SCENARIO 02 · LIQUID SEPARATION · PARALLEL UF TRAINS
Fouling-aware ultrafiltration and cleaning planning
Predicting fouling trajectory, remaining run length and cleaning effectiveness across parallel trains running different feed families.
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
Parallel ultrafiltration trains process feed families with different solids and temperature profiles. Cleaning is triggered by fixed thresholds, but run length and the recovery achieved after each clean vary considerably.
CHALLENGE
- Raw transmembrane-pressure trends distorted by temperature and flux changes
- No reliable forecast of remaining run length for the operators
- Cleaning that happens either too early or after severe performance loss
- Train imbalance that complicates production 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
- Standardise transmembrane pressure, flux, recovery and temperature-normalised permeability
- Segment production, backwash, relaxation and cleaning states automatically
- Train feed-aware fouling trajectory and threshold-time models
- Score every clean by permeability recovery and by the decline rate that follows it
- Expose train-balancing scenarios to production and maintenance planners
MODELLING & DECISION LOGIC
Time-to-event forecasting estimates when an agreed pressure or permeability boundary may be reached. Change-point detection separates an abrupt integrity or valve event from gradual fouling, and a digital twin tests candidate flux and backwash schedules before anyone runs them.
DELIVERY WORKFLOW
- S1Baseline membrane and analyser audit
- S2Derived-variable verification with the process engineers
- S3Historical run and cleaning replay by train and feed family
- S4Shadow run-length forecasting
- S5Advisory cleaning and train-sequencing pilot
- S6Maintenance and schedule integration after acceptance
DATA THE SCENARIO WOULD USE
Train-level feed, retentate and permeate flow; inlet and outlet pressure with transmembrane pressure; temperature, conductivity, turbidity or product quality; backwash, relaxation and valve-state sequences; cleaning recipe, chemical concentration, duration and rinse; membrane age and maintenance events; feed family or production order; and laboratory checks. Derived permeability and recovery are verified against the membrane vendor’s definitions before modelling.
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.
- Normalised permeability decline
- Hours or volume between effective cleans
- Cleaning recovery and repeat-clean rate
- Energy, water and chemical use per permeate volume
- Permeate quality and product retention
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 measured customer outcome exists for this scenario. A site pilot would compare like-for-like runs and report forecast error for remaining run length, normalised permeability decline, effective cleaning recovery, repeat-clean rate, water, chemical and energy intensity, and permeate quality. Any improvement claim needs a controlled baseline, agreed exclusion rules and customer approval.
WHAT THE PILOT WOULD LEARN
How to separate gradual fouling from an abrupt integrity, valve or sensor event; which feed families need their own decline model; which cleaning signatures precede incomplete recovery; and whether train-balancing advice improves schedule reliability without costing separation quality.
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
Better-timed cleaning and more predictable filtration capacity. Any chemical, water or availability gain requires a site baseline and a controlled comparison before it can be stated.
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.