(147c) How Adding Machine Learning to Simulation Brings Greater Value for Process Industries

Machine learning (ML) had surges in popularity every decade before breaking the pattern to yield a permanent adoption of ML into numerous aspects of business, engineering, and society as a whole. Many factors, including but not limited to advances in parallel computing capabilities from GPUs to facilitate the inception of more sophisticated and capable ML algorithms, have supported this AI industrial revolution. Engineers today have the distinct opportunity to utilize these advancements in ML capabilities for our domain area of simulation and physical testing.

For the process industry, the digital tools for design and operation of equipment is ripe for supplementation with machine learning. Design engineers can benefit from the constant stream of innovations in the ML community when productizing new equipment, while operators of the equipment can be enhanced in productivity from on-demand information when these ML models are deployed for their easy access. There are several themes by which machine learning can cause a material impact: augmented user experiences, optimization, better utilization of physical measurement, and time savings via faster insights.

Utilization of a limited number of experiments and complimenting them with simulated data can result in significant time savings and valuable insights into engineering processes. Specifically, how simulation experts, a widely used engineering method, can realize large time savings when employing ML models in their sensitivity and design exploration, which can help reduce time to market. Further, how physical measurement of in-operation equipment can be combined with simulations with as input to (ML powered) executive digital twins (xDTs) to provide the operator with simulation-level detail of performance from limited discrete physical measurement points.

In this study, we perform multiple batches of computational fluid dynamics (CFD) simulations to characterize the performance of a mixing vessel with respect to changes in the geometric design. More significantly, we then explore a series of three independent scenarios whereby unique ML-based ROMs are created and analyzed. First, we conduct an optimization study in a hybrid modality: machine learning models are used when possible, during the optimization study in conjunction with traditional simulation for the design points. Second, we create and leverage ML-based ROMs for adjacent design characterization akin to typical robustness, reliability, and sensitivity analysis. Finally, we create a third ML-based ROM that can impute simulation-level detailed results from only discrete measurements taken while the equipment is in operation.

Our observation is that the impact of simulation-based, detail-dense ROMs, is under-demonstrated in industry and has tremendous value; operators can take typical measurements in their equipment and as such use it as inputs to the ML-based ROMs created. These ML-based ROMs will provide fully 3D results, full of information as if a simulation model was ran, from only a few discrete measurement points (such as a RPM or torque measurement, which are 0D in nature).

These models can be deployed in web-based application to bridge the designer to the operator. It will aid in responding to changes either due to operating conditions or input conditions at a rapid pace while maintaining quality and safety of chemical operations at quicker pace.