Web-Based Tools for Data-Informed Remedy Optimization: Software Theory and User Guide

Song, Xuehang, Lopez Jr, Francisco J., Corney, Mikaela C. et al. · 2025

Read Web-Based Tools for Data-Informed Remedy Optimization: Software Theory and User Guide by Song, Xuehang, Lopez Jr, Francisco J., Corney, Mikaela C. et al. (2025) in the Tower of Babel, a free online library of public-domain works preserved in clean, searchable text. No sign-up. The full text is served as a plain-text file at a predictable URL, where text is present: https://antarctic-labs.com/tower-of-babel/doe-osti-3001001-web-based-tools-for-data-informed-remedy.txt.

Introduction

Xuehang Song, Frank Lopez Jr., Mikaela Corney and a Pacific Northwest team document two web tools in PNNL-38413, dated September 2025, for cleaning up contaminated groundwater at Hanford. Remedy decisions there rest on huge monitoring datasets and slow physics models. The tools speed the loop: PTOLEMY builds fast surrogate models of contaminant spread, and OPTIMA uses them to test remedy designs.

This software theory and user guide explains both systems, their maths and their screens, for the analysts who run them. Current surrogates cover carbon tetrachloride and technetium-99 in Hanford's 200 West Area. PTOLEMY is operational, with periodic retraining as new monitoring data arrives; OPTIMA is still under internal development and validation before release. Planned work widens coverage to more contaminants and sites. The guide mixes theory chapters with step-by-step usage, so a new analyst can reproduce a remedy optimisation run end to end.

Catalog record

CREATORSong, Xuehang, Lopez Jr, Francisco J., Corney, Mikaela C. et al.
YEAR2025
COLLECTIONPAPERS
FORMATTXT
FILE SIZE90 KB
LICENSEPublic Domain
RIGHTSPUBLIC DOMAIN
SOURCEDOE OSTI

Source: https://www.osti.gov/biblio/3001001

Browse the full library · About the Tower of Babel

Full text

The opening of the work, from the library's hosted text:

