International Mining | BY Paul Moore

A detailed insight into Accenture’s capabilities in helping shape mining’s future in terms of advanced data analytics and machine learning

IM Editorial Director Paul Moore took some time to catch up with Constantino Seixas, an Accenture Managing Director and Latin America lead for Industry X Digital Manufacturing and Operations.

The company sees the value of a connected mine in digging deep into the wealth of data available across the entire value chain to provide integrated, end-to-end situational awareness and systemic management, and it has been one of the leading companies helping mines progress from structuring through to implementing data-led transformation with AI front and centre

Where does Accenture see the future of mineral processing in terms of the industry moving to greater use of AI and machine learning?

Data analytics is essential to mining activities. Almost all mining companies implemented data historians more than a decade ago and accumulated huge quantities of data from all production processes, including those in the mines, concentration plants, smelters, refining units and logistics – including ports. However, very few companies are using this data to its full potential yet. Data is fundamental to understanding process behaviour, identifying operating patterns, detecting where the bottlenecks are, calculating variability, predicting quality and asset failures and for root cause analysis. Absolutely every unitary process in mining will benefit from data analytics, automation and artificial intelligence (AI) technologies. If you are drilling, you can capture much more data than before, including for example on the rock hardness, based on the drill torque, and rock composition, using new instruments that are able to characterise the ore. After blasting, the fragmentation can be analysed by image processing, using drones, or by measuring rock size distribution inside the shovel bucket. This improves possibilities in optimising fragmentation and selecting the ore that will be fed into the concentration plants; that is part of the precision mining concept. Of course, all of that depends on determining the exact position of mining equipment, using a high-precision Geographical Positioning Systems (HPGPS), and having real time connectivity. Safety applications also depend on data analytics. Video analytics can detect the presence of operators in the plant red zones, workers positioned under a hanging load, and the level of fatigue of truck drivers or crane operators, to name a few examples. New sensors are used to monitor tailing dams and this data is analysed by real-time algorithms that can detect an alarm condition. The mining industry is using AI in building models to make inferences about the future state of a system, or to create diagnostics to find the root cause of an anomaly.

Constantino Seixas, Accenture Managing Director and Latin America lead for Industry X Digital Manufacturing and Operations

What processes do you through with customers to help them achieve a transformation in their data analytics using AI?

The areas in which AI can be applied are manifold. Accenture is using AI in production planning, fleet dispatching, process control, quality forecasting, instrumentation data validation, creation of virtual sensors (soft sensors), asset failure prediction, multi-plant synchronisation and product-blending optimisation, to name but a few applications. In all cases, value realisation can be measured. Achieving a good result is the expected consequence. Sustaining the result is another story and depends on defining permanent optimisation programs. Accenture also deploys data-driven consulting in its day-by-day activities. When consultants are visiting a plant, they collect data that helps to understand if there are opportunities for improvement and invariably there are. Suppose the team is visiting a concentration plant and collecting flotation plant data. By studying the variability in flotation recovery and in fine copper production, it is possible to measure the productivity gap and to estimate the benefits to be achieved from addressing this.

In some cases, the journey begins as an ad hoc problem-solving project, with the client facing a challenge to understand a quality, production or maintenance problem and we then help them deploy data analytics to solve it. From this experience, the client learns about the power of data and understands that the data they already have is an asset to be explored. Accenture can then structure a data-led transformation roadmap, showing all possible applications of AI across several key dimensions: operations, maintenance, asset management, logistics, sales, procurement and HR, to name a few. Accenture and the client can then collaborate to prioritise the projects and define a roadmap for implementation. One of the first steps in this process is the definition of the data analytics platform. It will collect data in a data lake and supply data analytics tools to extract knowledge from the data.

In general, the client already has data in a historian such as Osisoft-IP, Aspentech IP.21 or IBA, which can be used to generate immediate results. When focusing on any program, we can also detect information gaps – things that are not being measured – because there is no instrumentation, or the instrumentation is there, but there is no connectivity with the historian and the data needs to be collected manually. Part of the data-led transformation roadmap involves creating an infrastructure, involving edge computing and the cloud to collect all the necessary data for each prioritised program. Edge computing is used to process data locally. If the objective is to measure bubble size, speed, collapse rate and colour in flotation, it is better to process the information close to the site and edge is optimal for this purpose. If the objective is to use deep learning that requires more powerful processing and a GPU, the image can be sent to the cloud to be processed there.

