AspenTech Hybrid Modeling Helps Deliver a Performance Engineering Approach

Summary

To optimize their processes and their businesses, industrial organizations must address multiple dimensions and factors simultaneously. The most successful industrial companies bring together multiple disciplines and perspectives to optimize design and operations. Using a well-crafted engineering process and cross-disciplinary teamwork, industrial manufacturers can leverage an integrated, “big-picture” approach to engineering design and optimization.  This enables them to deliver not just a project on schedule or budget, but superior business performance, including reduced CAPEX and OPEX, faster time to market, greater energy efficiency, and higher profit margins.

Projects today must address capital and energy concerns; embrace safety; meet sustainability and environmental goals; and manage controllability and yields. Advances in digitalization and the confluence of digital engineering technology and daily work processes are helping organizations improve collaboration and integration to deliver comprehensive asset optimization strategies that yield significant financial returns.

In a recent briefing from AspenTech,  we learned that the company has identified best practices in engineering and is incorporating these into a machine learning- and artificial intelligence (AI)-enabled technology solution designed to support asset owners in their efforts to drive a performance engineering lifecycle approach.

To achieve the desired model accuracy, plant data is used to calibrate these first principles models to observe plant conditions and performance.  Since effective model calibration requires considerable process expertise, AI and machine learning are built into the AspenTech solutions and can help accelerate the ability to calibrate first principles models and create data-based models and processes quickly.  Ultimately, the company believes AI has the potential to lower the expertise bar needed to model process systems.  However, human domain expertise is still needed to create the real-world “guard- rails” that make the models work safely and reliably.

A Performance Engineering Lifecycle Approach

As AspenTech explained, performance engineering is about pushing the boundaries of existing concepts, designs, and asset constraints to create new, higher performing designs and operations. The company believes that using asset models consistently across the CAPEX and OPEX cycle multiplies the value delivered.  This optimized engineering work process with shared models can support conceptual engineering, front-end engineering, design (FEED) and the economic, safety, sustainability, and energy evaluation to enhance collaboration and drive performance improvements at all stages of plant design and operation.  The recently acquired OptiPlant solution provides AI-driven 3D conceptual layout capabilities to deliver an integrated FEED solution to help accelerate collaboration between engineering, estimating, and layout disciplines. The table on the next page summarizes characteristics of AspenTech’s Performance Engineering approach.

AspenTech’s Hybrid Modeling Approach

AspenTech’s hybrid models combine AI and engineering first principles to deliver accurate models more quickly and without requiring significant process or AI expertise.  Machine learning is used to create the model, leveraging simulation and plant/pilot plant data.  Domain knowledge, including first principles and engineering constraints, is used to enrich the model.  The goal for this next generation of AspenTech solutions is to democratize AI within hybrid models to optimally design, operate, and maintain assets.

AspenTech will be deploying hybrid modeling capabilities across its existing software suite through a model alliance approach.  This synchronizes fit-for purpose models in different functional areas needed to operate a given asset safely, reliably, sustainably, and profitably.  For example, the model alliance can  be used to create reduced order unit models in planning, dynamic optimization, and online equipment monitoring - all derived from the same root refining unit operating data set and simulation model to achieve closed-loop production optimization.  Other types of hybrid models include:

AI-driven Hybrid Models

Machine learning to create an empirical model based on plant or experimental data, augmented with first principles (e.g., thermodynamic properties etc.), constraints (e.g., mass balance) and domain knowledge.

Reduced Order Hybrid Models

Machine learning is used to create an empirical model based on data from numerous simulation runs.  This is augmented with constraints and domain expertise to build a fit-for-purpose, high-fidelity, and high-performant model that is accurate within the range for which it has been trained.

First Principles-driven Hybrid Models

This approach augments an existing first principles model with AI using data from operations to calculate unknown variables and relationships not captured by the original model. Machine learning determines the unknown value and its relationships to continuously calibrate the model as conditions change.

Conclusion

ARC Advisory Group believes AspenTech’s vision for hybrid modeling in the process industry will help democratize and expand use of accurate models across a wide variety of assets and companies.  The ability of machine learning and AI to accelerate model development and accuracy is a crucial step in understanding how a specific process will behave or respond to unexpected change.  As plants and their systems have increased in complexity, these models have become essential to operations.

For many industrial asset owners, AI is a highly customized, ”bolt-on” platform or data lake platform that consumes considerable resources to develop and maintain. AspenTech’s hybrid model solution combines AI and engineering first principles to deliver a comprehensive, accurate model more quickly, without requiring the user to have significant domain knowledge or AI expertise.  Machine learning is used to create the model leveraging simulation or plant data.  Domain knowledge, including first principles and engineering constraints, is used to enrich the model.  With this hybrid approach, users don’t need to have deep process expertise or become an AI expert.

With hybrid models, users can model processes and assets that cannot easily be modeled with first principles alone.  Combining the accuracy of empirical models with the strength of first principles models creates a more robust and predictive model.

AspenTech is uniquely positioned to leverage over 40 years’ domain expertise to make AI applicable to process industries, delivering industry-ready AI.  AspenTech brings together three fundamental capabilities:

• Strong and deep domain expertise in process industries

• Strong capabilities to capture and analyze the quantities of data available through the proliferation of connected sensors

• Innovative leadership in turning machine learning and AI into industrial solutions.

ARC Advisory Group clients can view the complete report at  ARC Client Portal

If you would like to buy this report or obtain information about how to become a client, please Contact Us

Keywords: AspenTech, Engineering Simulation, Modeling, AI, First Principles Models, ARC Advisory Group.