Rotary Kiln Burning State Recognition Based on POD Snapshots Method

摘要

Rotary kilns are widely used in cement and other industries. During the process of cement production, the calcination of the material in the rotary kiln is one of the critical links. The state and stability of the sintered material directly affect the calcination quality of the clinker. However, it is difficult to monitor the temperature straightly. Soft measurement is needed in this case, but traditional soft measurement methods are often inaccurate or have many limitations. We apply a strategy based on the image process and proper orthogonal decomposition(POD) to solve this problem. We first edit the video to some images, and then exact the RGB values of these images, making POD transformation to reduce the dimension. Finally, we put the RGB values of the image sequence into three-dimensional coordinates to classify the burning state to increase production capacity.