Motion Control Resources
- This technical feature is filed under:
Motion Control Trends to Watch in 2021
by Kristin Lewotsky, Contributing Editor
Motion Control & Motor Association Posted 11/16/2020
Editor’s note: for more details on the trends below, plus additional motion control insights, register for our free webinar, Trends to Watch in Motion Control.
Industrial end-users face challenges on all sides. Customers expect more performance at a lower price point. The growing demand for personalization means that mass production has been replaced by the quest for equipment and software that will support fast changeovers and production flexibility leading to the elusive batch of one. Meanwhile, the competitive market forced many industries to operate on razor thin margins. And that was before the coronavirus sent shockwaves through the global economy.
According to the World Economic Forum, US GDP alone dropped 5% in Q1 2020 and 32.9% in Q2. The motion control sector and the industries it supports have fared quite a bit better. However. Alex Shikany, Vice President of Membership & Business Intelligence for the Association for Advancing Automation stated that orders reported by MCMA members for Q2 2020 were $828 million, down 6.7% year-over-year from Q2 2019, and shipments dropped 5%. Orders for the first half of 2020 were down year-over-year by $1.67 billion, or 5.8%.
Nobody will deny the challenging market conditions we face. That said, there are reasons for cautious optimism in manufacturing, as this Special A3 Report shows. Meanwhile, the technology trends identified in this article present OEMs, integrators, and end-users with tools to tackle these challenges head-on.
Industry 4.0, smart factories, cyber physical systems… In our era of software-defined everything, OEMs and end-users are turning to digital technologies to not just survive the challenges of the market but thrive. The trend toward digitization – converting analog, mechanical, and even manual inputs into digital format – has been underway for decades. Digitalization – using digital data to run and improve processes – is gaining momentum across industries as more and more companies seek to boost productivity and product quality while cutting unscheduled downtime and operating costs.
Digitalization leverages technologies like the IIoT, edge computing, and 5G to support use cases such as predictive maintenance, digital twins, and augmented reality. Here, we will take a closer look at each example.
The Industrial Internet of Things (IIoT)
End users across industries all want better visibility into status and performance of their equipment, the productivity of their operation, and the quality of the products they turn out. The industrial Internet of things (IIoT) can satisfy these needs. Properly executed, an IIoT strategy can lead to significant improvements in uptime and productivity while reducing costs.
Strictly speaking, the IIoT is a fabric of sensors and smart components that are networked together to exchange data, often in near real time. When people talk about the IIoT, however, it is in the context of the ways data from those industrial things can be analyzed to provide insights into the performance, condition, and throughput of the equipment. The data can be aggregated in some centralized platform for analytics and visualization. Use cases include online equipment health monitoring, energy analytics, process control, quality assurance, and more.
The devices of the IIoT can send data to the cloud for analysis by artificial intelligence (AI) or machine learning (ML) applications. A dashboard from an IIoT application might present key performance indicators (KPIs) or condition data from temperature sensors, current monitors, or vibration monitors that can be analyzed to detect developing defects.
The connectivity of the IIoT can make it possible for any permissioned user to access data from anywhere. Managers can monitor equipment from their desk or their den. OEMs and maintenance staff can troubleshoot issues from across the facility, across town, or across the country (see figure 1). The visibility enables best practices to be propagated across an organization, from shift to shift, facility to facility, or country to country.
Although the IIoT offers many benefits, it also has some issues. The first challenge results from the sheer volume of data generated by all of the industrial things. Sending that data to the cloud (whether a private enterprise cloud or a public cloud) requires an enormous amount of bandwidth. The demand is exacerbated by the fact that many sensors for the most part report steady-state operation. Sending all of the data captured can overwhelm a network, in some cases leading to the need for an infrastructure upgrade.
Second, sending data to the cloud introduces latency. While that is not a major issue for a slowly developing defect in a condition monitoring application, a performance issue in a high-speed production line can lead to a significant volume of scrap (and losses) in a very short time.
Finally, benefits of the IIoT aside, industrial organizations have a great deal of concern over data breaches and security vulnerabilities. In many cases, and users are not just reluctant to allow controls and upgrades from external sources, but are even unwilling to expose data for viewing and analytics. Edge computing provides a solution for all three of these concerns.
In edge computing, data acquisition, analysis, and storage take place inside the “edge” of the enterprise network, before the data is passed to the Internet. Performing the analysis as close as possible to the devices generating the data minimizes bandwidth demand and latency. The level of sophistication varies from implementation to implementation. In some cases, edge computing just involves preprocessing steps like filtering and normalizing to minimize the volume of data passing through the network. In other cases, the edge devices themselves are capable of running AI and ML applications to permit not just visualization and analytics but also automated alerts and responses.
Keeping processes within the enterprise network dramatically decreases security vulnerabilities, enabling end-users to feel comfortable not only viewing data but allowing machine control and updates.
Edge computing is best applied to applications that are highly time sensitive, involve large amounts of data, or both. Carsten Baumann, director of strategic initiatives & solution architect for Schneider Electric’s Secure Power division points to a beverage customer that applies edge processing for quality assurance and process control as a good example. The organization uses machine vision to monitor the performance of the motion axes, checking the fill levels, the label position, etc. Initially, the company performed analytics in the cloud. Streaming the data cost money in terms of bandwidth. Sending the data off premises presented a security risk. Most important, the cloud-based solution introduced latency. By the time an issue was detected, multiple substandard products would have been produced.
