Milestones
List view
- Due by June 30, 2023
Pedro
Due by June 30, 2023•7/7 issues closed#### 4.1 Assignees: - Sarah - Pedro - Ludovico
Due by December 16, 2022•2/3 issues closedInclude scalable virtual machines dedicated for each user #### 4.1 Assignees: - JB
Due by December 16, 2022•1/1 issues closed#### Collaborating Subtasks: - 4.2 Sequential Design of Experiments - Abby & Miranda #### 4.1 Assignees - Pedro
Due by December 16, 2022•2/2 issues closed- Include support for vector variables in the ML/AI plugin - Add support for additional surrogate model tools (e.g. PyTorch, Sci-kit Learn) and additional normalization function forms in the ML/AI plugin #### Collaborating Subtasks: 2.2 – Fluor Small Pilot (expected in EY23) - Jay 3.1 – Absorber Intensification Modeling - Gregorios 3.2 - Absorber Intensification Prototyping and Experimentation - Charles Finny (ORNL) 4.3 – Machine Learning Support - Phan, Brian Bartoldson #### 4.1 Assignees: - Brandon - Anuja
Due by December 16, 2022•10/10 issues closedCollaborating Subtasks: 4.2 - Sequential Design of Experiments - Abby & Miranda 4.1 Assignees: JB, Pedro
Due by December 16, 2022•1/1 issues closed4.1 Assignee: Sarah
Due by September 30, 2022Collaborating Subtasks: 2.2 – Fluor Small Pilot - Anuja & John M 3.1 – Absorber Intensification Modeling - Gregorios 4.1 Assignee: Pedro, Anuja
Due by June 30, 2022•4/4 issues closedImprove the existing FOQUS cloud functionality to support parallel execution of multiple flowsheet simulations for large-scale integrated models (e.g. combined Aspen Plus, Python, IDAES models) in UQ, optimization modules Collaborating Subtasks: 2.2 - Fluor Small Pilot - Anuja & Josh Morgan 4.1 Assignee: JB
Due by December 16, 2022•9/12 issues closedCollaborating subtasks: 4.2 Sequential Design of Experiments - Mirand & Abby 4.1 Assignees: Pedro
Due by December 16, 2022•2/2 issues closed(Go/No Go) Evaluate the software development requirements in the FOQUS base code and SimSinter to enable a direct interface with dynamic process models in Aspen Custom Modeler (ACM) Collaborating Subtasks: 2.1 – Chevron Small Pilot Support - Daison 2.6 – TDA TCM Pilot - Ryan 4.1 assignee: Anuja
Due by December 16, 2022•3/4 issues closedIDAES has tools for using machine learning techniques to generate surrogate models. It is common to have an Aspen simulation that you would like to generate a surrogate model from to be used in an IDAES model. It would be nice to provide a turbine client back end for the IDAES tools to simplify the process. A Turbine client backend could also be useful for things like the trust region tool that allows black box models to be incorporated into equations oriented models like the ones used in IDAES.
No due dateWe can run parallel FOQUS flowsheets on AWS now, but it would also be useful to be able to run an optimization or surrogate model generation with adaptive sampling fully on the cloud. Optimization requires a head node to coordinate the flowsheet runs and use results to generate new runs. Currently the head node is the local FOQUS instance. Since the optimization can take a very long time to run, it would be really nice to send to full optimization job to the cloud and be able to monitor the job and get the results later.
No due dateImplement FOQUS Cloud/parallel computing capabilities. Josh, Byron
Due by September 30, 2020•5/5 issues closedImprove FOQUS capabilities for handling advanced flowsheets. Surrogate based optimization solver. Anuja
Due by June 30, 2020•1/1 issues closed- Integrate UQ within SDoE module - Implement in FOQUS other sampling approaches to map input space and response range Pedro, Brenda
Due by December 18, 2020•7/7 issues closed- FOQUS GUI usability and user experience improvements, specifically for SDOE Sarah, Brenda, Pedro, Christine, Towfiq, Miranda
Due by September 30, 2020•5/5 issues closed4.B Extend FOQUS usability with open-source software (i.e., Sci-lab, IDAES framework, CFD software).
- Open-Source: Sci-lab, IDAES, CFD and advanced flowsheets - Improving FOQUS capabilities for handling advanced flowsheets Anuja
Due by June 30, 2020•1/1 issues closedExpand FOQUS capabilities to interface with “AML-algebraic modeling language” frameworks and solvers Anuja
Due by April 30, 2020•1/1 issues closedAdding new features to perform NUSF.
Due by December 31, 2019•4/4 issues closed- Due by January 31, 2019•3/3 issues closed
Support Python 3 in main branch of FOQUS Support of Python 2 will still be maintain, but via the use of conda or virtual environments.
Due by February 28, 2019•9/9 issues closedholder
Due by November 1, 2018•6/6 issues closed## Sync user's Simulations use git across turbine deployments rather than "HTTP PUT" via turbine web api. ## Fill out Turbine Web API on AWS API Gateway/lambda -- implement all parts of API that are used by FOQUS GUI. ## Auto scaling AWS ### AWS Alarm on SQS - if len(SQS) > 0, start up EC2 instances - if len(SQS) == 0 for 10 minutes, terminate workers shutdown NAT
Due by September 30, 2018•3/3 issues closedKeith Beattie Create enough automated tests to thoroughly test FOQUS without human interaction. Failed tests should report a log that gives some indication of what succeeded and what failed. These tests should have the ability to run in an automated server (Jenkins or some equivalent). Tests should be added as new features are included in FOQUS to ensure all aspects of FOQUS are tested.
No due date•19/19 issues closedWe would like to improve the Heat Integration feature in the FOQUS Flowsheet. This was delayed. Byron Soepyan
No due date•2/2 issues closedAdding multi-objective optimization under uncertainty (MOOUU). This is currently/somewhat on hold, with possibly a new LLNL person taking it on... Pedro & Brenda
Due by September 30, 2020•18/18 issues closedComplete the initial, internal release of the Sequential DoE module (Task 3.1 for FY 2018) Brenda Ng Towfiq Ahmed Christine Anderson-Cook
Due by December 31, 2018•13/13 issues closedWe would like to give FOQUS the capability to perform parallel computing of simulations. This capability will reduce the time required to perform UQ (Uncertainty Quantification), OUU (Optimization Under Uncertainty), and SDOE (Sequential Design of Experiments). This corresponds to the FWP 6/30/18 milestone: Public release of integrated Cloud based (AWS) computing configuration and management system with FOQUS release. Josh Boverhof
Due by June 30, 2018•3/3 issues closed