Suppose that, during the construction of the data-led transformation roadmap, tailings dam management is selected as a program. Data from piezometers, inclinometers, water level sensors, flowmeters, radars, seismic sensors, robotic total stations, and GPRs (ground penetrating radar) will be needed to feed the data lake, as well as images collected by fixed cameras and drones, InSAR satellite information and data from geotechnical inspections. It is possible that not all this information will be available, and firms must continue to evolve their capabilities until it is possible to gather all necessary data in a reliable way, reducing manual data input.

For automated data, coming directly from instrumentation, it is common to build a data validation layer that is called a data integrity engine, on top of the data historian or one level above, to check if the acquired data is correct. Univariate and multivariate algorithms are used for data validation. To clarify, if an instrument is always indicating the same value and that measurement is frozen, the instrument would then be selected for local inspection. An algorithm that calculates the standard deviation of this measurement can detect the problem automatically. Smart instruments facilitate the diagnostic and are able to inform the user about multiple abnormal conditions. In fact, data analytics verification works in any condition, even with non-intelligent instrumentation.

Accenture supplies all services for executing this kind of program, including those associated with instrumentation, communications infrastructure, edge computing and cloud, historian and applied intelligence. We deploy data engineers with deep mining expertise who can contribute effectively, bringing knowledge from their wealth of previous experience. Most of the mining companies are also structuring internal data analytics teams inside their IT department, or as an independent entity, closer to their operations. Accenture provides a collaborative program to support those teams. This includes theoretical and on-the-job data analytics training that speeds up deployment of the client’s internal programs. Accenture also provides services to sustain those programs so the momentum won in the project phase will not be lost.

Can you give an example of how you have been able to work with clients to add value using real time data analytics?

Accenture developed a solution that identifies several abnormal behaviours in the zinc electrowinning process, based on real-time measurements. The normal way to identify a short circuit in a pot line is to detect a hot spot that represents the short circuit point where a kind of polypus will short-circuit anode and cathode plates. To make this detection the operator typically uses a manual infrared scanner and needs to walk on the top of electrolytic cells. However, Score Zinc, the system developed by Accenture, detects this as well as other potential problems, based on pot voltage, electric current and temperature measurements. This system then alerts personnel to an abnormal condition and indicates where the problem is located, saving time and reducing the fault duration. This improves the current and energy efficiency in the pot line. A model, or digital twin, is used to compare the normal and bad electrolysis behaviours and is the crux of this solution.

What unique strengths do you bring to mining and mineral processing customers and what are your differentiators?

Accenture delivers end-to-end value. We help in all project phases from conceptual design to value realisation. Given any business situation, Accenture will look for improvement opportunities. Accenture is an agnostic company, meaning that we can deploy any technology, product, or application according to client needs. We have alliances with most technology providers, and we are trained in their solution suites. Accenture can deploy any ERP or cloud platform and we have thousands of trained people in each technology. Depending on the client’s choice, it can sometimes happen that we don’t have a trained team in some product or technology in a certain geography or that all our team members are already working on another project with another client. In this case, it is very easy to train another team in a new product, because for an expert in a certain area, migration to another product brand is very straightforward.

What differentiates Accenture when supplying a solution is the industry knowledge that we bring and our internal ecosystem that allows us to benchmark our solution to what other colleagues are doing with leading clients all over the world. When we talk about doing data analytics inside a remote operations centre, for example, we bring the experience of 25 previous projects, most of them in the mining segment. We deploy all solution layers from instrumentation to ERP, including robotics. For a decade now, Accenture has been building the broadest suite of engineering and manufacturing services, which we call Industry X. We have acquired almost 30 companies to help clients with their core operations. Recent examples include Pollux, a provider of industrial robotics and automation solutions in Brazil, and umlaut, an international 4,200-people company providing engineering consulting and services.

How far off is a true autonomous mineral processing plant in your view with only minimal maintenance inputs?

