Pressure drop prediction in a pneumatic conveying system with different curvature radius pipes for conveying particles

To investigate the system pressure drop distribution when conveying particle using different curvature radius pipes for the pneumatic conveying system, this paper measured the particle velocity distribution, particle-particle collision characteristics, collision energy loss, minimum pressure drop gas velocity, system pressure drop distribution, and power dissipation for R/D = 3.75, R/D = 5, and R/D = 6.25 pipes. Subsequently, the artificial neural network technique is used to predict the pressure drop of the pneumatic conveying system. It is found that the pressure drop of the system is lower when using the pipe with R/D = 6.25 for conveying particles. Compared to the pipe with R/D = 3.75, the reduction in power dissipation is 3.18 and 5.27% for conveying pellets when using R/D = 5 and R/D = 6.25 pipes, respectively. In addition, the energy loss of the system can be effectively reduced when using the pipe with R/D = 6.25 for conveying particles, which is more beneficial for the particles move in the pipe. The pressure drop model built with artificial neural network can predict the pressure drop value of the system more accurately within +/- 1.5%.