Integrated Genetic Algorithm and Back-Propagation Neural Network Approach for Fault Detection and Prediction in the Tennessee Eastman Process

Qi Ji; Azura Che Soh; Ribhan Zafira Abdul Rahman; Raja Kamil Raja Ahmad; Siti Anom Ahmad.

The current era emphasizes the application of intelligent algorithms for automating industrial processes. Among these, fault detection and prediction take precedence. This research introduces fault detection method that combine the genetic algorithm with back-propagation neural networks (GA-BPNN). The integration of these two methods, GA-BPNN, enhances their effectiveness. GA-BPNN effectively addresses the challenges of poor convergence in traditional genetic algorithms (GAs) and the difficulty of accurately defining parameters in back-propagation neural networks (BPNNs). In this approach, BPNN serves as the foundational framework, while GA dynamically optimizes various parameters within the BPNN. The proposed method GA-BPNN exhibits excellent parameter self-regulation ability and can adapt to various training conditions. This optimization process enhances precision and speed, making GA-BPNN a powerful and efficient solution. The Tennessee Eastman (TE) chemical process is employed as the simulated domain to validate the efficacy and superiority of the GA-BPNN approach in process control. The simulation results indicate that the GA-BPNN method outperforms the traditional BPNN. Additionally, the proposed method demonstrates excellent self-regulation ability, automatically optimizing parameters, and ensuring outstanding adaptability and learning ability in various situations.

KEYWORDS: Back-propagation neural network; Fault detection; Genetic algorithm; Optimization; Prediction



Download this PDF file

REFERENCES

1. Downs, J.J. & Vogel, E. F. 1993. A plant-wide industrial process control problem. Computers and Chemical Engineering, 17(3), 245–255.
2. Adeli, M. & Mazinan, A. 2020. High efficiency fault-detection and fault-tolerant control approach in Tennessee Eastman process via fuzzy-based neural network representation. Complex & Intelligent Systems, 6 (1), 199–212.
3. Heo, S. & Lee, J. H. 2019. Statistical process monitoring of the Tennessee Eastman process using parallel autoassociative neural networks and a large dataset. Processes, 7(411), 1-21.
4. Lomov, I., Lyubimov, M., Makarov, I. & Zhukov, L.E. 2021. Fault detection in Tennessee Eastman process with temporal deep learning models. Journal of Industrial Information Integration, 23(1), 100216.
5. Xie, D. & Bai, L. 2015. A Hierarchical Deep Neural Network for Fault Diagnosis on Tennessee-Eastman Process. Proceedings of the 14th International Conference on Machine Learning and Applications (ICMLA). 9-11 December, 2015, Miami, Florida, USA. pp 745–748.
6. Garud, K. S., Jayaraj, S. & Lee, M. 2021. A review on modeling of solar photovoltaic systems using artificial neural networks, fuzzy logic, genetic algorithm and hybrid models. International Journal of Energy Research, 45(1),1-30.
7. Xiang, L., Wang, P., Yang, X., Hu, A. & Su, H. 2021. Fault detection of wind turbine based on SCADA data analysis using CNN and LSTM with attention mechanism. Measurement, 175(4),1-10.
8. Zhang, X., Wang, H., Song, J., He, S. & Sun, C. 2023. Co-design of adaptive event generator and asynchronous fault detection filter for markov jump systems via genetic algorithm. IEEE Transactions on Cybernetics, 53(8), 5059-5068.
9. Yu, H. & Wang, T. 2021. A method for real-time fault detection of liquid rocket engine based on adaptive genetic algorithm optimizing back propagation neural network. Sensors, 15(21),1-15
10. Zhang, C., Guo, Q. & Li, Y. 2020. Fault detection in the Tennessee Eastman benchmark process using principal component difference based on K-nearest neighbors, IEEE Access, 8,49999–50009.
11. Chen, X. 2019. Tennessee Eastman simulation dataset. (https://ieee-dataport.org/documents/tennessee-eastman-simulation-dataset). Last accessed on 1 July 2022.