This horizon for achieving a truly autonomous mineral processing plant is not very far away. Mining companies are very eager, when it comes to adopting autonomous systems. We have already seen several companies implementing autonomous drilling, autonomous hauling, autonomous train loading, autonomous stacking and reclaiming, autonomous sampling, to name but a few applications. Some of these not mature yet as solutions, but are being deployed. Autonomous ship loading will be the next big focus. Maintenance is still a laggard when compared with operations. Most miners already have condition-based maintenance in place, and several are implementing predictive maintenance that will reduce unplanned downtime and reduce maintenance costs, but it is important to emphasise that self-healing plants are just a dream right now. Mechanical equipment requires maintenance. Robots can perform some functions like exchanging fabric filter elements of a press filter, yet they are not able to replace people in all maintenance tasks, as it stands. There are very good examples of advances in automatic equipment monitoring, however.  Today, a small wireless box can be used to monitor a motor in real time and detect more than a dozen possible problems in a centralised way. 5G has also arrived, connecting everything in a mineral plant. This will drastically reduce the cost of implementation of digital solutions. When it comes to people safety, operations, maintenance and logistics, all will be positively impacted by the convergence of AI and improved connectivity.

This blogpost is originally from Aquantico | Challenge, innovate & deliver value

Mining.com | BY Martin Provencher, OSIsoft

With the coming age of Industry 4.0, it’s no longer prudent, strategic or economically smart to wait until a critical mining asset has broken down to fix the machine. Equipment breakdowns are often costly on many levels — downtime means productivity is affected, parts can be expensive and then there are the wasted outlays on labor and energy.

Data analytics paired with predictive maintenance can be a virtual goldmine for mining operations, with initial cost reduction and productivity gains of an estimated 10% to 20%.

The five stages of mining maintenance

The evolution of maintenance in the mining industry has come a long way in the last decade or so, aided by the availability of real-time data. There are five common maintenance approaches that can be applied to mine assets: reactive, preventative, condition-based, predictive and prescriptive.

Moving from reactive to preventative maintenance will help improve asset reliability, but it might not be effective as there will still be unplanned downtime and costly repairs that could have been avoided. Calendar-based maintenance often proves to be inefficient because 82% of machine failures occur at random patterns.

Condition-based monitoring, the monitoring of machines when they are still running, is the first step toward adopting a forward-looking maintenance strategy. Data can be collected online through network connectivity to sensors or offline through operator rounds or other means, depending on the criticality of the machine.

Predictive maintenance advances the condition-based approach further by using model-based anomaly detection. It relies on the online collation of sensing data and uses data analytics to predict machine reliability.

The ultimate level of maintenance, prescriptive maintenance, involves the integration of big data, analytics, machine learning and artificial intelligence. It takes predictive maintenance a step further by implementing an action to solve an impending issue, rather than simply recommending an action. 

The most important factor in the digital transformation of your maintenance strategy is access to real-time operational data. Reaching the fourth level of maintenance (predictive) is achievable now, and requires the application of advanced analytics.

Establish an operational data infrastructure

The first step is establishing an enterprise operational data infrastructure that can capture real-time data coming from sensors, manufacturing equipment and other devices, and transform it into rich, real-time insights, connecting sensor-based data to systems and people.

This first step is fundamental to providing insights for later analysis. Not only will a real-time operational data infrastructure help improve asset reliability, but having a single infrastructure in place will improve process productivity, energy and water management, environment, health and safety, quality, as well as KPI and reporting.

Enhance and contextualize data

This step refers to how data is stored and enhanced (in other words, providing context) to become information. For example, even though data is collected from a sensor, analysts need to know if the equipment is running or has stopped either due to a failure or activation of an emergency stop button. Without this context, data doesn’t have much value. Also, recognizing what data is important and relevant to an organization is equally as vital.

Implement condition-based maintenance

Implementing data involves prioritizing certain assets, identifying the conditions that lead to eventual failure and implementing those conditions on specific assets within a real-time operational data infrastructure to automate condition-based monitoring. For example, when a bearing temperature starts increasing outside of its normal operating temperature, it means the bearing will eventually fail.

Reliability engineers already know a lot of these failure patterns, which are often found through reliability-centered maintenance analysis following a failure. All of these known patterns should then be implemented as CBM inside a real-time enterprise operational data infrastructure.

