"The industry has been practicing digitization for a long time, and recent changes have made these applications more powerful, agile, portable, intelligent, and autonomous.The change in CPI is incremental. It is incremental due to the inherent responsibilities associated with operating an industrial plant." says Sergio Fernandes, Head of Chemical Markets, Yokogawa Electric Corporation, USA. That said, CPI companies have successfully deployed digital technologies at scale, driven in part by the shift from software running on users' laptops to high-performance applications and tools that can now be accessed from virtually anywhere.
"Cloud computing not only reduces CAPEX budgets, but also facilitates the availability of process models, whether steady-state or dynamic, regardless of the location of the end user," Fernandes explains. However, he cautions against simply assuming that digital plant models (no matter how advanced) will be completely accurate in terms of permanence. "Industrial processes are like living entities; they change over time. Any mathematical representation, such as a digital twin, will need to be adapted and will need to be updated through some mechanism. Beyond that, they will eventually be discarded. Assets need attention; they require budgets to maintain their sustainability." He added. Looking ahead, as more and more autonomous operations emerge, the pressing need for security and sustainability means that a smart balance must be struck when deploying human resources alongside digital assets. "Hazardous field operations, repetitive actions, routine activities, unnecessary trips in the field to collect data and inspections in hazardous areas can be intelligently addressed with current and upcoming technologies," Fernandes said, adding that considering cutting-edge digital technologies as key elements can enable an architecture that can inspire more human innovation. "This means continuous improvement of operations, anticipation of the next disruption and optimization of the entire value chain." Assets need attention; they require budgets to maintain their sustainability." He added.
Simply developing a software model that mimics a process or asset is not enough to truly leverage digitization, reiterated Rajesh Ramachandran, chief digital officer, ABB Industrial Automation). . "The trend now is towards industrial AI twins for digital factories. He looks at how to predict and optimize a set of results for a specific process scenario, which gives the opportunity to fine-tune different parameters." Ramachandran emphasizes that pure AI cannot be applied "as is" in industrial environments, and that corresponding domain expertise is essential to capture the complexity of CPI operations, such as the quality specifications of the final product or the presence of raw material impurities. expertise, culminating in a cognitive model built through its ABB Ability Genix software platform. "Genix builds what we call a cognitive model that is based on data from different systems such as maintenance, instrumentation and laboratories. This means it helps make more accurate optimization predictions." Ramachandran added. Citing industry studies that show that, on average, a plant may only use about 27 percent of its production data, while engineers may spend as much as 80 percent of their time aggregating data, he predicted that advanced software platforms will help mitigate these imbalances. . He says: "We are fundamentally addressing areas where we need to unlock the value in unused data and apply industrial AI at scale for maximum productivity and operational gains, while also simplifying data integration."
Intelligent platforms
There's no doubt that industrial software platforms have become more powerful in recent years as more companies adopt industrial AI and machine learning (ML) use cases. "At CPI, these types of technologies are being integrated everywhere from asset monitoring to AI-powered drones than can inspect torch towers," explains Michael Tworzydlo, product manager for analytics and machine learning at Emerson (St. Louis, Missouri; ). But Tworzydlo warns against over-hyping the value of these solutions without having to recognize the importance of the underlying engineering principles. "As a chemical engineer, the basics of analytics are the best place to start, beginning with a principles-based analysis, such as one based on how a heat exchanger works. The organization can then evolve to a data-driven approach using AI or ML to deal with more complex processes or plant-wide problems." He adds.
" AI offers powerful capabilities for CPI, but some companies struggle to apply it effectively to manufacturing challenges," explains Paige Morse, chemical industry director at Aspen Technology, Inc.
In response, AspenTech has begun embedding AI into its software platform, which makes it more accessible to a wider range of users.Morse notes that combining first-principles engineering with AI and domain expertise can help users better find solutions to complex problems that must be solved in CPI.AspenTech's hybrid modeling approach not only helps to optimize processes, but also enables engineers to create custom soft sensors, design new devices and integrate asset-wide processes. "Engineers can now use ML to build rich models faster to leverage simulation or plant data, adding domain expertise, engineering principles and design constraints without deep process or AI expertise." Morse said. With many CPI companies facing legitimate skills gaps.
In addition to overcoming labor gaps, sustainability initiatives are another area where CPI companies are increasingly concentrating their efforts. "Cost savings have driven much of the digitization effort, but companies are increasingly focused on waste and emissions from production units, as well as efficiency and reliability improvements," Morse says. She added: "Process simulation helps develop new products to meet the technical challenges of the circular economy, such as molecular recycling and new plastics design, and with the help of AI this activity is even faster."
