Our future with predictive maintenance

Late last year, US President Biden issued an executive order on safe, secure and trustworthy artificial intelligence (AI). Outside of the controversial AI usage in politics, social media, journalism and other consumer uses, the industrial application of machine learning (ML) and AI has already become a standard and essential tool.

In heavy industries, like cement, steel, energy and chemicals, the ‘digital revolution’ of capturing manufacturing data, and the rise of ML/AI analysis to enhance maintenance and process control is active and ongoing. Global AI in cement production alone is expected to grow at a compound annual growth rate of 28.5% over a forecast period of 2021 – 2026.

Industry 4.0 has cement manufacturers around the world experiencing exponential growth in data collection along with the ML/AI analysis that has appeared in recent years. “Expert consultations, available remotely, aid interpretation of the condition monitoring analysis in predictive maintenance in concert with the plant in-house experts”, says Scott Ziegler, CEO of CemAI. “Unplanned downtime is minimised, and repairs are made at early stages, prior to failure, reducing cost and the need for total replacement.”

CemAI’s cement industry specific solution for plant end-to-end predictive and prescriptive maintenance delivers on the technology’s ability to maximise equipment availability and production output. By using existing plants’ sensors, process flow diagrams, and historical data, CemAI first builds a plant digital model that incorporates process and equipment hierarchies and dependencies, establishing an operational baseline.

Next, the system applies analytics to the continuous data flow from the plant. Meaningful differences or abnormalities are detected, compared to historical operating conditions, which predict and direct maintenance interventions.

The ‘machine learning’ through the correlations and interactions of operational data obtained, continuously updates and improves the AI system and the plant maintenance plan alike.

Energy efficiency

Cement manufacturing is an energy-intensive process, both in fuel and electrical consumption. Facilities can achieve environmental gains through ML/AI solutions. The complexity of cement manufacturing lends itself to energy-saving benefits that can range from heat utilisation efficiencies, specific energy consumption improvements, and end-to-end process control improvements that can make environmentally beneficial changes while continuing to control their impact on product output and quality.

ML/AI – predictive and prescriptive maintenance solutions – like CemAI, are showing tangible results in energy efficiency. In 2022, Titan America’s Pennsuco plant in Medley, FL, moved beyond its initial successes with ML/AI. At this point in time, digitalisation was a movement in cement manufacturing for at least a decade.

As data became more available and could be adequately preserved in operating systems, the evolution of acting on real-time information began in earnest.

By utilising sensors and analogue signals to map the digital model for normalised plant operations, and then analysing in real-time deviations in temperature, pressure, vibration, lubrication, and other parameters, the AI models installed at Pennsuco provided the teams with powerful solutions to drive production outputs and equipment availability to new higher levels, while improving energy efficiency beyond original estimates. “Avoiding unnecessary cool down and reheat of equipment and end-to-end optimising of processes and machine function produces demonstrable gains”, says Ziegler.

Building on a few years of strong effort in digitalisation, Pennsuco achieved a new level of optimisation, combining the process and maintenance AI systems to achieve a new level of Overall Equipment Effectiveness (OEE), better process stability with higher alternative fuel (AF) rates and lower energy consumption.

AI-induced innovations at Pennsuco optimised operations in two critical areas:

• Upgrade of the existing maintenance and process control AI system with a new set of over 100 high resolution sensors installed to measure vibrations, temperature and their deviations, allowing for the AI systems to reduce idling and restarting time of major equipment and providing significant thermal and electrical energy savings. AI-based process control solutions were also updated, based on the earlier learning experience, for finish mills’ process control that allowed higher throughput of each mill and decrease in electrical energy consumption.
• AI solutions at the kiln contributed to significant stability improvements, so that despite higher AF consumption, Pennsuco delivered a new production record. Significant portions of the fuels are produced in its own processed engineered fuel (PEF) facility. This PEF facility can receive a wide array of waste streams and was already integrated with advanced technical controls that ensure consistent fuel properties. However, in 2022, it was enrolled in the maintenance digitalisation process to improve the reliability and efficiency.

Utilising CemAI and process optimisation software aid, Titan America’s Pennsuco plant’s electrical energy consumption declined by 6% and the operation doubled its AF consumption, saving landfill capacity, delivering on circularity, and lowering greenhouse gas emissions.

Enjoyed what you've read so far? Read the full article and the rest of the February Issue of World Cement by registering today for free!

Read the article online at: https://www.worldcement.com/the-americas/22022024/our-future-with-predictive-maintenance/