Implement Predictive Maintenance 4.0

Your chosen enterprise operational data infrastructure — in conjunction with advanced analytics and pattern recognition tools — will provide real-time, actionable intelligence that empowers your business to optimize operations. Used together, these tools will automatically determine the patterns that lead to an eventual failure.

Perhaps the best way to see the benefits of condition-based and predictive maintenance is through real-life circumstances.

For example, Barrick Gold’s Pueblo Viejo, the largest producer of gold in the Caribbean, wanted to improve the asset health monitoring system of their haul truck fleet to improve maintenance efficiency and costs. The challenge involved providing real-time information of 34 haul trucks using installed systems and at minimum cost.

Prior to using condition-based maintenance, reliability and maintenance managers relied on incomplete or delayed information to make decisions.

“We used to use the in-vehicle sensors to investigate why a truck failure had happened,” Ted Olsen-Tank, a senior metallurgist for Barrick Gold, said. “Now we can be one step ahead of a failure and be more proactive.”

SINCE ITS INCEPTION IN 2018, THE PREDICTIVE MAINTENANCE PROJECT AT BARRICK’S CORTEZ MINE HAS BEGUN TO PAY OFF AS MORE THAN HALF A DOZEN MAJOR EQUIPMENT FAILURES HAVE BEEN AVOIDED

Barrick Gold saved $500,000 due to their newfound ability to detect and address failures as well as reduced the total number of failures from engine, brake or suspension issues by 30%.

Barrick is using predictive maintenance at their Cortez Mine in Nevada. The machinery involved in the gold refining process at that location is expensive, and so is downtime caused by mechanical failure. Barrick Gold invested in predictive maintenance at Cortez Mine by using sensors and machine learning to detect potential equipment problems before they escalated into failure. The machine learning algorithms use frequently collected sensor data to generate an equipment health score, which can be tracked for declines that might indicate a potential problem.

Since its inception in 2018, the predictive maintenance project at Cortez has begun to pay off quickly as more than half a dozen major equipment failures have been avoided. To put that into perceptive, a single early fault detection for one piece of equipment alone saves the company $600,000.

There are powerful economic incentives to leverage real-time data and predictive maintenance. Companies can benefit from reduced costs, open new revenue streams, extend equipment life and increase production capacity.  

(Martin Provencher is industry principal of mining, metals and materials)

This blogpost is originally from Aquantico | Challenge, innovate & deliver value

PWE Magazine

Svendborg Brakes says its newly launched Industrial Internet of Things (IIoT) solution is empowering the mining industry to boost its operations. PWE reports.

IIOT and Big data optimisation

Businesses can upgrade their existing systems to reach the next level in advanced braking control and predictive maintenance. Users can benefit from the combination of Big Data analytics, Cloud and Edge Computing to optimise equipment’s reliability and service life while reducing the cost and time associated with maintenance.

Predictive maintenance applied to critical machinery

The reliability of brake systems for key mining components, such as conveyor drive stations, bucket wheel excavators and crane applications, is crucial to maintain peak productivity and throughput. Relying on highquality brakes is a good starting point, which can be complemented by advanced data-driven technologies for real-time remote control and condition monitoring, such as the latest offering from Svendborg Brakes, a brand of Altra Industrial Motion Corp.

The solution combines IIoT and data mining technologies to offer remote, real-time monitoring and predictive maintenance to any existing or new braking systems from Svendborg Brakes’ portfolio. These can be connected to all versions of closed loop control SOBO (Soft Braking Option) systems, including the latest SOBO iQ solution. Users can also benefit from a dedicated IIoT technology to connect with alternative control systems. In all these cases, key information on braking operations and the status of its components is collected on the brake system. Key parameters include system pressure, current state of the brake and its piston, brake fluid level and temperature.

These data are pre-processed by the controller, using Edge Computing, to address time-critical tasks and support a prompt action in the brakes, such as stopping and holding as well as generally all internal processes within brake control systems. Advanced analytics is then performed in the Cloud, in order to assess the conditions of the brakes and develop key predictions to optimise their maintenance. More precisely, actual data and models on usage and wear provide an early warning when component servicing or replacement are required.

End users can access the actionable insight generated by the IIoT solution via a user-friendly condition monitoring platform. As a result, they can benefit from an immediate and comprehensive overview of the braking system. Consequently, they can also schedule accurate maintenance activities and regimes.

