Courtesy of Ali Vazirizadeh

Advanced Process Control (APC) is a technology that uses computer algorithms to optimize industrial processes and improve product quality. Despite promising results in applying APC in mineral processing plants, a significant majority of such plants still employ basic strategies. The deployment of APC faces several challenges, including technical, cultural and economic ones, which have hindered its widespread adoption.

From a technical standpoint, the primary obstacles to the deployment of APC are a lack of accurate process models, complex dynamic operating conditions, limited data availability and difficult handling constraints. However, mineral processing is not overly complicated, and there are established technical solutions to these issues that have been around for several decades. The question remains as to why these solutions are not yet widely used in operations.

Cultural challenges can be one of the biggest obstacles to the deployment of APC. There is a general skepticism among operations personnel towards the effectiveness and sustainability of any APC, which may be because many metallurgists and process engineers that wear iron rings have not received proper training on APC in their university courses. A review of the syllabi of 16 undergraduate programs most Canadian mineral processing engineers and metallurgists graduate from revealed that only 25 per cent included APC in their curriculum. Further analysis showed that a small fraction of graduates from those programs are absorbed by mining companies, with the majority finding careers in other industries. In contrast, many graduate programs in mineral processing have focused more on the application of APC and the development of better control and optimization solutions.

Despite recognizing the value of APC for mineral processing plants as far back as the mid-80s with hundreds of papers published, there remains a gap in the transfer of knowledge from graduate to undergraduate programs. If undergraduate metallurgists and mineral processing engineers were to receive training on the fundamentals of APC, they might be less skeptical and more willing to collaborate in its deployment. A mid-term solution to fill the gap could be operational training. To begin with, it is essential to provide an overview of the benefits of implementing APC, which include increased efficiency, reduced waste and improved product quality. This should be followed by an explanation of the fundamental components of APC, including sensors, controllers and actuators. It is also important to describe the various types of APC systems, such as model-based predictive control and fuzzy logic control. Next, it is recommended to provide hands-on training that is relevant to the trainees’ operations. For instance, showcasing successful APC implementations in similar plants through case studies or real-world examples can help to illustrate the practical applications of APC. Additionally, practical training exercises that utilize operational data to build a simple APC in a simulation with the trainees can reinforce the knowledge transfer. And finally encourage questions and feedback, and continuously monitor progress. This approach can help to ensure that the operation’s staff understands and appreciates APC and is better able to implement the technology effectively in their plant.

Related: RCT’s ControlMaster Automation platform boasts new features for a wide variety of mining equipment

Another obstacle is the absence of standardization across commodities, processes, instruments and controllers, which makes it challenging to develop a solid business case for APC deployment. One outcome is that it is difficult to compare results across various operations and commodities. Conducting case-by-case studies appears to be a more practical approach to ensure a thorough evaluation of the business case.

Business case studies of brownfield projects must answer the question of the return on investment. While qualitative analysis and benchmark studies can demonstrate the value of the investment, quantifying the payback for a particular operation can be complex. One potential solution is to adopt a stepwise approach to mitigate investment risk. This can be done by first evaluating the required infrastructure and then modelling and simulating based on historical data. Finally, the results are compared to actual data. This approach of using training and validation sets (pioneered by the artificial intelligence/machine learning fraternity) will provide clarity in value proposition even before engaging with APC technology providers.

Greenfield projects face a challenge in that historical data do not exist to be modelled, which makes it even more difficult to make a compelling argument for the deployment of APC. To define the value proposition for APC in these projects, we should take a step back and look at the entire project. During the design phase of a plant, technical and financial studies are carried out using lab tests and piloting data. However, achieving similar results in full-scale operations can be tricky due to the variability of ores and changing operational conditions.

While adding extra capacity has been a solution in the past, it may no longer be the most appropriate or cost-effective approach as ores and processes become more complex. For example, to ensure that the reported flotation recovery achieved at the laboratory scale can also be achieved at the full scale, it is necessary to consider additional capacity to allow sufficient time for reactions. However, flotation recovery is also highly sensitive to disturbances, which may result in decreased recovery. This is where APC can play a crucial role in mitigating the risk of recovery decline. Therefore, it is recommended to incorporate a suitable APC system, along with reasonable design factors, during the design phase for optimal effectiveness. It is important to note that all the necessary provisions for developing such a system should be considered early on in the project life cycle. This includes adding more instruments, selecting advanced control systems and employing APC experts. While this may require additional capital, installing similar systems after building the plant will be more expensive, if feasible. Embracing new technology at an early stage can prevent resistance to change from people and ensure the long-term sustainability of the solution.

Engineering companies should therefore incorporate APC strategies into control philosophies and not solely rely on the basic packages. Process technology providers should offer at least preliminary APC systems as a default rather than as an option. Surprisingly, some APC technologies, such as expert control and fuzzy logic, considered advanced decades ago, are still regarded as cutting-edge in the mineral processing industry today. While these technologies were indeed advanced at the time, other industries have since evolved and adopted more advanced strategies such as model predictive control, adaptive control, and, more recently, reinforcement learning. The mineral processing industry must demand more from technology providers and expect them to deliver effective solutions and play an active role in educating process engineers and metallurgists about these advanced technologies and the benefits they have shown in practice in mineral processing and other industries.

Ali Vazirizadeh, a metallurgist,  is co-founder and director of Aisimpro, an AI implementation in mining company.