PNNL-38413, Rev 0 DVZ-RPT-121, Rev 0 Web-Based Tools for Data-Informed Remedy Optimization: Software Theory and User Guide September 2025 Xuehang Song Frank Lopez Jr. Mikaela Corney Paul Tran Yusuf S. Afzal Tommy E. Joppich Sophie N. Baur Hung Luu Reem Osman Christian D. Johnson Inci Demirkanli Prepared for the U.S. Department of Energy under Contract DE-AC05-76RL01830 Choose an item. This report was prepared as an account of work sponsored by an agency of the United States Government. Neither the United States Government nor any agency thereof, nor Battelle Memorial Institute, nor any of their employees, makes any warranty, express or implied, or assumes any legal liability or responsibility for the accuracy, completeness, or usefulness of any information, apparatus, product, or process disclosed, or represents that its use would not infringe privately owned rights. Reference herein to any specific commercial product, process, or service by trade name, trademark, manufacturer, or otherwise does not necessarily constitute or imply its endorsement, recommendation, or favoring by the United States Government or any agency thereof, or Battelle Memorial Institute. The views and opinions of authors expressed herein do not necessarily state or reflect those of the United States Government or any agency thereof. PACIFIC NORTHWEST NATIONAL LABORATORY 76RL01830 Acknowledgments iv Acronyms and Abbreviations Adam Adaptive Moment Estimation CPU central processing unit CSV comma-separated values DOE U.S. Department of Energy eSTOMP (STOMP family) groundwater simulator HFO Hanford Field Office JPEG Joint Photographic Experts Group (image format) MC3D-CNN multi-channel 3-D convolutional neural network NSGA-II Non-dominated Sorting Genetic Algorithm II OPTIMA Optimization for Pump-and-Treat Implementation, Management, and Assessment P2R Plateau-to-River (groundwater model) PDF Portable Document Format P&T pump-and-treat PTOLEMY Pump-and-Treat Optimized Location Evaluation to Maximize Yields RMSE root mean square error ROD Record of Decision SOCRATES Suite Of Comprehensive Rapid Analysis Tools for Environmental Sites U-Net U-Net convolutional neural network architecture ZIP ZIP archive (compressed file format) Acronyms and Abbreviations v Contents Summary ...................................................................................................................................................... iii Acknowledgments........................................................................................................................................ iv Acronyms and Abbreviations ....................................................................................................................... v 1.0 Introduction ...................................................................................................................................... 1 1.1 Purpose and Scope .............................................................................................................. 1 1.2 Site Background .................................................................................................................. 1 1.3 Overview of PTOLEMY and OPTIMA ............................................................................. 4 1.4 Intended User and Applications .......................................................................................... 5 1.5 Report Organization ............................................................................................................ 5 2.0 PTOLEMY Theory .......................................................................................................................... 7 2.1 Multi-Channel 3D Convolutional Neural Network Approach ............................................ 7 2.2 Training Data Development and Performance Classification............................................. 8 2.3 Model Training and Performance Evaluation ..................................................................... 9 2.4 Decision Criteria and Constraints ....................................................................................... 9 2.5 Current Limitations and Assumptions .............................................................................. 10 3.0 PTOLEMY User Guide ................................................................................................................. 11 3.1 Interface Overview ........................................................................................................... 11 3.1.1 Map Panel ......................................................................................................... 11 3.1.2 Performance Probability Distribution Graph Panel .......................................... 12 3.2 Interface Overview ........................................................................................................... 12 3.2.1 High Performance Thresholds and Contaminants ............................................ 12 3.2.2 Suggested Well Locations ................................................................................ 13 3.2.3 Adding New Well Locations ............................................................................ 15 3.3 Analysis Results ................................................................................................................ 15 3.3.1 Interactive Chart–Map Linking ........................................................................ 16 3.3.2 Chart Settings ................................................................................................... 17 3.4 Exporting Results and Reports.......................................................................................... 17 3.5 Summary ........................................................................................................................... 19 4.0 OPTIMA Theory............................................................................................................................ 20 4.1 Framework Overview ....................................................................................................... 20 4.2 Surrogate Modeling: U-Net with Physics-Informed Well Drawdown Channel ............... 21 4.2.1 U-Net Surrogate for Multi-Variable Prediction ................................................ 21 4.2.2 Input Representation and Steady-State Drawdown Channel ............................ 21 4.2.3 Training Data and Performance ........................................................................ 22 4.3 Objective Formulation ...................................................................................................... 23 4.3.1 Remediation Objectives.................................................................................... 23 4.3.2 Decision Variables and Design Representation ................................................ 23 Contents vi 4.3.3 Operational Constraints .................................................................................... 23 4.4 NSGA-II Evolutionary Optimization Engine ................................................................... 24 4.4.1 Evolutionary Search Process ............................................................................ 24 4.4.2 Surrogate-in-the-Loop Evaluation and Outputs................................................ 24 4.5 Current Limitations & Assumptions ................................................................................. 24 5.0 OPTIMA Design and User Guide .................................................................................................. 26 5.1 Interface Overview ........................................................................................................... 26 5.1.1 Simulation Overview Page ............................................................................... 