Reduce Carbon Footprint With Advanced Analytics-Driven Energy Management

When the low hanging fruit for energy savings has been picked and more knowledge is applied to improve operational performance, you get the added benefit of improving overall profitability and increased safety.

Mnet 126683 Carbon Footprint
Edwin van Dijk, VP of Marketing, TrendMinerEdwin van Dijk, VP of Marketing, TrendMiner

The discussion about climate changes has been taking place for many years and is still a hot topic, now more than ever before. This debate has led to global initiatives to reduce carbon footprints, which is high on the agenda of almost every country’s government. Regulations on a global, regional and local scale have been established to reduce greenhouse gas emissions, which heavily impacts the chemical industry. To achieve those goals and prove regulatory compliance, the industry is rapidly adopting the ISO 50001 standard to improve the organization’s energy performance and make climate part of their corporate strategy.

Reducing your carbon footprint also has an overall profitability benefit. Within the chemical industry, energy is often one of the largest components of the company's cost structure. Energy management for reducing costs is not new, but has become more important due to the imposed regulations. Most companies have formalized energy management programs and use automation and control technologies to help minimize energy costs. It is clear, however, that many companies need to take their efforts to the next level by monitoring and optimizing energy use in real time, and leveraging IIoT-generated data.

For many years process data has been captured in historians. All this data needs to be unlocked and leveraged for continuous improvement to lower the carbon footprint of the company. To some extent, data analytics have been utilized by large companies for their larger on-site energy issues. These time-consuming, centrally led data modeling projects are less suited for process-related optimization projects that require subject matter expertise. New tools put advanced analytics in the hands of subject matter experts such as process and field engineers. This allows them to handle 80% of energy related cases that contribute to the corporate goals for reducing the carbon footprint.

Energy Management 4.0

Global interest in Industry 4.0 has accelerated digital transformation in the process manufacturing industry, and specifically in the commodity and specialty chemicals industry. Many companies have engaged in technology pilots to explore options for reducing costs, increasing overall equipment effectiveness (OEE) or regulatory compliance. One of the best ways to leverage these new innovations is to apply advanced industrial analytics to production data generated by sensors. Every piece of data provides unique opportunities for improving energy efficiency.

In general, energy savings can be achieved in various ways: through change in daily behavior (switching of the light), through installations of more energy efficient equipment, through equipment maintenance, or through process optimization and ensuring the use within the best operating zones. Process and asset performance optimization is probably the biggest area for energy savings, but requires a deeper understanding of operational process and asset data (available in the historian).

(Source: DOE presentation “Introduction to the Superior Energy Performance Program” July 2016)(Source: DOE presentation “Introduction to the Superior Energy Performance Program” July 2016)

Analyze, Monitor and Predict W.A.G.E.S. consumption

The major process and asset related energy consumers include Water, Air, Gas, Electricity and Steam (WAGES) and can be directly or indirectly analyzed through all sensor data. Utilities and energy are often neglected at the plant level, however, since there are more pressing needs and analyzing WAGES inefficiencies is laborious. Looking for WAGES inefficiencies is time consuming because the utilities are used all across assets, plants and/or sites. For example, a plant could have hundreds of unit operations that require nitrogen or steam. For an individual to find the root cause of what caused an increase in overall steam usage to the plant would require looking at possibly hundreds of tags from their historian-like finding a needle in the haystack. It is rare for plant personnel to have the time available to deep dive something like this. Most of the time, no one would even look at the steam flow meter unless there was a major issue.

Subject matter experts such as process, operations and maintenance engineers have deep knowledge of the production process. They are in the best position to understand process behavior and assess root causes when using self-service advanced analytics tools, without the need to gather data manually, make a complex data model and validate the results. Through the self-service analytics tool, the data can be descriptively analyzed to determine what has happened. And if a long period of performance can be assessed, the performance can be better understood. Sometimes, certain issues happen only a couple of times per year but can have a big impact on energy consumption (a trip causing a shutdown). Discovery analytics helps understand what has happened and through diagnostic analytics, the organization can start monitoring the performance of the site.

Since asset performance is contextualized by the process they function in, the best operating zones or performance windows need to be extracted from actual process behavior rather than theoretical data. Based on the historical data, fingerprints with an energy consumption focus can be created to monitor good and bad behavior. Additionally, monitoring live operational performance can be used for predictive analytics e.g. performance downstream is caused by behavior an hour or more upstream.

