The Project
One of Australia’s leading mining company’s sought to improve its operational efficiency by leveraging modern data technologies to unlock a vast set of operational data and in-house knowledge.
The project aimed to identify and assess tools that could boost production, comparing various market-available technologies, including software with Machine Learning (ML) capabilities and advanced data analytics tools.
During the assessment process, a specific opportunity was identified in the copper solvent extraction area (CuSX), characterised by:
- Variable Conditions: The Pregnant Leach Solution (PLS) conditions, including copper concentration and pH levels, fluctuate every shift, leading to inconsistent copper recoverability and potential revenue loss.
- Continuous Adjustments: Metallurgists must adjust operational parameters every six hours to maximise copper transfer, a task complicated by the process’s continuous 24/7 nature.
Technical Solution
Phase 1: Evaluation of Technological Solutions
Phase 1 involved evaluating the various tools available in the market to compare:
- Cost: Analysing financial feasibility and potential ROI.
- Features: Examining the functionality and scalability of each tool.
- Integration Potential: Ensuring compatibility with the client’s existing systems.
It was evident that the CuSX area offered substantial opportunities for improving operations. Due to the complexity involved, the client concurred that creating a tailored solution for this specific area was the best option. This understanding shifted the project’s emphasis to the CuSX area for Phase 2.
Phase 2: Development of a Recommendation Model
The technical methodology involved using mass balance calculations for precise monitoring of material flows and employing machine learning to model extraction isotherms. This model can swiftly simulate the operation and evaluate thousands of scenarios, offering real-time recommendations for optimal parameter adjustments based on the most recent PLS data.
The data analytics team at WGA created an Excel dashboard linked with Python code, enabling the client to directly observe the model’s results in Excel and adjust input parameters via the Excel interface. The backend logic was managed by Python. This method improved decision-making and set the stage for future automation.
Client Outcome
The recommendation model enables metallurgists to receive data-driven insights, allowing for more accurate and timely adjustments to operational parameters. The client can achieve higher efficiency and productivity by optimising copper production processes.
Intermediate Step Towards Automation
Implementing the recommendation model moves the client closer to full automation. The objective is to link this model with the control system for automatic parameter adjustments based on new site data, thereby reducing human error and enhancing operational consistency.
This project showcases the team’s expertise in incorporating cutting-edge technologies into mining operations, resulting in notable enhancements in efficiency and productivity. By thoroughly assessing and deploying optimal tools, we have furnished the client with a solid solution that tackles present issues while also setting the stage for future innovations in automation.
The Copper Solvent Extraction and Electrowinning Recommendation Model is included in the following sectors & services.
Click on one to find out more about that sector or service.