Modeling & Data Analytics in Process Optimization

Session Chairs:

• Thomas Froese, atlan-tec Systems GmbH

Session Description:

Modelling and Digital Twins are core technologies for understanding and optimizing processes. There are three basic ways to create models: White Box, Grey Box and Black Box. The first is based on data, the second is hybrid and the third is based on explicit process knowledge. The session introduces the three ways and their advantages and disadvantages.

*All session and speaker information is subject to change pending finalization

Confirmed Speakers:

• Norbert Jung, AVEVA
• Heiko Kulinna, BASF
• Andreas Krüger, atlan-tec Systems

Abstracts:

The Role of Process Engineering in the Digital Transformation

Norbert Jung, AVEVA

While Digital Transformation hits every corner of industry, the process engineering discipline has been largely excluded from this trend so far. PwC foresee that for process licensors / EPCs, data driven digital services can displace process technology as a differentiator1).

The objective of this presentation is to provide an overview of the obstacles to Digital Transformation for the process discipline and explain how these can be overcome. A special emphasis will be given on the role of the process simulation tool (white box modelling) as a catalyst for transformational change.

The presentation will examine several challenges specific to process engineering:

1)  According to DECHEMA, one major obstacle for digital transformation is the division of process simulators into single-purpose point solutions. Today, separate models may be created – often in different tools – for process design, control strategy design, operator training simulation, performance monitoring and online optimization. This drives up the total cost of modelling to a degree than can become prohibitively expensive. One single model should be able to cover the entire plant lifecycle from idea to operations2.

2)  Engineering disciplines, other than process, collaborate closely with database-driven tools and assured workflows. Process simulators are typically poorly integrated into this workflow, and if so, with a single directional information flow. This lack of integration makes it nearly impossible to break down silos and enable new ways of working, like the agile development methods that have transformed software engineering.

3)  Over the decades, legacy process simulators got overloaded with niche features and functions only usable by experts. This bloat makes them not suitable for automation and operations.

4)  The potential benefits of Artificial Intelligence and Machine Learning for process engineering are not widely understood. Appropriate black box tools are not deeply embedded within white box tools.

More recently, stakeholders see the Digital Twin as the most important building block for Digital Transformation of the process industries. The process discipline must be a central part to this transformation representing the process behavior of the Digital Twin3).

While legacy simulators are well-suited to accurately simulate processes, their decades-old architectures mean they are not ideal to serve the entire plant lifecycle and support the Digital Transformation. We will use the AVEVA SimCentral Simulation Platform4) as an example of how the identified obstacles can be overcome with a next generation process simulator. Case examples from leading EPC and Operating companies will be outlined and will serve as a way to explain the related benefits.

1) Digital business models in plant engineering and construction in an international comparison, A benchmarking study of PwC and VDMA, May 2019 (link)

2) DECHEMA Whitepaper Digitalisierung in der Chemieindustrie, September 2016 (link)

3) DECHEMA Tutzingen Thesen 2018 (link in english)

4) AVEVA SimCentral Simulation Platform (link)

Grey Box Modelling – The optimal way to optimize your Process

Dr. Heiko Kulinna, BASF Lampertheim GmbH, Lampertheim, Germany

The producing Industry creates new valuable Products from raw materials, beside it generates a lot of Data. These Data combined with Process Knowledge and Laboratory Data give us the possibility to improve our process efficient to new limits.

The Grey-Box Approach provides us with a very effective and efficient method to improve a process or a process step.

We will have a closer look on a few examples in detail:

• Drying processes are often cycletime bottlenecks. The combination of a few production data and some knowledge about physical-chemical relationships give us the tool to speed up this type of drying equipment.
• In a complex 2-step hydrogenation process a mixed catalyst is used. The combination of Laboratory Data, Kinetic Measurements and Process Data give us all information for a stepwise process improvement in several relevant aspects.

The process gets step by step very close to the optimum for throughput, quality and cost efficiency.