Practical Use Cases

Using best-in-class self-service analytics is easier than making data models and it is much more efficient to process and, to stick with our previous example, compare hundreds of tags to see what caused a spike in the unit battery limit steam meter. The users can quickly get answer to questions such as:

  • "Is my steam/N2/plant air/demin water usage abnormal?"
  • "Is there anything that correlates with this abnormal usage right now?"
  • “How can I quickly figure out why my WAGES are abnormal so I can take timely corrective action?”

With traditional data modeling tools, these questions could take weeks to answer, which is why they are frequently neglected... no one has that kind of time. There are many instances where advanced self-service analytics has been successfully used to analyze, monitor and predict the process and asset performance of energy management.

One example is related to energy consumption within a cooling water network. A large number of reactors were consuming cooling capacity from the utility network for cooling water. Sufficient cooling capacity is critical for many of the reactors as thermal runaway could occur when the available capacity is insufficient. To avoid this undesirable situation, advanced analytics was set up to monitor the cooling capacity in real time. Early warnings were created and only triggered on actual problematic situations, avoiding false positive alarms that could be triggered by measurement noise or spikes in the data. Upon receiving a warning, there is ample time for the process engineer and operators to re-balance the reactors and de-prioritize other equipment so that the critical ones can consume the maximal cooling capacity and overall energy consumption is within target boundaries.

A second use case is related to a burner oven, which was suffering from many trips causing production loss and increased gas consumption. Through advanced process data analytics, multiple previously unknown root causes were found to be responsible for the trips. With this increased understanding of the process, monitors were created to alert the key stakeholders. The monitors allow them to take appropriate action when a specific cause of a trip occurs and thus avoid a trip from actually happening. The events are now also automatically annotated with the explanation of the root cause. This way, the organization is actively learning to control the process based on combining actual process behavior with subject matter expertise. With downtime reduction achieved, gas consumption has also decreased significantly, but more important the energy consumption can be improved continuously over time.

A third example is a predictive maintenance case for the fouling of heat exchangers. In a reactor with subsequent heating and cooling phases, the controlled cooling phase is the most time consuming. Fouling of heat exchangers increases the cooling time, but scheduling maintenance too early leads to unwarranted downtime and scheduling too late leads to degraded performance, increased energy consumption and potential risks. To enable timely maintenance, a cooling time monitor was set up, extending the asset availability, reducing the maintenance cost and safety risks. All of these benefits, including controlled energy consumption, ultimately led to 1%+ overall revenue increase of the production line.

Continuous Improvement 4.0

In general, finding and solving root causes for process deviations and anomalies results in a more energy efficient operation. Monitoring the live production performance allows for the control of various production parameters, including energy consumption. When the total energy consumption of a specific year is taken as a base line, the monitoring of performance against corporate goals becomes possible.

Energy consumption per production line for three consecutive years showing performance against the reference year.Energy consumption per production line for three consecutive years showing performance against the reference year.

Covestro, a manufacturer of high-tech polymer materials for many major industries, initiated three major energy savings projects for their polyether plant in Antwerp as a part of their energy-savings goals and ISO50001 directives. Self-service industrial analytics solutions were implemented for online detecting (including root cause analysis and hypothesis generation), logging and explaining unexpected energy consumption, and for comparing the results with the reference year 2013. Using specific formulas and calculated tags, various energy consumers are monitored and controlled. Through monitoring the performance against the reference year, it is shown that the energy consumption is effectively decreased year over year, meeting their corporate goals. More importantly, with a growing knowledge and insight into the production process, the company is continuously improving their overall performance.

Concluding thoughts

Energy management is not new; many companies have a structured energy management program in place. However, new self-service analytics tools now allow subject matter experts to analyze, monitor and predict process and asset performance, which can result in a huge contribution towards meeting organizational carbon footprint goals. And when the low hanging fruit for energy savings has been picked and more knowledge is applied to improve operational performance, you get the added benefit of improving overall profitability and increased safety.

“For us, energy efficiency is the key to combining climate protection, conservation of resources and competitive economic advantages.” - BASF Global Strategy

More in Energy