Research and Application of Wavelet Neural Networks of Particle Swarm Optimization Algorithm in the Performance Prediction of Centrifugal Compressor

Abstract:

The traditional method of centrifugal compressor performance prediction is usually the BP neural network, however, the problems are that prediction accuracy is not high enough, convergence is slow and it is apt to fall into local optimal solution. In order to predict the performance of centrifugal compressors more accurately and identify the implicit problems in advance, now we combine the particle swarm optimization, wavelet theory and neural networks, to establish performance prediction model of centrifugal compressor based on wavelet neural network of PSO. First, set the various parameters of wavelet neural network as the particle position vector X and the energy function of mean square error as the optimized objective function. By particle swarm optimization algorithm to iterate the basic formula to obtain the corresponding WNN coefficient and then use back-propagation algorithm to train WNN to approach any nonlinear function. Simulation results show that application of the prediction model can achieve the accurate prediction of performance and monitoring of centrifugal compressor. The prediction model has the advantages of simple algorithm, stable structure, fast calculation of convergence speed and strong generalization ability with a prediction accuracy of 99%, 13% higher than prediction accuracy of traditional methods, which has a certain theoretical research value and practical value.

• Periodical: Advanced Materials Research (Volume 187)
• Pages: 271-276
• DOI: https://doi.org/10.4028/www.scientific.net/AMR.187.271
• Online since: February 2011
• Authors: Sheng Zhong Huang
• Keywords: Centrifugal Compressor, Particle Swarm Optimization Algorithm (PSO), Performance Prediction, Wavelet Neural Network (WNN)

References

[1] Liyin Guo, Zhang Bangli, battalion commander repair. Wavelet neural network and its design method [J]. Pattern Recognition and Artificial Intelligence, 1997, lO (3): 197.

[2] ZHANG Q, BEAVENISTS A. Wavelet etworks[J]. IEEE Trans 0n Neural Networks，1992，3(6): 889.

[3] DAUBECHIES. The Wavelet Trandform. Time-frequen-cy Localizmion and Signal Analysis[J]. IEEE Trans On Information Theory, 1990, 36(5): 226.

[4] Xiang-Jun Wang, east, Jiang Tao, et al. A two-population evolutionary programming algorithm [J]. Journal of Computers, 2006, 29 (5) : 835-840.

[5] Lu Gang, Tan Dejian. Improvement on regulating definition of antibodydensity of immune algorithm. Proceedings of the 9th international confer2ence on neural information processing, 2002, 5: 266922672.

DOI: 10.1109/iconip.2002.1201980

[6] Yongyong He, Fulei Chu, B inglin Zhong. A h ierarch ical evo lut ionary algo rithm fo r const ruct ing and t rainingw avelet netwo rk s[J ]. N eural Comput ing &App licat ions, Sp ringer2V erlag, 2002 (10) : 357～ 366.

[7] V M Janardhanan, V Heuveline, O Deutschmann. Performance analysis of a SOFC under direct internal reforming conditions[J]. Journal of Power Sources, 2007, 172(1): 296-307.

DOI: 10.1016/j.jpowsour.2007.07.008

[8] F Calise, et al. Simulation and exergy analysis of a hybrid SolidOxide Fuel Cell (SOFC)- Gas Turbine System [ J ]. Energy, 2006, 31 (15) : 3278 - 3299.

DOI: 10.1016/j.energy.2006.03.006

[9] Chenzhen Wei, Zheng Guo crisis. Wavelet neural network predictive model Simulation [J]. Computer Simulation, 2008, 25 (6): 147-150.

[10] Zhourong Yi, Li Shuqing, cows will never , Wavelet Neural Network on Safety Management Evaluation[J], Coal Science and Technology, 2006(5): 67-70.

[11] Zhao Guorong, Wang Xibin, Gao Qing Wei , PSO neural network transfer of INS Alignment[J], System Simulation, 2010, 22(3): 670-673.

[12] Lidi Wei, Shao-hua, SHEN Yuan-tong. PSO based wavelet neural network [J]. Engineering Geophysics, 2007, 4 (6): 529-532.

[13] Liu Hongbo, Xiukun, Mengjun. Neural Network Learning Based on Particle Swarm Optimization Algorithm Research Study [J]. Mini-Micro Systems, 2005, 26 (4): 638-640.