Hydrocyclone underflow monitoring using image processing methods

Until a few years ago there has been little development of image processing systems in the minerals industry, mainly owing to the computational demands of image recognition and interpretation. Due to rapid progress in computer technology at present however, image processing systems are increasingly being investigated in order to solve complex process engineering problems. In this investigation videographic analysis of the hydrocyclone underflow discharge is applied primarily as a diagnostic tool to assess the performance of the classifier. The discharge spray angle is a visual parameter which immediately identifies normal (umbrella profile discharge) or anomalous (roping discharge) operation, and can be directly interpreted to characterise internal flow properties such as pressure drop. Also, this information can be applied to estimate the feed stream particle concentration and ultimately the hydrocycloneʹs cut size and can serve as a basis for considerable improvement in the on-line control of mineral processing circuits as a whole.