GFIS Model Control Room
Main Level 2 interactive interface for methane prediction, soft sensor output, stability state, optimization, evaluation, and research evidence.
Launch model control roomGFIS Level 2
Physics-Guided AI Digital Twin for Biogas Methane Production
Level 2 keeps the existing GFIS website intact and adds a serious working layer for BITS Pilani dissertation evidence. It is designed to connect the ML engine, industrial simulation, reports, diary, and deployment pipeline in one place.
Accepted Abstract Feedback
The BITS abstract evaluation accepted GFIS with an Excellent remark and specifically pushed the work away from black-box prediction toward PIML, soft sensing, and digital-twin simulation. In this portal, PIML is used as Physics-Informed Machine Learning; the credit here is for the evaluator's framing and guidance that sharpened GFIS into a physics-guided digital-twin research prototype.
Live GFIS Workbench
Main Level 2 interactive interface for methane prediction, soft sensor output, stability state, optimization, evaluation, and research evidence.
Launch model control roomPlant-style simulation surface for process behavior, methane production, operating drift, and industrial demonstration scenarios.
Launch industrial simulationRapid dissertation dashboard for dataset viewing, model checks, metrics, and supervisor demos. The production target is deployed on AWS App Runner instead of localhost.
Open AWS Streamlit dashboardEarlier M.Tech dashboard upgraded into the Level 2 workspace for live monitoring, anomaly controls, research paper context, and industrial AI storytelling.
Open industrial AI dashboardCompact live monitoring dashboard from the original M.Tech work, retained as a reference implementation and dashboard lineage artifact.
Open core dashboardDashboard Integration
The dashboard layer is arranged like an engineering console: public demonstrations remain open, research records stay gated, and older M.Tech surfaces are preserved as useful operating views rather than disconnected pages.
Primary command layer for project scope, engines, secure documents, diary, readings, and deployment evidence.
Stay in workbenchPrediction, simulation, optimization, soft sensor, explainability, and evaluation interface connected to the GFIS API.
Open control roomPremium SCADA-style plant run with telemetry, threshold warnings, system logs, and process behavior visualisation.
Open simulationUpgraded earlier dashboard for supervisor demos, architecture storytelling, anomaly controls, and industrial AI framing.
Open AI dashboardCompact monitoring surface retained as lineage evidence and as a quick view for reactor metrics and model behavior.
Open core dashboardEngine Console
Runs XGBoost, LSTM, ensemble, and physics-guided yield checks.
/predict
Changes temperature, pH, OLR, HRT, TS, and VS to study methane response.
/simulate
Plays a 48 hour plant scenario with process drift and stability transitions.
/plant-run
Estimates VFA/ALK ratio and maps process state into stable, warning, or critical.
/soft-sensor
Live Readings
Temperature36.4 C
pH7.08
OLR2.9 kg VS/m3/day
VFA/ALK0.24
Physics Violations0
Secure Prototype Workspace
Restricted Sections
Documentation, Library, Diary, Notes, and user administration are gated here. Public Level 2 simulations remain open for review and demonstration.
Profile
Role
Secure Documentation Vault
| Folder | Document | Type | Status |
|---|---|---|---|
| Login complete. Loading secure manifest... | |||
Collaboration And Support
Deployment