Grinding circuits typically contribute disproportionately to the overall cost of ore beneficiation and their optimal operation is therefore of critical importance in the cost-effective operation of mineral processing plants. This can be challenging, as these circuits can also exhibit complex, nonlinear behavior that can be difficult to model. In this paper, it is shown that key time series variables of grinding circuits can be recast into sets of descriptor variables that can be used in

more »

... modelling and control of the mill. Two real-world case studies are considered. In the first, it is shown that the controller states of an autogenous mill can be identified from the load measurements of the mill by using a support vector machine and the abovementioned descriptor variables as predictors. In the second case study, it is shown that power and temperature measurements in a horizontally stirred mill can be used for online estimation of the particle size of the mill product. These distances are collected in a matrix that is amenable to analysis by a large body of multivariate methods. As demonstrated in this paper, it includes the use of textural analysis, whereby information-rich descriptors of the dynamics of the comminution equipment can be extracted. These descriptors can subsequently be used as predictors in process models or to visualize the behavior of the equipment, as discussed in more detail in the paper. The paper is organized as follows. In Section 2, the proposed analytical methodology is described, followed by two illustrative case studies in Section 3. The results of the analyses are discussed in Section 4, while the conclusions of the study and further work are considered in Section 5. Recurrence Texture Analysis of Time Series Data A general overview of the analytical method is given in Figure 1 . It starts by dividing the generally multivariate time series (A) into segments of equal lengths, specified by the user (B). A distance matrix is calculated for each time series segment (C) and, from this matrix, a set of features is extracted (D) that can subsequently be used for further analysis, such as predictors in a classification model (E). A more formal description follows below. Processes 2018, 6, x FOR PEER REVIEW 2 of 19 These distances are collected in a matrix that is amenable to analysis by a large body of multivariate methods. As demonstrated in this paper, it includes the use of textural analysis, whereby information-rich descriptors of the dynamics of the comminution equipment can be extracted. These descriptors can subsequently be used as predictors in process models or to visualize the behavior of the equipment, as discussed in more detail in the paper. The paper is organized as follows. In Section 2, the proposed analytical methodology is described, followed by two illustrative case studies in Section 3. The results of the analyses are discussed in Section 4, while the conclusions of the study and further work are considered in Section 5. Recurrence Texture Analysis of Time Series Data A general overview of the analytical method is given in Figure 1 . It starts by dividing the generally multivariate time series (A) into segments of equal lengths, specified by the user (B). A distance matrix is calculated for each time series segment (C) and, from this matrix, a set of features is extracted (D) that can subsequently be used for further analysis, such as predictors in a classification model (E). A more formal description follows below.