Alvaro Rendón Montoya
Scientific Director
ECN Automation
Expert Contributor

Digital Twins: The Foundation for Autonomy in Mineral Processing

By Alvaro Rendón Montoya | Tue, 07/12/2022 - 13:00

With the new challenge established in the roadmaps of the leading companies in the mining sector, which seek to achieve process autonomy by 2030, automation and digitalization are prerequisites to start projects related to equipment and process autonomy.

With autonomy in mind, interoperability, connectivity and cybersecurity requirements emerge to give life and move toward the democratization of digital twins (DT), a concept that was used the first time by NASA in 2010 but, due to the limitations of the technologies, such as low computing power, data storage, low or no connectivity of devices with the internet, etc., DT had no practical industrial applications at the time.

What is a digital twin?

Digital twin refers to the virtual copy or model of any physical entity (physical twin). Both are interconnected via exchange of data in real time.

A DT platform reads from the Distributed Control System (DCS) the required process variables and, using simulation, phenomenological models, machine learning and optimization algorithms, creates a virtual representation of the mining equipment or process. This virtual entity has the capacity to predict wearing conditions, future process performance, etc., and transfer these new variables to the DCS.

DTs available for mining today

At this time, some DT projects that are in the piloting stage or first implementation include:

  1. Drill and blasting to obtain the better fines factor and ore particle size distribution.
  2. Mineral transporting to predict the mineral ore fed to the process plant and improve ore control planning.
  3. SAG Mills to predict liner wear, ball and mineral charge level.

Cloud, On-premise or Edge DT Implementation?

The cybersecurity policies defined by each mining company could determine how DT platforms can be implemented:

  1. Cloud. Only the process variables are exchanged between the mining plant and the DT in the cloud; an IoT device is installed on site to connect the data in real time to the cloud.
  2. On-Premise. Only local computing is used to implement the DT platform; no information is shared by internet.
  3. Edge. This architecture has the on-premise criteria but uses cloud capabilities to automate version actualization, re-train the models and remote monitoring of the DT.

Interoperability: the software capability required to play in the DT platforms game

Interoperability enabled” is the most powerful feature in a simulation and similar software to play with DT.

Interoperability is related to the app or software capability for sharing information in real time to modify their configuration and report results in an open architecture.

For example, in a SAG Mill digital twin, where the liner wear prediction model is trained using particle simulation software based on the Discrete Element Method (DEM), only DEM software can be used, where the simulation is created and started as a remote command. This model needs to be trained as an automatic task to solve the liner wearing rating for each mill operating condition. This internal autonomy of DT to manage multiple DEM simulations to populate a model with data, train and validate the model is intensive in the interoperability between all the DT platform components.

Autonomy = Hyperautomation + DT

The hyperautomation is resumed as the required automatic coordination between applications or systems to put in action the result obtained by artificial intelligence (AI) or the DT system.

With hyperautomation, we can integrate a SAG Mill TD with a Drill & Blasting DT to obtain a new Mine to Mill DT.

Joint DTs using hyperautomation create a digital twin or a real-time virtual representation of the organization (DTO).

Autonomous mine site operation needs a DTO at the top of the system.

Cybersecurity restrictions

Cybersecurity restrictions are greatly encouraging advanced analytics projects that are required prior to implementing DTs, so we must turn it around by executing OT (operation technologies)-only projects, and then implement DT in on-premise mode.

‘Digital Factory’ professional team

To face digital transformation projects and successfully adopt Digital Twins, mining companies must have a minimum team for their digital factory:

  1. Process Translator/Process Expert. For projects in a mineral processing plant, the expert could be a metallurgical engineer with experience in process control or a process control engineer with advanced knowledge in metallurgy.
  2. Data Scientist. Expert mathematician in machine learning algorithms.
  3. Python Programmer and UX/UI Designer.

The COVID-19 pandemic accelerated the adoption of DTs. The value DTs bring by increasing production, reducing maintenance costs, increasing user engagement and optimizing operations, is indisputable.

In the coming years, DT projects and autonomous systems will multiply in mining operations and this will imply important challenges in training current collaborators in systems to remotely control equipment.

To support the new autonomous systems, new professional skills will be required in the areas of cybersecurity, wireless networks, cloud computing, embedded computing and artificial intelligence.