This predictive capability is increasingly valuable in achieving specific sustainability goals, such as reducing air pollution through predictive emissions monitoring systems (PEMS), a function of Emerson's Plantweb Optics Analytics platform, which deploys ML and AI control systems through digital twins and distributed deployment. "As part of Plantweb Optics Analytics, we can deploy PEMS to monitor and estimate emissions using models and ML to dynamically optimize production. With PEMS, we can build models based on process variables that have been captured and use them to estimate and ultimately reduce emissions." Tworzydlo said.
The software's commitment to a sustainability strategy goes beyond emission reductions. "The rise of sustainable products and technologies that reuse or recycle waste is a growth area for the process simulation industry, presenting new problems and new opportunities. Recent growth areas include process simulation of hemp derivatives (e.g., CBD) and improved controls to reduce emissions from renewable energy sources. More established growth areas for the chemical industry include biofuels, methane recovery, CO2 recovery and solvent selection." says David Hill, Technical Support Manager at Chemstations Inc.(Houston, TX).
Hill believes that the prospects for process simulators can be further enhanced by creating product alliances with ancillary tools in CPI. Engineers who don't use process simulators often have tools that can be improved by connecting to a process simulator. In the safety, process control and energy sectors, there are many opportunities to combine industry-specific tools with the first principles of process simulators." Hill explains. Hill believes the drivers of this shift will include greenhouse gas reduction, energy efficiency, optimization based on thermodynamic models, improved safety and opportunities for advanced process control using rigorous simulation.
Extended Reality
In addition to AI and ML, augmented reality (AR) and virtual reality (VR) software platforms are also on the rise in industrial plants-and are no longer seeing this type of technology as a "luxury" item, more useful than ever before due to the increasing demand for remote work during pandemics. With fewer people in plants due to pandemics, plants are embracing new technologies. ar can overlay digital information into the real world, which helps better equip workers to perform tasks more accurately and with greater ease." Emerson's Tworzydlo said. As for the future of AI, ML and AR in industrial software, the use cases will certainly continue to expand. "There is still a huge amount of untapped potential. Eventually, we will start to target certain processes for autonomous operations.
Aveva Group plc (Cambridge, UK;) has bundled AR and VR concepts into its Extended Reality (XR) platform, and one particularly relevant application is personnel training. " The XR immersive training system allows companies to capture and retain operational knowledge when replacing retiring experienced operators, which is critical to plant safety and performance. This behavioral training can be applied not only to front-line operators, but also to engineers, technicians and emergency responders." Ravi Gopinath, Aveva's Chief Cloud Officer and Chief Product Officer, explains.
In one example, an operator training program developed by Aveva and Shell (The Hague, The Netherlands;) focuses on behavioral training to improve safety competencies.Gopinath says, "With this behavioral approach, an operator can be trained and evaluated on how he or she performs when faced with an accidental or episodic situation in the plant. " In another project, Aveva and Adnoc (Abu Dhabi;) created a real-time data visualization center that brings together more than 120 dashboards and 200,000 data points on a giant interactive screen.
Training is only part of XR's potential. AR tablet-based applications have been used to support field staff. Using AR to connect VR models in the tablets with real-time information and guided processes allows for better execution of work, thus avoiding costly breakdowns and reducing downtime. Looking ahead, Aveva believes that XR software can dramatically improve facility design and capital project engineering by automating the importation of traditional 3D plant models used during the design phase into an immersive environment. The conversion to VR will allow you to review and improve ergonomic designs even before you purchase any equipment. Virtual factories can exist entirely in the cloud, allowing collaboration between engineers located in different
offices or even on different continents.
With the development of data capture and analysis capabilities in software platforms, powerful analytical tools have become self-service, scalable decision-making tools for chemical engineers, who can build their own functionality into them to meet specific process needs. with such democratized tools, engineers can leverage data from different sources - for example, batch quality, etc. - to improve the quality of their processes, says Edwin van Dijk, vice president of marketing at TrendMiner NV. data from different sources-for example, lab information such as batch quality can be linked to process data with maintenance data. ). "The goal of democratizing analytics is to make actionable insights available to every operator, from the control room to the boardroom, to make data-driven decisions. By allowing users to create their own dashboards based on fingerprints, monitors and contextual views, this goes beyond traditional dashboard tools." Van Dijk adds. With pattern recognition, engineers can investigate operational performance and use good operational behavior for process monitoring. In addition, they can create their own "soft" sensors to monitor what physical sensors cannot measure, such as product quality specifications.