Ultimately, businesses in the mining sector can shift from preventative to predictive maintenance. This, in turn heavily reduces the time and cost associated with on-site inspections and maintenance activities at mining facilities, which are often in isolated, outlying locations. Also, the service lives and uptime of braking components are maximised while reducing the likelihood of unexpected failures.

The IIoT solution from Svendborg Brakes is also able to flag any unauthorised access or attempt to open the cabinet door. Therefore, it offers a tool to prevent interference or tampering with the system.

https://www.linkedin.com/company/svendborg-brakes-as/

This blogpost is originally from Aquantico | Challenge, innovate & deliver value

Plantservices.com | By Sheila Kennedy, CMRP, contributing editor

Maintenance and reliability best practices are continually improving and so are the technologies that support them. Firmly embedded in the realm of the “best” is predictive maintenance (PdM), which combines real-time monitoring of asset condition, environmental, and/or operational data with smart analytics to detect, assess, and forewarn of impending problems.

Prescriptive maintenance (RxM) takes PdM a step further by prescribing corrective actions for deteriorating conditions and including them in the alerts. RxM is a newer concept just gaining ground that is made possible by machine learning (ML), artificial intelligence (AI), and the internet of things (IoT).

“Preventing dreaded unplanned downtime provides a clear business benefit and driver for adoption of PdM,” said Ralph Rio, vice president at ARC Advisory Group. “Unplanned downtime often cascades into major losses including revenue, WIP materials, and larger equipment repair costs.”

Additionally, the costs for implementation and support of a PdM application have dramatically lowered with the IoT, cloud platforms, microservices for analytics, and wireless sensors, Rio observes. “Clear benefits and lower costs have improved financial justification and driven adoption in a wide range of industries,” he added.

The six case studies summarized below illustrate some of the numerous ways PdM and RxM are transformative.

BASF / Schneider Electric

BASF, the largest chemical company in the world, has digitalization as a corporate strategy. This includes using data to better forecast maintenance requirements and reduce unexpected shutdowns.

One such initiative involved an electrical infrastructure expansion at BASF’s Beaumont, TX, plant. Since production efficiency depends on having predictable electrical power, the plant chose to enable remote monitoring and management of its new power distribution substation’s operations and asset health with EcoStruxure Asset Advisor from Schneider Electric. The data-driven, IIoT-enabled service employs asset sensors for continuous condition monitoring along with predictive analytics to identify threats that could lead to asset failures.

BASF also has access to customized, proactive advice about how to prevent failures and improve its maintenance strategies through its partnership with Schneider Electric’s Connected Services Hub (formerly Service Bureau).

Read “How 7 companies are accelerating PdM and RxM at their plants”
The use case involved having more than 100 condition variables continually collected, measured, and computed for 63 substation assets. The asset data is monitored and analyzed from a digital dashboard that provides 24/7 insight into the substation’s global health index and specific asset statuses, from any location.

“We were surprised by some of the things that we found out. Asset Advisor is helping us to prevent catastrophic failures,” said Lee Perry, electrical design engineer at BASF. The cloud-based service “connects assets so we can look at the health of not just our electrical distribution equipment, but also the motor control centers and the motors that actually drive the process,” he explained.

“If there’s an issue, then we get an email from the Service Bureau saying, ‘We’re noticing this.’ I can log in to it remotely, look at the same information, and help troubleshoot that issue,” added Perry.

With around-the-clock access to data and expert guidance, the plant has the information it needs to make the right decisions at the right time, perform PdM to optimize asset health, and take steps to improve the efficiency of critical electrical distribution assets. These actions consequently benefit plant uptime, performance, productivity, and safety.

ALCOA / Senseye

PdM is a key part of Alcoa’s strategy to become a more stable and profitable organization. To help automate PdM and reduce downtime and maintenance costs, a proof-of-concept (POC) project involving 50 assets across two casthouse systems was conducted at its Fjar∂aál aluminum smelter in Iceland. The solution is now being expanded to at least 1,000 assets in Fjar∂aál and is scalable enterprise wide.

