Customers encounter daily uncertainties surrounding the SAG grinding process, primarily due to the many variables involved and the inability to directly observe or measure internal mill operations. The complexity of numerous factors adds to the challenge of understanding and managing the process effectively.
Questions regarding the total filling level and balls, the evolution of liners’ or balls’ wear, optimal load/ball ratio, and similar inquiries remained unanswered due to the absence of a reliable tool that could provide insights into the internal conditions of the mill. The lack of such a tool made it challenging to accurately assess and address these critical aspects of the grinding process.
In response to the information gap, ME Elecmetal has directed its efforts toward addressing these questions by leveraging cutting-edge technologies like artificial intelligence, machine learning, and more. This endeavor led to the development of the SAG Digital Twin, an operational digital twin for SAG grinding. This innovation aims to enhance analytical capabilities and comprehension of the intricate interactions among various variables governing the efficiency of the comminution process in SAG mills.
Empirical and phenomenological models might not be universally applicable across all mills, lacking specific focus and adaptability. Conversely, the SAG Digital Twin offers a distinct advantage due to its specialized and targeted approach.
Some methods enable the measurement or estimation of certain parameters, but they often do so discreetly and come with numerous disadvantages.
The operational digital twin for SAG mills, SAG Dtwin, serves as a virtual replica of the SAG mill. This innovative system offers an operational intelligence solution that:
This digital platform consolidates various prediction, simulation, and variable analysis mechanisms to diagnose the mill’s operation and enable control strategies focused on operational stability and productivity enhancement. It assimilates real-time process data from different sources such as DCS/PI/PHD and others, utilizing this information to develop operational variables. These variables, in conjunction with phenomenological models, DEM simulation, and optimization algorithms, mirror the operational state of the mill. Additionally, they offer recommendations for operational adjustments to optimize mill performance.
The system consists of crucial modules:
The creation of these core modules serves as a foundation for tailoring specific applications (custom modules) for individual clients:
The SAG Digital Twin provides a platform to showcase critical variables like the total filling level and ball filling level.
The integration of SAG Digital Twin facilitates obtaining crucial insights for optimizing the grinding process. It achieves this by continuously and in real-time monitoring various inputs, outputs, and other pivotal variables.
Ariel Medina, Director ME Digital Lab
Minera Los Pelambres has conveyed a strong appreciation for the accuracy of the diverse modules, the stability of the platform, and the adaptability to integrate new applications. Presently, they are actively engaged in direct connection tasks to integrate these variables into the advanced control of the SAG mill.
The insights derived from this tool have spurred the development of valuable new applications for the client. One such application involves the control and inventory management of the SAG mill, proving highly beneficial for both process control and enhancing conditions — both operationally and safety-wise — during scheduled maintenance.
Currently, we are piloting the operational digital twin with four of the most important mining operations in Chile, which have adopted this technology, including it in their own digital transformation roadmap. The miners’ enthusiasm and willingness to validate and adopt this technology reflect the perceived value and potential benefits they anticipate from integrating the digital twin into their operational strategies.
