A platform for research: civil engineering, architecture and urbanism
A platform for research: civil engineering, architecture and urbanism
Frankenstein’s ROMster: Avoiding pitfalls of reduced-order model development
Highlights Reduced-order models (ROMs) are a powerful approach to reducing the complexity of physics-based numerical simulations. We show that a traditional ROM can display excellent overall statistics and yet have poor predictive power. We introduce a novel approach where a set of sub-ROMs are generated to overcome potential pitfalls in traditional ROMs. The effectiveness of the new approach—the ROMster framework—is demonstrated using injection rates for CO2 sequestration. In our case study, the ROMster framework reduces average CO2 injection prediction errors from 200% to only 4%.
Abstract Reduced-order models (ROMs) are a widely used and powerful approach to reducing the complexity of predictive physics-based numerical simulations for a wide range of applications, including electronics and fluid mechanics such as geologic CO2 sequestration (GCS). ROMs are critical for optimization, sensitivity analysis, model calibration and uncertainty quantification where full-order models cannot be feasibly executed many times. Traditional approaches generate a single ROM for each simulated response (e.g., CO2 injection rates, pH changes) based on a set of training simulations. Here, we demonstrate that a single ROM can display excellent overall predictive statistics, but have predictions that dramatically and unacceptably deviate from simulator responses especially when the response variable has a large range (i.e., vary over multiple orders of magnitude). For example, we show that a traditional statistically-high-performing GCS ROM (coefficient of determination R2 of 0.99) can have average absolute relative errors of over 200%. To address this, we propose a new and novel approach where a set of sub-ROMs are generated to overcome the potential pitfalls in traditional single ROM development. The effectiveness of the proposed approach—the ROMster framework—is demonstrated using a case study of predicted CO2 injection rates for GCS. We find our approach is a robust and general framework for ROM development, reducing the average “error” from 200% to only 4% in our case study.
Frankenstein’s ROMster: Avoiding pitfalls of reduced-order model development
Highlights Reduced-order models (ROMs) are a powerful approach to reducing the complexity of physics-based numerical simulations. We show that a traditional ROM can display excellent overall statistics and yet have poor predictive power. We introduce a novel approach where a set of sub-ROMs are generated to overcome potential pitfalls in traditional ROMs. The effectiveness of the new approach—the ROMster framework—is demonstrated using injection rates for CO2 sequestration. In our case study, the ROMster framework reduces average CO2 injection prediction errors from 200% to only 4%.
Abstract Reduced-order models (ROMs) are a widely used and powerful approach to reducing the complexity of predictive physics-based numerical simulations for a wide range of applications, including electronics and fluid mechanics such as geologic CO2 sequestration (GCS). ROMs are critical for optimization, sensitivity analysis, model calibration and uncertainty quantification where full-order models cannot be feasibly executed many times. Traditional approaches generate a single ROM for each simulated response (e.g., CO2 injection rates, pH changes) based on a set of training simulations. Here, we demonstrate that a single ROM can display excellent overall predictive statistics, but have predictions that dramatically and unacceptably deviate from simulator responses especially when the response variable has a large range (i.e., vary over multiple orders of magnitude). For example, we show that a traditional statistically-high-performing GCS ROM (coefficient of determination R2 of 0.99) can have average absolute relative errors of over 200%. To address this, we propose a new and novel approach where a set of sub-ROMs are generated to overcome the potential pitfalls in traditional single ROM development. The effectiveness of the proposed approach—the ROMster framework—is demonstrated using a case study of predicted CO2 injection rates for GCS. We find our approach is a robust and general framework for ROM development, reducing the average “error” from 200% to only 4% in our case study.
Frankenstein’s ROMster: Avoiding pitfalls of reduced-order model development
Chen, Bailian (author) / Harp, Dylan R. (author) / Pawar, Rajesh J. (author) / Stauffer, Philip H. (author) / Viswanathan, Hari S. (author) / Middleton, Richard S. (author)
2019-10-29
Article (Journal)
Electronic Resource
English
Frankenstein's mechanised Birmingham tree-houses
| British Library Online Contents | 2002
Avoiding polyurethane grouting pitfalls
Online Contents | 1998
Avoiding the Pitfalls of Assisted Living Development
| British Library Online Contents | 1997
Jonathan Meades - Frankenstein's mechanised Birmingham tree-houses
Online Contents | 2002
Avoiding the Pitfalls of Partnering
| Wiley | 2005