Senseye was chosen to unify and synchronize the equipment sensor data in the plant’s OSIsoft PI ecosystem with maintenance data in its Oracle eAM solution. From the current and historical ingested data, it automatically builds models and starts learning, without having to set up parameters or alarm levels. Isolated peaks in raw data create signatures that reveal trends and hidden failures. Predictive analytics identify what is happening and why, and can provide prognostic insights on the asset’s remaining useful life.

Operators are automatically notified of cases needing attention and can drill down for further details. It is “like having a thousand eyes in the plant” letting you know when data streams start to move away from normal, explained Árni Einarsson, reliability implementation manager at Alcoa in his presentation at OSIsoft PI World San Francisco 2020.

For instance, an idling current increase provided an indication of a fault in the HDC saw motor system. It was determined a belt guard had come loose and was in contact with the sawing drive, damaging the belt. Replacing the belts and fastening the belt cover during a maintenance shutdown resolved the issue, avoiding 12 hours of unplanned downtime.

A sensor failure detected in a coiler rod cropping shear motor prompted the discovery that a lower pinch roller sensor had come loose, leading to a sharp increase in shear motor torque. Re-fastening the sensor returned the torque to normal levels, avoiding three hours of unplanned downtime.

In total, Alcoa reduced unplanned downtime of the machinery by up to 20 percent and achieved full ROI in 4-6 months. “With the POC completed, we were able to scale the solution to other parts of the business,” said Einarsson. “The focus is on finding the best business cases to continue.”

Duke Energy Renewables / Seeq

Duke Energy Renewables, an owner/operator of wind and solar farms across the U.S., used advanced analytics and ML to automate profiling and detection of failing contactors on one of its wind turbines. “This was a trial run with one use case that was interesting, the data was available, and it seemed like a good candidate. We wanted to get up and running fast and to see if this even works for us,” said Abhi Hullatti, manager of performance analytics at Duke Energy Renewables, in his presentation at the ARC Industry Forum Orlando 2020.

Hullatti described how each turbine has six contactors that help to ensure that when the generator kicks in, it ramps up smoothly and synchronizes with the electrical grid. When any of the six contactors fail, the turbine goes offline for 2-10 days for diagnosis and repair. Turbine contactors at one site tended to fail more frequently than at others. Automating the prediction of impending failure would enable PdM, improve uptime, reduce maintenance costs, and allow better management of spare parts inventory.

Using an automated profiling tool from Seeq, a model was trained to look for the contactor fault error code and plausible leading-indicator signals (reactive power, active power, current, wind speed, rotor speed, and generator speed), to recognize what normal behavior looks like, and to provide one-hour advance notification of a fault.

The model ran against 2.5 years of cleansed signal data for the turbine and it found 12 occurrences of the error code, mostly within the few months preceding a failure. There were no false positives leading up to that outage, and no further error codes or predictions followed once all the contactors were replaced. The model can run continuously once it is validated and it will give automatic notification when something is going wrong, said Hullatti.

“The results are extremely promising, so we want to next expand this model to the rest of our turbines at that site, and also, more interestingly, look at other failure modes. That will be a lot more complex but also far more financially rewarding,” observed Hullatti. The return on the ability to catch a generator that is about to fail, and act proactively instead of reactively, is in the hundreds of thousands of dollars per event, he explained.

Global Mining Company / Uptake

A leading global mining company needed to predict and prevent failures that can cause unplanned downtime in its material transport operations. The company owns and operates a network of private railway tracks, rolling stock, and signals that move iron ore from its mining sites to the seaport. Any faults, delays, or breakdowns in the process of moving the high-value material is not only costly from an equipment maintenance and operational performance perspective, but it can also be potentially catastrophic.

The trains and rail infrastructure already produce immense amounts of data from wayside telematics devices. If harnessed, the Wheel Impact Load Detection (WILD), Wheel Condition Monitoring (WCM), Hot-Box Temperature Detection (HBD), and Bearing Acoustic Measurement (BAM) data could provide operational and condition insights into anomalies with sufficient time to take corrective actions.

The mining company chose Uptake, an industrial AI software company, to provide the necessary visibility into degrading asset conditions across the network and advanced analytics to prescribe recommended solutions. The software is now analyzing the four telematics data sources for indications of faulty wheels, wheel bearings, and axles on 10,000 rail cars.

