100067
Joo, S., Kwak, S., Lee, J., and Yoon, Y. (2021). Thermoacoustic instability and flame transfer function in a lean direct injection model gas turbine combustor. Aerospace Science and Technology, 116:106872. https://doi.org/10.1016/j.ast.2021.106872
Le, X.H., Ho, H.V., Lee, G., and Jung, S. (2019). Application of long short-term memory (LSTM) neural network for flood forecasting. Water, 11(7):1387. https://doi.org/10.3390/w11071387
Lyu, Z., Jia, X., Yang, Y., Hu, K., Zhang, F., and Wang, G. (2021). A comprehensive investigation of LSTM-CNN deep learning model for fast detection of combustion instability. Fuel, 303:121300. https://doi.org/10.1016/j.fuel.2021.121300
Meziane, S. and Bentebbiche, A. (2019). Numerical study of blended fuel natural gas-hydrogen combustion in rich/quench/lean combustor of a micro gas turbine. International journal of hydrogen energy, 44(29):15610-15621. https://doi.org/10.1016/j.ijhydene.2019.04.128
Nassini, P.C., Pampaloni, D., Meloni, R., and Andreini, A. (2021). Lean blow-out prediction in an industrial gas turbine combustor through a LES-based CFD analysis. Combustion and Flame, 229:111391. https://doi.org/10.1016/j.combustflame.2021.02.037
Nguyen, T.H., Park, J., Jung, S., and Kim, S. (2019). A numerical study on NO x formation behavior in a lean-premixed gas turbine combustor using CFD-CRN method. Journal of Mechanical Science and Technology, 33:5051-5060. https://doi.org/10.1007/s12206-019-0944-3
Niszczota, P. and Gieras, M. (2021). Impact of the application of fuel and water emulsion on CO and NOx emission and fuel consumption in a miniature gas turbine. Energies, 14(8):2224. https://doi.org/10.3390/en14082224
Niu, Y., Kang, J., Li, F., Ge, W., and Zhou, G. (2020). Case-based reasoning based on grey-relational theory for the optimization of boiler combustion systems. ISA Transactions, 103, 166-176. https://doi.org/10.1016/j.isatra.2020.03.024
Okafor, E.C., Somarathne, K.K.A., Ratthanan, R., Hayakawa, A., Kudo, T., Kurata, O., and Kobayashi, H. (2020). Control of NOx and other emissions in micro gas turbine combustors fuelled with mixtures of methane and ammonia. Combustion and Flame, 211:406-416. https://doi.org/10.1016/j.combustflame.2019.10.012
Park, Y., Choi, M., Kim, K., Li, X., Jung, C., Na, S., and Choi, G. (2020). Prediction of operating characteristics for industrial gas turbine combustor using an optimized artificial neural network. Energy, 213:118769. https://doi.org/10.1016/j.energy.2020.118769
Shao, C., Liu, Y., Zhang, Z., Lei, F., and Fu, J. (2023). Fast prediction method of combustion chamber parameters based on artificial neural network. Electronics, 12(23):4774. https://doi.org/10.3390/electronics12234774
Shi, Y., Zhong, W., Chen, X., Yu, A. B., and Li, J. (2019). Combustion optimization of ultra supercritical boiler based on artificial intelligence. Energy, 170:804-817. https://doi.org/10.1016/j.energy.2018.12.172
Singh, D. and Singh, B. (2020). Investigating the impact of data normalization on classification performance. Applied Soft Computing, 97(Part B):105524. https://doi.org/10.1016/j.asoc.2019.105524
Srilakshmi, K., Rao, G.S., Balachandran, P.K., and Senjyu, T. (2024). Green energy-sourced AI-controlled multilevel UPQC parameter selection using football game optimization. Frontiers in Energy Research, 12:1325865. https://doi.org/10.3389/fenrg.2024.1325865
Sundararaj, R.H., Kumar, R.D., Raut, A.K., Sekar, T.C., Pandey, V., Kushari, A., and Puri, S.K. (2019). Combustion and emission characteristics from biojet fuel blends in a gas turbine combustor. Energy, 182:689-705. https://doi.org/10.1016/j.energy.2019.06.060
Ursueguía, D., Marín, P., Díaz, E., and Ordonez, S. (2021). A new strategy for upgrading ventilation air methane emissions combining adsorption and combustion in a lean-gas turbine. Journal of Natural Gas Science and Engineering, 88:103808. https://doi.org/10.1016/j.jngse.2021.103808
Zhou, Y., Zhang, C., Han, X., and Lin, Y. (2021). Monitoring combustion instabilities of stratified swirl flames by feature extractions of time-averaged flame images using deep learning method. Aerospace Science and Technology, 109:106443. https://doi.org/10.1016/j.ast.2020.106443
