SCENARIO 04 · STARCH PROCESSING · CROSS-STAGE
Cross-stage liquefaction and saccharification optimisation
Material genealogy from slurry to assay, stage-specific soft sensors, and a cross-stage twin for the enzyme, severity and residence-time trade-off.
ILLUSTRATIVE CASE STUDYIllustrative scenario. This is not a named customer project and claims no achieved result. It shows how NOVYRON would structure a technically governed engagement.
OPERATING CONTEXT
A starch-processing line sees seasonal feedstock variation. Liquefaction and saccharification are controlled as separate stages, and delayed assays make it hard to identify where the conversion variability actually starts.
CHALLENGE
- Feed solids and starch characteristics shift both the cooking and the enzyme response
- Residence-time cohorts are difficult to match with the later laboratory samples
- Over-severe liquefaction wastes energy; weak liquefaction burdens saccharification
- Slow saccharification ties up tanks and disrupts downstream scheduling
SYSTEM BOUNDARY
NOVYRON supplies the software intelligence, integration and lifecycle layer. The process equipment, instruments, PLC/DCS logic, safety systems and approved operating procedures stay under the control of the customer and their equipment vendors.
WHAT NOVYRON WOULD BUILD
- Build material genealogy from slurry preparation through to the saccharification assay
- Calculate solids, pH, enzyme and thermal-history features for each cohort
- Develop liquefaction viscosity and DE and saccharification glucose soft sensors
- Compare enzyme, severity and residence-time scenarios in a cross-stage twin
- Detect stalled conversion and forecast endpoint time with uncertainty
MODELLING & DECISION LOGIC
Hybrid kinetic models preserve the known reaction structure while learned residual models capture feedstock and equipment effects. Multi-objective optimisation explores enzyme cost, steam use, endpoint quality and tank time without crossing validated recipe limits.
DELIVERY WORKFLOW
- S1Assay, genealogy and residence-time mapping
- S2Historical cohort reconstruction
- S3Stage-specific soft sensors
- S4Cross-stage replay by feedstock lot
- S5Shadow endpoint and stall alerts
- S6Controlled advisory trials with quality approval
DATA THE SCENARIO WOULD USE
Feedstock lot with its starch and solids characteristics, linked to slurry preparation, pH, temperature, pressure, steam, flow and residence-time history; enzyme product, lot and dose; viscosity or DE checks after liquefaction; glucose, reducing-sugar or other approved assays through saccharification; tank occupancy and transfer times; operator interventions; and the downstream quality and scheduling outcome. Material cohorts are reconstructed so a delayed assay is assigned to the upstream history that actually produced it.
MEASUREMENT PLAN
A pilot starts from agreed definitions, baseline windows and exclusion rules. These are the measures that would be agreed, not results that have been obtained.
- Liquefaction DE and viscosity consistency
- Saccharification endpoint time and sugar variability
- Enzyme use per tonne of dry solids
- Specific steam use
- Stalled and extended batch frequency
- Tank utilisation and downstream schedule adherence
Results would be evaluated against comparable products, feed conditions and operating constraints, with model predictions, operator actions and external disturbances all kept traceable.
MEASURED RESULTS — STATUS
No customer deployment or quantified conversion benefit is claimed in this scenario. A controlled pilot would report soft-sensor and endpoint prediction error, assay variability, enzyme and steam intensity, stalled or extended batch frequency, tank utilisation and schedule adherence, split by comparable feedstock cohort. Only validated site results could replace this statement.
WHAT THE PILOT WOULD LEARN
How much downstream variability actually originates in liquefaction; which feedstock features change the enzyme response; when a conversion trajectory is genuinely stalled rather than slow; and whether cross-stage recommendations reduce total cost or merely move burden between stages. It should also mark the operating regions where the hybrid model loses validity and must defer to the laboratory.
DATA & GOVERNANCE
- Read-only and shadow-mode deployment come before any approved write-back.
- Access is role-based, and data and model versions are logged.
- Site safety systems and validated quality release stay authoritative.
- Customer data use, retention and model-training rights are documented.
- Every recommendation shows its assumptions, its confidence and its constraint status.
EXPECTED OUTCOME CATEGORY
Less cross-stage variability and a clearer time, enzyme and energy trade-off. Any quantified benefit has to come from controlled site data.
DOMAINS USED IN THIS SCENARIO
Discuss a comparable process
Give us the process boundary, the tags you already have, the quality measures, the constraints and your current baseline. We will propose a data assessment and a shadow pilot — without a hardware-vendor change.