One data analytics success story reported by TrendMiner involved a chemical plant that was experiencing "sticky" valves, which caused delays between changes in valve output and actual process response. The plant wanted to accurately identify when the valves began to stick, so they needed to monitor any deviations from the expected behavior of the valves and then find parameters that would differentiate between "normal" and "bad" valve behavior periods. These parameters are converted into alerts for out-of-phase behavior, which not only notifies personnel of the situation, but also suggests possible corrective actions. "By using a self-service analytics solution, process experts are able to use embedded AI and ML capabilities to search for and validate production issues using high-speed trend analysis.
Even with the vast array of available software tools and mobile apps to choose from, some users still require highly customized solutions to meet their business needs. This is where in-house programming can come in handy.JourneyApps (Denver, Colo.;) offers a high-productivity application development platform that users can use to write their own code, resulting in more sophisticated applications than non-coded application builders, which are targeted at non-programmers and limited by their simplicity.JourneyApps CEO Conrad Hofmeyr explains, "This means that advanced business logic, engineering calculations, and highly customized integrations can be implemented in a matter of days without much of the traditional overhead associated with software development." He notes that most chemical engineers have some basic coding or scripting experience through tools like Microsoft Excel Macros or Matlab, so they can quickly acquire the programming skills necessary to use JourneyApps to build complex applications that automate and streamline critical business functions.
For example, Hofmeyr cites an example of a CPI company that built a dedicated application for Standard Operating Procedures (SOPs), enabling them to move from manual spreadsheet-based SOPs to a centrally-controlled application with a complete audit trail. He adds, " The customizability offered by JourneyApps means that individual global applications can be tailored for local needs and system integration." In another example, an oilfield chemical manufacturer developed its own application to run key calculations used in its daily field reporting process and generate report documentation, all while users were offline on an offline site.
Looking ahead to end use
Advanced software and modeling tools are also enabling the creation of safer, more efficient end products in many industries, from automotive parts to pharmaceuticals. One example is BASF SE's (Ludwigshafen, Germany) Ultrasim computer-aided engineering (CAE) tool for modeling material properties, which was recently updated to model a range of thermoplastic elastomer materials from initial processing through the entire processing chain. end-use products. Shorter development cycles and aggressive schedules are putting increasing pressure on engineers to get product performance right the first time. Predictive accuracy is a huge advantage," says Marios Lambi, CAE team leader for simulation engineering at BASF in North America.Ultrasim can simulate initial and cyclic loading of components, which has proven to be particularly important for automotive parts made from elastomeric materials. "From creep loads to crash simulations, thermal loads and vibration behavior, together with processing simulations describing process-induced material properties, as well as numerical optimization tools that allow for rapid geometry changes, Ultrasim lays the foundation for designing better parts," emphasizes Andreas Wüst, BASF Europe's Dynamic Structural Analysis Team Leader.
"The material characterization process generates the necessary data that is essential for the accuracy of predicting the behavior of real parts. Theoretical material models developed for this purpose are being calibrated using information from tests, thus ensuring that the behavior represents real manufacturing conditions and not an arbitrary situation that is far from reality." "There are many examples of complex assemblies, such as automotive seats, that have been crash tested, and these tests utilize Ultrasim's predictive accuracy to create parts that pass validation tests. This dramatically shortens the development cycle and minimizes, if not eliminates, design changes," he added.
For high-precision processes in R&D and quality analysis labs for biopharmaceutical ingredients and other high-value products, the software tool can be used for a variety of purposes, including facilitating an organization's business continuity plan (BCP). "Efficient software can mitigate or reduce the number of risks during laboratory inspections, simplify event testing, and automated procedures can be used to restore systems after an event or even keep them running during an event, all of which simplify BCP," said Thermo Fisher Scientific's Enterprise Chromeleon Data Systems Organization chromatography software product marketing manager Barbara van Cann said. In addition, labs can further simplify BCP by choosing integrated software that includes a chromatography data system (CDS), a laboratory information management system (LIMS), and a laboratory execution system (LES). van Cann explains, " Both the LIMS and CDS software should provide the tools to monitor instrument qualification, calibration, and maintenance, even for individual parts." CDS software should also help users deal with analytical irregularities and have built-in network failsafe features to ensure that operation can continue without human intervention in the event of a network outage. To avoid disruptions due to cybersecurity attacks, Van Cann recommends that labs run CDS and other software in a domain separate from the main office system to avoid potential cyber threats from e-mail. Finally, as with any automated software platform, the human factor must be considered. "Human error can be minimized by controlling what users can and cannot do, and what they can and cannot access. In addition, tools should be available to automate as many actions as possible. Less user interaction equals less error." She added.