Maintenance is one area of business in which what’s possible is being transformed by the Internet of Things (IoT) and Industry 4.0 technologies.
A combination of developments are working together to strip away traditional barriers of cost and scalability and bring wide-ranging predictive maintenance within the grasp of many process operators for the first time.
The buzz is that everyone can look forward to slashing unplanned downtime and maximising overall equipment effectiveness as a result.
It sounds great, yet this ‘big picture, high-concept’ narrative leaves many process operators unsure about how to begin their own digital revolution.
At a basic level you probably already have sensors in your system anyway. Even a non-critical system will have, say, pressure gauges or temperature sensors
Jake Mitchell, product manager for hydraulics, Bosch Rexroth
The bad news is that they can’t afford to ignore what’s happening for too long if they want to remain competitive. The good news is that specialist predictive maintenance and software companies are coming up with solutions designed to make life much easier for digital novices that want to enjoy the benefits of condition monitoring and predictive maintenance.
And beyond these IoT specialists, many of the same equipment and component suppliers that process operators are already familiar with are also working hard to make it easier to get going.
In many cases, the first step will be to get a better view of what’s happening around the plant floor by deploying a network of sensors and a system to bring the resulting data together. Only once the plant data is being effectively curated can operators take the next step and deploy a smart system to analyse the data and recommend when to take the most effective breakdown-busting maintenance measures.
Building on legacy
Quality new industrial equipment will probably have the necessary sensors built in, but what about legacy machinery? Even older equipment is likely to be instrumented in some way and users can build on that, says Jake Mitchell, product manager for hydraulics with Bosch Rexroth.
“At a basic level you probably already have sensors in your system anyway. Even a non-critical system will have, say, pressure gauges or temperature sensors,” he says.
A data gathering tool such as Rexroth’s IoT Gateway can take what you have and bring this data together in one place, making it possible for a single, desk-based operator to monitor multiple machines and identify trends.
The next step is to elevate this basic condition monitoring approach with a machine learning system that can issue warnings and instructions to optimise maintenance. In the case of Rexroth, that’s the company’s ODiN software.
“ODiN takes data from a system over a two-week period and will look at the performance of the machine and the times where you rest and hours you operate per day. It will use that to predict that within ‘x’ amount of time you will need to repair the pump or you may have a problem. It will also send a message, if you want, to an operator, or to a PLC. It’s the brains that will interpret the information in order to give you warnings and time to prepare.
“So condition monitoring is what the IoT Gateway does, but ODiN provides actual predictive maintenance because it predicts when you need to do things.”
[Without IoT monitoring technology] every interaction is a single point in time, whether that’s a phone call or a site visit
Paul McKeithan, head of digital services, Buhler Aeroglide
As a provider of the kind of hardware that can typically be found all over a process plant, such as hydraulic systems, drives and controls and linear motion technologies, a company such as Rexroth is well-placed to help process operators monitor the condition of a wide range of assets in a way that’s not too dissimilar to dedicated condition monitoring providers.
However, many hardware providers that focus on more specific unit operations are also looking to support operators as they digitise their processes.
For example, Buhler Aeroglide manufactures hot air-based ovens for drying and cooking food such as nuts and extruded snacks. Head of digital services, Paul McKeithan, says that the ability to monitor processes remotely and in real time is enabling the company to offer a game-changing level of support compared with the traditional approach: “[Without IoT monitoring technology] every interaction is a single point in time, whether that’s a phone call or a site visit. Now with having the process data on the Cloud and being able to share that data constantly it’s much more intimate.”
In a similar way, bearing company Schaeffler’s SmartCheck monitoring system can give an early warning when bearings begin to deteriorate.
“As the machine’s condition is being monitored continuously and autonomously, a change in the asset’s condition is recognised immediately at any given time. This is a large step forward in comparison to manual single measurements that only offer ‘snapshot’ information of the current machine condition,” confirms Martin Wolf, Industry 4.0 portfolio and sector solutions manager for Schaeffler. “Lead time prior to a machine failure can date back as far as six months for certain applications.”
SmartCheck was recently implemented successfully at a French steel mill owned by ArcelorMittal, where it monitors the deflection rolls that support the running and tensioning of steel belt. By spotting problems early and preventing unplanned furnace stoppages through improved maintenance scheduling, the implementation is saving ArcelorMittal an estimated €179,000.
In the same way that the level of intimacy is increasing between process operators and trusted suppliers, the component companies further up the supply chain are working more closely with OEMs to improve the reliability of major assets.
An ongoing project between ZF Friedrichshafen (ZF), which supplies wind turbine gearboxes, and Schaeffler is a great example of this. Since September last year, the companies have been collaborating to develop new solutions to predict the operating life of turbine gearbox components based on the actual loads that occur during operation.
The project uses ZF’s software solution for wind turbine gearboxes as part of a smart system to combine the expertise of a wider range of specialists in a single platform and therefore provide wind farm operators with an aggregated overview of each gearbox.
Schaeffler is a preferred partner for rolling bearings here and supplies rolling bearing load analyses, while ZF itself assesses the loads placed on the gearbox components. The project is designed to be open for the involvement of further companies such as lubricant experts and control system suppliers.
“Wind turbine technology has made continuous improvements in ensuring reliable performance of the power trains, also under harsh operating conditions, and of lowering the levelised cost of energy (LCOE),” says Wolf. “Further improvements can be made on the basis of an intensive collaboration amongst component suppliers. Using their expertise to analyze in-service operational data and perform predictive analytics will give operators and OEMs better methods to improve the total cost of ownership (TCO).”
In the solution that is currently installed, pre-processed data from the condition monitoring system and other sensors are transmitted to the ZF Cloud, while torque and speed data are forwarded to the Schaeffler cloud, where a detailed simulation model of the ZF gearbox has been implemented as a virtual twin. The calculation results from the virtual twin are transmitted back to the ZF Cloud and are then available on the ZF software’s dashboard for use in monitoring the gearbox’s condition.
“Schaeffler and ZF are long-term development partners in different fields and applications and also for digital solutions. Due to the fact that bearings are a key component within wind drivetrains, the Schaeffler domain knowhow can help the customer to lowering the LCOE,” says Wolf.
“Moreover, the data access will help Schaeffler to further improve its simulation and modelling capabilities, offer specific product recommendations in the spare part business, better forecast internal production, and influence product modifications and the development of new products.”