Speed Prediction of ESPCP System Based on Neural Network

Abstract:

ESPCP is a new way of well lifting. The speed optimization is the main way to improve economic index of ESPCP system. By analyzing the main factors of speed of ESPCP system and using the technology of artificial neural networks, established an artificial neural network model for output speed about oil viscosity, pressure difference of the pump two ends, magnitude of interference between stator and rotor as the input variables. Through the use of additional momentum and adaptive learning rate method to predict the learning samples, the results can fit experimental data. The results of recalling and forecasting are accurate. It is shown that the model is of high precision and reliability. It also provides a new calculation for ESPCP system speed prediction and optimization.

• Periodical: Applied Mechanics and Materials (Volumes 44-47)
• Pages: 3569-3573
• DOI: https://doi.org/10.4028/www.scientific.net/AMM.44-47.3569
• Online since: December 2010
• Authors: Xuan Luo, Shi Jie Wang, Xiao Ren Lv, Hao Sun
• Keywords: Artificial Neural Network (ANN), ESPCP System, Output Speed

References

[1] Wang S J, Li Q: Electrical submersible-motor-driven Progressive Cavity Pumping Technology and system design (Metallurgy Industry Press, 2006), in Chinese.

[2] Li C, Wang S J: Development & Innovation of machinery & electrical products Vol. 1 (2005), pp.102-104, in Chinese.

[3] Fei si technological products research and development center. Neural network theory and MATLAB7 realize(Beijing: Publishing House of Electronics Industry, 2005), in Chinese.