How one mining company dug up control loop optimization’s benefits

Control loop optimization insights

• Mining company Rio Tinto employed Canadian engineering firm BBA to audit its Salt Lake City Kennecott site to seek out optimization targets.
• Control loop optimization was used to gather data and make appropriate upgrades.

Rio Tinto, founded more than 150 years ago, is among the largest mining enterprises in the world. The strength and scale of the company provides the opportunity for the operations team to explore innovations for improving its mining activities while accelerating both asset decarbonization and sustainable practices. At the Rio Tinto Kennecott site in Salt Lake City, the organization renewed its focus on process control to actively pursue optimization and efficiency initiatives following a disciplined approach.

In 2022, Rio Tinto engaged Canadian engineering and consulting firm BBA to conduct a detailed audit of the Kennecott site’s process control capabilities. The primary objectives were to identify opportunities for improving the stability of critical production processes and reducing the risk of unplanned downtime. These goals would be realized by first baselining the status of process instrumentation and other field devices and then optimizing the performance of the site’s many regulatory control loops. To accomplish their objectives, the team relied on analytics from a control loop performance monitoring (CLPM) software solution.

Laying the groundwork for control loop optimization

 Rio Tinto fields a large technical staff but performing an audit of this scale while maintaining normal plant operations necessitated additional outside support. With over 1,500 engineering and technical personnel, a global network of field offices and extensive experience in successfully executing complex process optimization projects, Quebec-based BBA was chosen to partner with staff from the Kennecott plant (see Figure 1).

Figure 1: The operations team at Rio Tinto’s Kennecott mining site partnered with Quebec-based BBA, to audit and optimize regulatory control loop performance. Courtesy: Rio Tinto and BBA

A group of BBA specialists took the lead, first assessing the plant’s instrumentation and mechanical assets. Spans for each input device were validated, and valves were both checked and tested for proper sizing and function. These preliminary findings revealed many field devices in need of repair, replacement or upgrade. More in-depth analysis would eventually be performed, uncovering other mechanical challenges that put reliability at risk.

Months were dedicated to meticulously performing the necessary instrumentation and equipment installation, configuration and performance assessments across a comprehensive portfolio of over 3,500 devices. The effort covered operational areas across the site such as flash smelting, an acid plant, a flash converter and anodes. With this initial phase of the audit completed, the team moved on to evaluating the performance of the site’s hundreds of proportional-integral-derivative (PID) control loops (see Figure 2).

Figure 2: Auditing Rio Tinto’s Kennecott mine operations involved an initial assessment of the site’s instrumentation prior to optimizing performance of hundreds of PID control loops. Courtesy: Rio Tinto and BBA

The quantity of PIDs employed in the regulation of production processes at the Kennecott site presented the team with a challenge. Control loops often interact with and among each other. With hundreds of control loops, it would be difficult to accurately assess each controller, let alone isolate and correct the true source of any performance issues. To achieve the audit’s goal, BBA staff procured Control Station’s PlantESP loop performance monitoring solution for the audit. Once installed, this CLPM solution would prove invaluable in analyzing control loop performance plantwide and allowing the team to achieve its goal of improving stability while reducing the threat of downtime.

The CLPM software was installed on an on-premises server, and it connected seamlessly with the plant’s existing AVEVA PI System. The software leverages the PI historian’s templates which provide a standard data structure and prescribes the tags associated with each type of automation element. Within hours of importing hundreds of controllers, the CLPM software began benchmarking the performance of each using the available historical data.

BBA staff trained Rio Tinto personnel on the software’s capabilities and its use to maximize the CLPM software’s benefits. Beyond the audit’s stated goal, Rio Tinto staff understood the value of proactively monitoring the Kennecott plant’s control loops to maintain performance and lock in any gains that were achieved. They took to BBA’s training and mentorship with that long-term view in mind.

Digging deeper into the data

While CLPM solutions provide insight into a range of PID performance challenges, many users find that analytics for identifying loop tuning issues and facilitating optimization often top the list of key CLPM features. For the audit team, the first step was to determine which loops were underperforming as opposed to being under the negative influence of other PIDs. The CLPM’s advanced forensics utilities would help the team to dig deeper and distinguish between symptoms and root causes.

