Prediction Method of Reciprocating Compressor Inlet Pressure and Shaft Power Based on GA – BP Neural Network
Author(s):
Y. H. Zhou, H. M. Liu, & X. Z. Yi
Source:
Y. H. Zhou, H. M. Liu, & X. Z. Yi, Prediction Method of Reciprocating Compressor Inlet Pressure and Shaft Power Based on GA – BP Neural Network, Journal of Mechanical Engineering Research and Developments, vol. 40, no. 1, pp. 81-87, 2017.
ABSTRACT: Aiming at the problems of inlet pressure and shaft power prediction in the optimization operation of natural gas compressor unit, a method of inlet pressure and shaft power prediction based on GA-BP neural network is proposed. The method by testing inlet pressure, the output flow rate and the output pressure are used to predict the inlet pressure and the shaft power of the compressor. The compressor operating condition is adjusted according to the inlet pressure, and the compressor operation station is adjusted according to the shaft power to achieve the energy saving and hence reducing consumption. The experimental results show that the method can predict the inlet pressure, and the error is less than 2.75%. By predicting the shaft power and adjusting the operation combination, the energy saving is about 10%.
Keywords: GA-BP neural network; Reciprocating compressor; Pressure prediction; Shaft power prediction.