“The process took way too long,” says Baumann. “They would have had to slow down the manufacturing process to catch up with the latency they experienced from performing the analytics in the cloud.” After analyzing the situation, the company determined that processing needed to take place directly on the shop floor using edge computing. “[With an edge computing solution], they were able to reduce latency to below one second. As a result, they were able to run the manufacturing process exactly at the speed as it was designed and optimized for without the data analytics impairing the manufacturing process itself.”
Intended to help prevent unscheduled downtime and minimize operating costs, predictive maintenance combines current and historical data from sensors and smart components with advanced modeling and analytics to identify issues well before they go critical. Sudden temperature increases or spikes in the vibration spectrum of a motor can indicate a developing bearing defect, for example (see figure 2). Predictive maintenance via remote monitoring not only preempts catastrophic failure, it also gives companies the opportunity to order parts and schedule maintenance at a time that is least disruptive to the schedule.
According to the US Department of Energy, the average savings for a predictive maintenance program is:
- 25% to 30% reduction in maintenance costs
- 70% to 75% elimination of failures
- 35% to 45% reduction in downtime
- 20%- to 25% increase in production
Predictive maintenance is far from new. Over the past decade or so, there has been a steadily, albeit slowly increasing interest in the technology, particularly from OEMs looking to streamline fleet management. The trend line changed abruptly with the advent of the COVID-19 pandemic, however. Suddenly, traveling to the customer to troubleshoot and repair faulty equipment was no longer an easy option. Engineering and reliability technicians needed real-time visibility and access from anywhere. Predictive maintenance, courtesy of the IIoT, not only captures the data but makes it broadly available to off-site experts throughout the food chain.
Simulation, Digital Twins, and Virtual Reality
Simulation plays an increasingly important role in the design, commissioning, operation, and maintenance of automated equipment. The days of designing a machine on paper and then discovering problems in the prototype phase are fast disappearing. Detailed simulations of machines enable virtual prototyping, commissioning, and maintenance. This is particularly effective in the case of the digital twin.
A digital twin consists of a detailed model of an actual physical device, constantly updated using sensor data from the real physical asset. Over time, the digital twin becomes a highly accurate representation of the equipment. It can be applied at the machine level, the production line level, or even the factory level. It can be used for virtual commissioning, to monitor component wear, troubleshoot issues, try out new control algorithms, etc.
Simulation also can be used to provide a virtual reality experience for staff such as operators and maintenance. Modeling enables users to learn new tasks or experiment with techniques while seeing an accurate representation of machine response. They can learn how to respond to emergency conditions such as fires and test their understanding after.
Manufacturing and other industries that use motion control are facing a skills gap. The “machine whisperers” of the baby boom generation are retiring, taking their decades of knowledge and experience with them. The connected worker leverages augmented reality to help plug that skills gap, speeding training, providing easy connection to off-site expertise, and offering rapid, mobile access to information that used to be only available from paper manuals.
Augmented reality supplements the real physical scene with additional information. If an operator needs to perform an unfamiliar process, an augmented reality display could show a sequence of steps or even an animation overlaying the actual scene as they go through the process. A maintenance technician looking at the housing of the gearbox might be shown a rendering of the components inside, as well as a list of operating parameters like bearing temperature, oil viscosity, etc. (see figure 3). The system would deliver the information and the appropriate response directly to the display so that staff no longer needs to guess.
Once again, pandemic limitations like social distancing, limited staffing, and even layoffs have dramatically increased the uptake of augmented reality. If an issue occurs when a skeleton staff is on-site and the sustaining engineer is at home, augmented reality would not only enable on-site staff to get additional information on the issue, the off-site expert could add arrows and circles to the image to show features of interest to help with the explanation of the repair.
“Before Covid, [augmented reality] was always a great concept,” says Craig Resnick, VP at ARC Advisory Group (Dedham, Massachusetts). “Companies would say, ‘We’ll get there some day but we can only spend money where we have a ROI of 12 months or less.’” Then, the pandemic brought shutdowns, social distancing, and staffing cuts, forcing many companies to reconsider their positions, not only adopting augmented reality but retaining it going forward. “Now, they’re thinking, ‘Can we guarantee that this is never going to happen again?’ Of course, the answer to that is no.”
It’s almost impossible to turn on the television or visit a news or entertainment website without seeing an ad for 5G cellular service. In reality, the biggest impact of 5G will not be driven by its ability to download a movie in a few seconds but by its ability to support massively parallel machine-to-machine communications with high reliability and millisecond latency. Tailor-made for the IIoT, 5G will support use cases like reconfigurable factory layouts and connected worker applications such as augmented reality.
3G and 4G LTE networks operate at the lower-frequency end of the RF spectrum (600 MHz to 2 GHz). This region requires high towers and larger antennae to generate radio waves. The signals are moderately fast and penetrate buildings well but the latencies are insufficient for most safety-critical applications. 5G networks operate in the mid-band (2 GHz to 10 GHz) and the high band (10 GHz to 37 GHz). Higher frequencies mean higher speeds, lower latencies, and smaller towers. The latter characteristic makes it easy for companies to establish private 5G networks such as plant-level networks, helping address concerns like privacy and security.
Wireless networking is still something of a hard sell in the industrial space. Expect 5G to change that.
There is no guarantee when the COVID-19 pandemic will be under control. One thing is certain, however: our interconnected economy will have other threats and other uncertainties. The technologies discussed here provide tools for OEMs, integrators, and end users to tackle current and future challenges. For more details and additional trends, don’t miss Trends to Watch in Motion Control.