Events are ranked according to criticality. Each high-severity wheel or bearing downtime event for a rail car costs the company at least $6,000 and requires immediate attention to return it to operation. Previously unforeseen, the mining company now has roughly a weeks’ advance notice on high-severity wheel alerts, 11 days’ notice of high-severity bearing alerts, two weeks on medium-severity alerts, and a month on opportunistic events.

With the advanced notice of issues and prescriptive alerts, the company can proactively bundle its maintenance activity, significantly reducing downtime and avoiding rail car failure. As a result, it is able to reduce its single-car unscheduled maintenance events by an estimated 50 percent, from 1,850+ per year to 925+ per year, which represents approximately $34 million in savings over five years.

Cement Plant / Aspentech

A cement plant that had struggled with cyclone blockage in its cement kilns was one of several PdM case studies shared by Chris Williams, global director of asset performance management (APM) services for AspenTech, at his ARC Industry Forum Orlando 2020 workshop. Frequent blockages in filters and cyclones limit production and increase maintenance costs, he explained. Subtle changes in the preheater process can eventually lead to blockage.

To reduce downtime and optimize maintenance, the plant sought a solution to predict and avoid cyclone blockages through early and accurate detection. Having the added benefit of RxM capabilities to resolve emerging problems quicker would minimize production losses.

The cement plant chose a PdM and RxM solution with AI and ML capabilities from AspenTech. Maestro for Aspen Mtell was used to build enhanced agents capable of predicting cyclone blockages. The failure agents trained on past blockages to identify the failure signatures that precede degradation, breakdowns, and process disruptions, and learned to recognize the pattern.

The multivariate (MV) models helped the plant to better understand the root cause of the condition and provided prescriptive guidance on feed composition and kiln operating conditions to the operators, enabling them to quickly bring the process back to normal operation. Within weeks, the solution was able to provide several days of early warning. “Eliminating just 25 percent of blockages generated over $1 million in savings,” said Williams.

Wärtsilä / Pega

Wärtsilä, a Finnish technology and service provider for energy and marine markets around the world, needed to improve its ability to process and analyze asset condition data received daily from thousands of equipment installations. An IoT engine upgrade was needed to help preempt performance issues with its customers’ power plants and vessels, and to automate feedback to each customer. Its legacy system could not sufficiently scale to handle the desired automation.

Many of Wärtsilä’s installations have multiple engines per site, and each engine has attached equipment and hundreds of sensors generating condition-based data. To better harness the global data, the company needed to map it to a normalized data structure for quick processing and apply a rules framework to assess the sensor data for anomalies. This would enable PdM practices to preempt equipment failures and also help to optimize service support.

Read “RxM: What is prescriptive maintenance, and how soon will you need it?”
The company worked previously with Pega on other projects and chose the Pega Platform to extend its existing IoT capabilities. The digital transformation platform for Digital Prescriptive Maintenance is designed to help optimize the cost and duty-cycle of devices and systems.

With the new IoT-powered solution, Wärtsilä now compiles all incoming sensor data in a data lake in the cloud, where it is normalized and fed into the Pega Platform. Using a complex rules framework, the data is processed to identify exceptions that may indicate a developing risk of engine failure or a reduction in equipment performance. Daily feedback reports document the findings and prescribe recommended actions, enabling the company to save time and money by avoiding unplanned downtime and production problems.

Benefits to the customers include improved environmental and economic performance and long-term predictability of their vessels or power plants. Wärtsilä is also benefitting from optimized equipment maintenance, improved reliability and performance, and increased sales of proactive maintenance contracts.

Each of these industry leaders is enjoying the rewards of using maintenance and reliability best practices. With the benefits of PdM and RxM being realized, the old ways of doing business seem antiquated and companies like these are not looking back.

This blogpost is originally from Aquantico | Challenge, innovate & deliver value

Reliability connect | BY Ken Madsen

Implementing IIoT-Based Asset Reliability Program at a Gold Mine

The wireless Predictive Maintenance program paid for itself within six months at a leading gold mining company. They demonstrated increases in asset reliability, personnel efficiency, and production as part of their digital transformation initiative.

STARTING WITH A DIGITAL FIRST MINDSET

One of the world’s largest gold mining companies in the world is undergoing a digital transformation to integrate data and technology throughout their operations. Part of the objective is to get the right information to the right people at the right time and to connect disparate operations. The digital transformation team identified an opportunity to get better machine health data, more often, in real time, by working with Petasense, an Industrial IoT startup.

