Talonite APC — powered by proven real-time control and model-predictive-control technology — stabilises and optimises minerals processing circuits continuously and autonomously: stabilise, optimise, sustain. 24/7, independent of shift changes or operator experience.
Most minerals processing operations consistently fall short of maximum achievable performance. The gap between actual and potential represents significant unrealised value.
Feed variability, instrument failures, inconsistent operator actions and poorly configured control systems directly translate to lower yields and reduced recoveries.
Multiple interacting process variables overwhelm traditional single-loop PID control — the best achievable overall performance stays out of reach.
Performance dips with every handover when optimisation depends on individual operator skill rather than continuous, autonomous control.
“With APC off, process deviations are wide, erratic and asymmetric. With APC on, they collapse to a tight, symmetric distribution centred near zero — exactly where the process needs to be.”Before-and-after histograms, live Talonite APC deployment
Model Predictive Control anticipates process behaviour and acts proactively — moving from an unstable process to full optimisation, with compounding improvements at each stage.
MPC continuously predicts process behaviour over a future horizon, keeping controlled variables in their desired limits and rejecting disturbances — feed surges, trips, dosing errors — before they propagate.
Optimisation algorithms dynamically adjust set points to maximise throughput and recovery while minimising energy, reagent and raw material consumption — within defined constraints.
Optimal operation maintained continuously, 24/7, without dependency on individual operator skill levels.
A web-based real-time calculation and control engine — visually configurable process logic, seamless DCS/SCADA integration via OPC, no thick client required.
Industrial-grade model predictive control built for multivariable, non-linear, delay-prone minerals processes — constraint-aware at all times, deployed across multiple South African mining sites.
Process analysis → model development (step tests) → controller configuration → commissioning — collaborative at every phase, with site teams fully equipped to operate long-term.
Talonite APC deployments consistently deliver significant, quantifiable improvements across key performance indicators at operating sites.
Every Talonite APC project includes a post-implementation performance review — value measured, documented and sustained.
Delivered through the Talonite Three Horizons journey — Foundation → Optimise → Improve — with ROI tracked, validated and compounded through our Smart Eye value layer. Every solution, every discipline.
See how we deliver value →Instability in kiln temperature and atmosphere control forced furnace operators to compensate with excess reductant. Talonite deployed a Model Predictive Control (MPC) controller built for the kiln's slow, interacting, multivariable dynamics — well beyond standard PID. The result: significantly improved kiln stability and pre-reduction consistency, a 5% reduction in kg coal per ton of feed, and extended baghouse bag lifespan cutting maintenance cost and unplanned downtime.
Real-time control and model predictive control (MPC) layered over the existing base control system. It continuously stabilises and optimises circuits — pushing throughput, recovery and stability closer to the plant's true potential.
Around 80% variance reduction on controlled circuits and recovery uplifts of 5% or more, with higher throughput and more stable operation — gains that compound because they apply to every tonne processed.
No. APC integrates with your existing PLC / SCADA / DCS environment rather than replacing it. Stable base-layer loops are the prerequisite — which is why control loop health comes first in the Smart Plant journey.
An APC feasibility assessment identifies your highest-value circuits, quantifies the opportunity, and scopes a deployment with measurable, reviewed outcomes.