26 5.1.2 Workspace Layout ............................................................................................ 27 5.2 Defining an Optimization Scenario .................................................................................. 27 5.2.1 Operational Constraints .................................................................................... 27 5.2.2 Objectives Selection ......................................................................................... 28 5.3 Narrowing Results ............................................................................................................ 29 5.3.1 Full Candidate Set ............................................................................................ 29 5.3.2 Filtered Subset and Progression ....................................................................... 30 5.4 Compare Solutions ............................................................................................................ 31 5.4.1 Initial State........................................................................................................ 31 5.4.2 Single Solution Selected ................................................................................... 32 5.4.3 Current vs. Pinned Time ................................................................................... 33 5.4.4 Layout Controls ................................................................................................ 33 5.5 Exporting Results and Reports.......................................................................................... 34 5.6 Summary ........................................................................................................................... 35 6.0 Conclusions .................................................................................................................................... 36 7.0 Continued Development ................................................................................................................ 37 8.0 Quality Assurance .......................................................................................................................... 38 9.0 References ...................................................................................................................................... 39 Contents vii Figures Figure 1.1. Extent of Contamination Plumes with Concentrations above Water Quality Standards on the Central Plateau (Source: DOE/RL-2020-60, Rev. 0) .............................. 2 Figure 1.2. Hydrostratigraphic Units and Generalized Hanford Site Stratigraphy (Source: Hammond and Lupton 2015) .............................................................................................. 3 Figure 2.1. Architecture of the MC3D-CNN Classification Model ...................................................... 8 Figure 2.2. Deep Learning Workflow for Well Performance Classification, from Training Data Generation to MC3D-CNN Prediction ....................................................................... 9 Figure 3.1. PTOLEMY User Interface ................................................................................................ 11 Figure 3.2. Map Options in PTOLEMY.............................................................................................. 12 Figure 3.3. High-Performance Threshold Adjustment Options .......................................................... 13 Figure 3.4. Suggested Well Locations on Map ................................................................................... 13 Figure 3.5. Suggested Well Locations and Distance from Nearest Extraction Well Options ............. 14 Figure 3.6. Distance From Nearest Extraction Well Display on Map................................................. 14 Figure 3.7. Adding a New Well Location to Map ............................................................................... 15 Figure 3.8. Performance Probability Distribution Graph .................................................................... 16 Figure 3.9. Corresponding Chart Lines and Map Well Locations ...................................................... 16 Figure 3.10. Changing Graph Y-Axis ................................................................................................... 17 Figure 3.11. Data Export Button ........................................................................................................... 18 Figure 3.12. Data Export Options ......................................................................................................... 18 Figure 4.1. OPTIMA Computational Framework Illustrating the Integrated Workflow .................... 20 Figure 4.2. U-Net Deep-Learning Framework Used by OPTIMA ..................................................... 21 Figure 5.1. OPTIMA Simulation Overview (landing page) ................................................................ 26 Figure 5.2. Set Parameters – Operational Constraints......................................................................... 28 Figure 5.3. Set Parameters – Custom Inputs and Objectives .............................................................. 29 Figure 5.4. Refine Results – Full Candidate Set ................................................................................. 30 Figure 5.5. Refine Results – Filtered Subset Highlighted ................................................................... 30 Figure 5.6. Compare Solutions – Initial (no selection) ....................................................................... 31 Figure 5.7. Compare Solutions – One Solution Selected .................................................................... 32 Figure 5.8. Compare Solutions – Current vs. Pinned .......................................................................... 33 Figure 5.9. Compare Solutions – Collapsed Left Panel ...................................................................... 34 Figure 5.10. Compare Solutions – Drag to Resize Panels..................................................................... 34 Figure 5.11. Download Assets Dialog .................................................................................................. 35 Contents viii 1.0 Introduction 1.1 Purpose and Scope This report documents the development and application of two complementary decision-support tools – PTOLEMY (Pump-and-Treat Optimized Location Evaluation to Maximize Yields) and OPTIMA (Optimization for Pump-and-Treat Implementation, Management, and Assessment) – designed to enhance groundwater remediation efforts, specifically focusing on pump and treat (P&T) operations, at the Hanford Site’s Central Plateau. These web-based tools are implemented as modules within the SOCRATES (Suite Of Comprehensive Rapid Analysis Tools for Environmental Sites) platform, a framework for data access, visualization, and analytics developed by Pacific Northwest National Laboratory to support remedy optimization. These tools were created to improve the effectiveness and efficiency of the 200 West Area pump-and-treat (P&T) system by guiding P&T well placement and operation strategies as pre-screening-level decision support. PTOLEMY and OPTIMA integrate advanced computational methods (e.g., deep learning surrogates and evolutionary algorithms) with the existing Hanford Site groundwater flow and transport model to support performance-based remedy management. The purpose of this report is to describe both tools, explain their context and rationale, and demonstrate how they work together to inform P&T system optimization. The scope includes an overview of contaminant and aquifer conditions and challenges at the Hanford 200 West Area, the technical architecture and features of each tool, their intended use

Read the complete text