Research Progress of Mill Load Detection and Modeling Methods

Abstract

This is a paper in the field of mining engineering. With the increasing demands for energy conservation and consumption reduction in mineral processing plants, ball mill load measurement is the key technology to realize mill control and optimize the grinding process. This paper summarizes the measurement methods of mill load in the recent years: differential pressure method, grinding sound method, vibration method, power method, ultrasonic method, indirect detection method based on multi-source signal fusion. Moreover, it summarizes and analyzes the mill load modeling methods that have emerged inrecent years. In the future, the indirect methods of multi-source information fusion will still be the main methods to detect mill load, modeling methods based on improvement of neural network and new online detection methods, as well as the establishment of efficient and accurate load detection models will be the main development direction of mill load detection.

Keywords

• mill load
• soft measurement
• modeling
• neural network
• online detection
• mining engineering