Particle size distribution detection of laboratory ball mill based on vibration signal

Real-time detection of particle size distribution (PSD) in a ball mill is essential for optimizing grinding efficiency and reducing energy consumption. This study proposes a vibration-signal-based method for soft sensing of PSD during milling. First, the influence of particle size on the impact energy of media was studied based on particle breakage mechanism and the motion law of grinding media; Second, a Discrete Element Model (DEM) was conducted to verify the relationship between particle size and the throwing height of the media; Finally, vibration signals were mined to determine the proportion of particles with different sizes in the mixed particles based on “white box” characteristics of dendritic neural network. The experimental results show that when some parameters, such as: Ball Charge Volume Ratio (BCVR), total particle mass and rotational speed, remain constant, the larger the particle size in ball mill, the smaller the amplitude of vibration signal collected; Meanwhile, the dendrite neural network can effectively achieve soft measurement of PSD in ball mill.