D06 · STARCH PROCESSING · LIQUEFACTION
AI-Assisted Liquefaction Control
Control viscosity reduction and starch conversion through a stable liquefaction trajectory despite feed variability.
NOVYRON is software-first and vendor-neutral: this is configurable software, integration and lifecycle service, not process equipment. The existing PLC or DCS keeps interlocks and regulatory control, and every recommendation stays inside a site-approved operating envelope under human oversight.
THE INDUSTRIAL PROBLEM
Starch liquefaction has to balance solids loading, pH, temperature, enzyme dose, mixing and hold time at once. Weak control leaves high viscosity or incomplete dextrinisation, constrains heat transfer and pumping, and passes the variability straight into saccharification.
THE NOVYRON LAYER
NOVYRON models the liquefaction train from slurry preparation through jet cooking and holding to cooling. It estimates conversion and viscosity proxies from the data already collected, predicts the downstream impact, and recommends bounded enzyme, pH, temperature and residence-time targets for the feed state in front of you.
TYPICAL DATA INPUTS
- Dry-solids or density measurement
- Slurry and jet-cooker temperature and pressure
- pH and alpha-amylase dose
- Flow, tank level and a residence-time distribution proxy
- Mixer and pump load as viscosity proxies
- DE, viscosity or iodine laboratory results and feedstock lot
MODELS AND DECISIONS
- Soft sensor for viscosity and the liquefaction endpoint
- Residence-time and thermal-history digital twin
- Feedstock and solids variability classification
- Enzyme-dose against energy trade-off optimisation
- Anomaly detection for poor mixing, dosing loss or heat-transfer drift
- Forecast of the downstream saccharification outcome
What each deployment reads, which models it runs, how it connects and how much authority it is given are designed for the site. None of it is pre-installed hardware, and none of it is a guaranteed autonomous function.
CORE CAPABILITIES
- LOT-AWARE CONVERSION TRAJECTORY
- ENZYME & THERMAL SEVERITY ACCOUNTING
- HOLD-TUBE RESIDENCE-TIME VISIBILITY
- LINK TO DOWNSTREAM GLUCOSE PERFORMANCE
- RECOMMENDATION WITH UNCERTAINTY
- SOLIDS & THROUGHPUT SCENARIOS
HOW IT WORKS
- S1Map the slurry, cooking, holding and cooling stages as one connected train
- S2Build solids, enzyme and thermal-history features per cohort
- S3Align laboratory DE and viscosity results with their residence-time cohorts
- S4Train endpoint and downstream-impact models on the aligned data
- S5Validate shadow recommendations across feedstock lots and seasons
- S6Deploy bounded guidance and cross-stage coordination with saccharification
INTEGRATION POINTS
- PLC/DCS, historian and recipe system
- Enzyme dosing skids and pH control
- Density, flow and energy meters
- LIMS and feedstock material genealogy
- Saccharification tanks and downstream scheduling
NOVYRON connects read-only and runs in shadow mode first. Any write-back is designed, tested and approved separately. Interfaces are chosen to fit the architecture already on site rather than forcing a new hardware stack.
MEASUREMENT EXAMPLES
These are measures for a baseline and target-setting workshop, agreed from your own data. None of them is a statement of a result NOVYRON has achieved.
- Liquefaction endpoint or DE consistency
- Viscosity proxy and pumping load
- Enzyme use per tonne of dry solids
- Specific steam and energy use
- Throughput and hold-time compliance
- Downstream saccharification time and glucose yield proxy
TYPICAL APPLICATIONS
- STARCH SWEETENERS
- GRAIN & STARCH FERMENTATION
- FOOD INGREDIENT MANUFACTURE
- BIO-BASED CHEMICAL FEED PREPARATION
- PILOT & SCALE-UP STUDIES
DEPLOYMENT PATH
- 01Process and assay mapping
- 02Residence-time and data-cohort model
- 03Soft-sensor development
- 04Cross-stage historical replay
- 05Shadow advisory by feed family
- 06Controlled production deployment
WHAT STAYS OUTSIDE
Enzyme supplier guidance, food and feed safety requirements and validated thermal limits define the allowed operating region, and the system stays inside it. Models are calibrated separately for materially different feedstocks rather than transferred between them.
SAFETY, QUALITY AND GOVERNANCE
- The deployment architecture can be designed to segregate customer data by tenant and deployment boundary, subject to the hosting, access-control and contractual terms agreed with you. Edge, on-premises, private-cloud and managed-cloud patterns are all evaluated against your constraints.
- Role-based access, encryption in transit and at rest, audit logging, model and version traceability and a secure update process are defined during solution design.
- Data provenance, retention and permitted model-training use are documented contractually. Customer data is not silently pooled into general models.
- Predictions carry quality flags and uncertainty where that is technically meaningful. Drift and out-of-distribution conditions trigger review rather than false precision.
Bring us the process boundary
Share the process, the signals you already collect, the control architecture, the constraints and the outcome you need. An engineer turns that into a scoped data assessment, a shadow pilot and a deployment plan.