The Research of the Temperature Decoupling Control Based on QDRNN in Ammonia Synthesis Converter

Abstract:

Considering the coupling of the bed temperature control of sections in ammonia synthesis converter, by means of the analysis for quasi-diagonal recurrent neural network (QDRNN) and the gradient descent method (GMD), a dual-variable decoupling control scheme based on QDRNN is designed in temperature control system of ammonia synthesis converter. The structure and principle of the algorithm has been illustrated in this paper. Through the comparison of simulation results, it shows that the proposed scheme has better control effect than the conventional PID decoupling control scheme.

• Periodical: Applied Mechanics and Materials (Volumes 556-562)
• Pages: 2502-2506
• DOI: https://doi.org/10.4028/www.scientific.net/AMM.556-562.2502
• Online since: May 2014
• Authors: Xiao Qing Lu, Li Jun Hao*
• Keywords: Ammonia Synthesis Converter, Dual Variable Decoupling Control, GMD, QDRNN

References

[1] Huacong Li, Liye Rong and Yubin Zhu. The Aeroengine Multivariable Decoupling Control Based on QDRNN, in Journal of Aerospace Power, Vol. 11 (2007).

[2] Dray JS. Dynamic recurrent neural networks; A dynamical analysis. IEEE Transaction on Systems, Man, and Cybernetics, Part B: Cybernetics, (1996), 26(5): 692-706.

DOI: 10.1109/3477.537312

[3] Hongru Li, Jianhui Wang and Shusheng Gu. The Quasi Diagonal Recurrent Neural Network and Its Algorithm, in Journal of System Simulation, (2004), 07: 1542-1544.

[4] Decui Tang, Yanfei Zeng. The Quantitative Decoupling Control of the Paper's Moisture Based on Improved DRNN, in Computer Engineering and Applications, Vol. 23 (2009).

[5] Wanzhao Wang and Li Li. The Research of Decoupling Control Based on DRNN in Steel Ball Coal Mill Pulverizing System, in Northeast Electric Power Technology, Vol. 1 (2008).