The effects of a single poorly performing control loop can echo throughout a plant, undermining control across units and even across an entire plant. When used in tandem, two forensic utilities available within select CLPM solutions allow for the systematic isolation of root causes:

• Power spectrum examines the unique spectral frequencies associated with control loops and it facilitates the identification of other loops that share the same or similar “fingerprint” (see Figure 3).
• Cross correlation is a separate tool which uses statistics to establish the lead/lag relationship that exists between two or more control loops such that the root cause of poor performance can be pinpointed.

Figure 3: Power spectrum visually depicts control loops that share a frequency peak associated with a disruption. Once identified, the lead-lag relationship between the loops can be established using the CLPM software’s cross correlation utility. Courtesy: Rio Tinto and BBA

Tuning a PID control loop is a proven way of eliminating unwanted variability, but the benefits are generally limited when the root cause is left uncorrected. By using power spectrum and cross correlation in sequence, the CLPM solution directed the project team to the origin of each performance issue. Doing so allowed the team to efficiently address the site’s tuning needs at scale.

Teaming up for success

Since the commencement of the audit, work has proceeded in an orderly fashion with steady advancement of the team’s established goals. BBA has supplied monthly detailed reports to their Rio Tinto counterparts, updating on progress and recommending the next batch of prioritized tasks. These reports leverage intelligence gained from the CLPM solution and have aided in directing the team’s activities (see Figure 4).

Figure 4: PlantESP from Control Station is the CLPM software used at Rio Tinto to assess PID control loop performance. The software provides insights which BBA integrates into monthly progress reports for establishing new priorities based on the findings. Courtesy: Rio Tinto and BBA

Based on the solution’s findings, the team has decided each month on different target areas such as overall device health, particular and troublesome oscillations, or a group of valves exhibiting signs of mechanical strain. Recommended tasks always pushed the team in the direction of tightening control, minimizing energy or material usage, and reducing excessive wear and tear on costly assets.

In one such case the CLPM software’s mechanical health metrics had identified a control valve in the site’s acid plant that had been moving far more than required. The software’s recommendation for a new proportional and integral terms were implemented resulting in a reduction of the amount of valve travel by a factor of 30, energy saving, control improvements, and extension of the asset’s life. (see Figure 5).

Figure 5: The CLPM solution identified a valve within the acid plant that exhibited excessive movement and risked premature failure. The software’s tuning recommendations significantly reduced the amount of effort required to effectively regulate the process. Courtesy: Rio Tinto and BBA

In other cases, the CLPM solution had enabled the team to detect leaking valves or identify actuators with backlash problems. The software proved to be a significant upgrade over previous tools like Microsoft Excel that had been used at the Kennecott site in an improvised and manual manner to capture and trend control loop performance attributes.

Like the CLPM software, Rio Tinto’s selection of a third-party expert in process control proved invaluable. On several occasions the result of an investigation required BBA to overhaul the existing process control strategy. One such example involved changes in the control of fans that regulated boiler airflow. BBA determined it would be better to stop some fans during certain operating states so others could remain in an appropriate range of their operating curve. Following the site’s change control process, the proposed changes were documented before to justify the work and after to quantify the results. In all cases, the CLPM software has been used to generate supporting documentation.

As the work has progressed, a steady stream of issues has been identified by the CLPM solution and resolved by the team. Though the low-hanging fruit is gone, each incremental improvement moves the team closer to its objectives.

Seeing mining success through control loop optimization

While some control problems are visible to the naked eye, many others can only be discovered with the assistance of advanced analytics available in a comprehensive CLPM solution. With the help of CLPM analytics and other utilities, the project team at the Kennecott site has enhanced controller performance and achieved much of the audit’s stated goals. With time remaining before the project officially winds down, the team is confident in its approach and in the technology tasked with uncovering the next opportunity.

By partnering with an experienced contractor and a leading CLPM software supplier, personnel from Rio Tinto have successfully balanced the site’s day-to-day operational needs with its objectives of enhanced stability and uptime. It is anticipated this model will extend elsewhere throughout the organization.

Do you have experience and expertise with the topics mentioned in this content? You should consider contributing to our WTWH Media editorial team and getting the recognition you and your company deserve. Click here to start this process.