D01 · THERMAL PROCESSING · DRYING
AI-Enabled Industrial Drying Optimisation
Stabilise final moisture, reduce avoidable thermal load and detect drift before off-spec product leaves the dryer.
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
Industrial dryers run with long dead times, changing feed moisture and uncertain heat- and mass-transfer conditions. A fixed recipe compensates with excess residence time or temperature, which costs energy and risks overdrying, colour change, degradation or lost capacity.
THE NOVYRON LAYER
NOVYRON reconciles inlet conditions, thermal states, airflow and quality measurements into a soft sensor for outlet moisture and a constrained recommendation engine, around the dryer you already run. Interlocks and regulatory control stay in the PLC or DCS.
TYPICAL DATA INPUTS
- Inlet and outlet temperature and humidity
- Product feed rate and estimated inlet moisture
- Airflow, pressure drop and recirculation position
- Heater duty, steam flow, gas or electrical consumption
- Dryer speed, residence-time proxy and exhaust conditions
- Laboratory or inline moisture and quality results
MODELS AND DECISIONS
- Hybrid moisture soft sensor: first-principles mass and energy balances with a learned residual
- Time-series forecasting of outlet moisture at the process dead-time horizon
- Constrained optimisation of temperature, airflow, recirculation and residence-time targets
- Change-point and residual monitoring for fouling, air leaks, sensor drift or feed change
- What-if digital twin for grade changes and start-up recipe validation
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
- GRADE CONTEXT & RECIPE VERSIONING
- MOISTURE CONFIDENCE INTERVAL
- ENERGY PER KILOGRAM EVAPORATED
- RECOMMENDATION WITH EXPECTED EFFECT
- GOLDEN-RUN COMPARISON
- MODEL DRIFT & SENSOR HEALTH
HOW IT WORKS
- S1Connect historian and PLC tags, energy meters, recipe context and quality records
- S2Build mass and energy features, and align delayed laboratory moisture with process data
- S3Learn grade-specific moisture trajectories and the normal operating envelope for each
- S4Run predictions in shadow mode and compare them with laboratory or inline analyser results
- S5Enable bounded operator guidance, then optional setpoint writing through approved controls
INTEGRATION POINTS
- OPC UA/DA, Modbus TCP and MQTT gateways
- Plant historian, MES and batch or recipe systems
- Inline NIR or microwave moisture analysers
- DCS/PLC advisory setpoints and alarm management
- Energy meters and steam or fuel accounting
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.
- Final-moisture mean and standard deviation
- Share of product inside the moisture specification
- Specific energy consumption per kilogram of water removed
- Throughput at equal quality
- Overdrying, rework and scrap rate
- Start-up and grade-change stabilisation time
TYPICAL APPLICATIONS
- FOOD INGREDIENTS & POWDERS
- STARCH & CARBOHYDRATE PRODUCTS
- CHEMICALS & MINERALS
- PULP, PAPER & BIOMASS
- SPECIALITY MATERIALS
DEPLOYMENT PATH
- 01Data and heat/mass-balance assessment
- 02Instrument and tag-quality review
- 03Offline soft sensor and baseline model
- 04Shadow-mode validation across representative grades
- 05Operator-advisory pilot
- 06Governed rollout with periodic recalibration
WHAT STAYS OUTSIDE
Burner management, over-temperature trips, explosion protection, airflow permissives and product-specific safety limits stay where they are. The system is designed not to bypass them and its recommendations are constrained to an approved operating envelope — a boundary that is established by site verification, independent safeguards and commissioning tests, not by the software asserting it.
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.