With a digital first mindset, the mining customer implemented Petasense’s Asset Reliability & Optimization (ARO) system and wireless sensors to monitor several critical pumps, fans, SAG and Ball mills in one facility. They had an existing Predictive Maintenance (PdM) program in place for over 20 years that relied on walkaround vibration data collection, and they identified opportunities to reduce failures further by increasing the frequency of data collection.

The challenge with walkaround programs is problems that develop during the intervals can be missed. Even with an experienced team, several failures had developed quickly in between monthly and quarterly routes, causing unplanned downtime.

EFFICIENCY, INTEGRATION, AND RUGGEDNESS

The Petasense Vibration Motes (permanently installed wireless vibration sensors) collect data automatically, and the data is analyzed with advanced machine learning algorithms to predict equipment problems. Each machine is assigned an asset health score and provides real-time alerts when the score falls below an acceptable number. This allows the analyst to focus only on problem equipment, with access to all the sophisticated vibration analysis features that they are used to using.

Using Petasense Motes, the customer was able to instrument remote assets that are difficult to access for regular vibration readings. Another important factor for selection was ruggedness of the Motes (IP67 rated) in the dusty, harsh mining and mineral processing environment.

The customer has been using OSIsoft Pi data historian for many years, and they didn’t want another standalone system. They appreciated Petasense’s open architecture that easily integrates with Pi. Vibration readings are brought into Pi, and the team uses this to analyze process and reliability data simultaneously to make better decisions and optimize equipment performance.

DEMONSTRATING ROI FROM REAL-TIME MONITORING

Petasense’s implementation was completed within a few weeks. This included installation of the Motes, configuration of the ARO Cloud software and Pi integration. Soon after going live, the PdM team was alerted by high vibrations on the drive end of a Mercury Scrubber Fan. The high vibration generated real-time alerts which allowed the maintenance team to shut it down proactively and fix the problem.

After troubleshooting for fan imbalance and visually inspecting, the PdM team determined that there were loose frame bolts holding the motor and fan frame. The looseness had caused structural damage and broken the weld in frame.

The Mercury Scrubber Fan helps with recovery percentage, and a failure would mean shutting down the kiln. While it doesn’t shut down the plant, it can help gold recovery by 1-2%, and over the long term, this can cause millions of dollars in lost revenue. Before working with Petasense, the customer had experienced a failure on the same fan and it was offline for six weeks, waiting for the fan assembly to arrive and causing significant production losses.

Another win for the PdM team was high-flow centrifugal pumps in the heap leach field. The heap leaching operations are used to recover lower-grade oxide ore. They consist of applying a leaching solution to piles of crushed ore to extract the gold. The customer had increased the amount of gold extracted, to the point that both the primary and standby pumps were needed to get the flow rates and the tank was undersized for current demand.

With the Petasense Motes, they were able to diagnose cavitation in the pumps. A failure on these pumps lowers the gold recovery amount. Knowing this, the PdM team were able to identify the need for a design change to increase reliability.

EXPANDING THE PREDICTIVE MAINTENANCE PROGRAM

The early success of detecting failures helped justify expansion of the program to other rotating assets and sites, as well as multi-parametric monitoring. Machine monitoring is limited when just a singular parameter, such as vibration, is used. It’s most effective when combined with related parameters such as load or running speed. By installing the Petasense Transmitter, the customer was able to wirelessly collect data from hundreds of different types of sensors, including temperature, pressure, ultrasound, current, vibration, and more.

In addition to expanding the program to more types of data, the PdM team has identified other assets that are not monitored today. They are planning to install Motes on critical equipment like conveyors, which the PdM team cannot access with normal route-based data collection. Some of these are critical, non-redundant assets that will take down the production line, making them prime candidates for Petasense wireless sensors.

Using Petasense, the PdM team was able to demonstrate tangible results that aligned perfectly with their digital transformation. One of the success factors was the true partnership between the corporate digital transformation team and plant personnel responsible for implementing and using the information. They were able to successfully navigate the culture change by picking technologies that demonstrated fast Return on Investment to both the shop floor and the digital team.

This blogpost is originally from Aquantico | Challenge, innovate & deliver value