The scroll flow and torque prediction with the wavelet neural network optimized by PSOA and BP
Ying Kong1, Xiao guang Chu2
COMPUTER MODELLING & NEW TECHNOLOGIES 2014 18(5) 303-307
1School of Medical Information and Technology, JiNing Medical University, Rizhao 276826, China
2School of Electrical information automation, QuFu Normal University, Rizhao 276826, China
A new Compressed Air Energy Storage(CAES) with scroll was proposed to promote the storage efficiency, which can be acquired by the scroll efficiency tracking control with the timely evaluation, but the flow and torque is not easy acquired because the sensors possessed the merits of high price, lower life-span and subjection to the disturbance, so a torque and flow prediction algorithm based on Wavelet Neural Network (WNN) is proposed adopting a hybrid learning algorithm combining Particle Swarm Optimization (PSO) with BP. Through the comparison between predictive and the experimental data and the scroll efficiency experiment, the proposed prediction method is validated and can be successfully used to improve Pneumatic conversion efficiency.