This simulator models a heated mixing vessel controlled by cascade architecture — an outer (master) temperature PI controller requests a utility-flow setpoint, and an inner (slave) flow PI controller manipulates a utility control valve to deliver that flow. Compare cascade control against a single temperature-to-valve loop and against fully manual valve operation while injecting utility supply disturbances.
• A real-time 3D cutaway of the heated vessel and agitator, internal heating coil, utility control valve, utility flow transmitter, vessel temperature probe, the two visibly separate master/slave controllers, and the utility source/disturbance valve, with home view, focus-selected-part, toggleable full enclosure, exploded view, auto-rotate, expand and label controls plus tappable numbered components. • Eleven live controls: temperature setpoint, a three-way control-architecture selector (cascade / single-loop / manual valve), temperature-loop proportional gain and integral time, flow-loop proportional gain and integral time, flow-process lag, thermal-process lag, utility supply factor, a cooling-disturbance magnitude, and a manual valve-opening percentage. • Play/pause, single-step (0.1 s) and larger-step (1 s) time controls, plus a playback-speed selector from 10× slow motion to 1-minute-per-second. • A dedicated "reduce utility capacity to 60%" disturbance action alongside start/stop trial controls. • Eight live metrics: temperature setpoint, vessel temperature, temperature error, outer-loop flow request, actual utility flow, valve demand percentage, current thermal equilibrium target, and utility availability factor. • A Curves & measurements tab with two charts (temperature target vs. actual, and flow target vs. actual), the full PI model equations, and snapshot measurements. • An Experiments tab with four guided scenarios (cascade baseline, utility disturbance, single-loop comparison, insufficient utility), a model-verification bench of independent automated checks, and a timestamped event log with a copyable trial report. • A Learn & assess tab with guided lessons (setting the outer target, closing the fast inner loop, applying a utility disturbance, recognizing saturation), a knowledge-check quiz and a written scope/reference statement.
In cascade mode, the outer temperature PI controller does not manipulate the valve directly — it outputs a flow setpoint that the inner flow PI controller tracks. Because the inner flow loop measures a fast variable (utility flow) rather than the slow vessel temperature, it can detect and correct a utility supply disturbance and adjust valve opening before the disturbance has had time to significantly shift vessel temperature.
In single-loop mode, the same PI logic instead compares temperature directly to setpoint and drives the valve without an inner flow correction, so the controller must wait for a temperature error to develop before it can respond to a utility disturbance — which the simulator's single-loop comparison experiment demonstrates directly against the same disturbance magnitude.
Both PI controllers use conditional (anti-windup) integration: flowSP is clamped between 0 and 60 flow units, and valve percent is clamped between 0% and 100%, with integration halted once an output limit is reached to prevent further windup. Vessel temperature approaches an equilibrium of 20 plus 1.5 times flow minus the cooling disturbance, and utility flow itself approaches 60 times valve fraction times the supply factor — so once the valve saturates at 100% open, no amount of additional gain or integral action can supply utility capacity that physically does not exist, as shown in the insufficient-utility experiment.
This is a lumped, lag-only model: normalized utility-flow units, first-order thermal and flow dynamics, and PI control with no thermodynamic steam properties, transport delay or derivative action. Manual-mode integrators track the manual operating point, so switching architecture mid-run can still produce a transient even though it is not itself a process disturbance.
In cascade mode the outer temperature PI controller outputs a requested flow setpoint (flowSP) for the inner loop — it does not drive the control valve directly. The inner flow PI controller compares that requested flow to actual measured utility flow and manipulates valve opening to correct any error.
The inner flow loop measures a fast variable (utility flow) and can react to a supply disturbance immediately. A single loop instead measures the slow vessel temperature directly, so it must wait for a temperature error to actually develop before it can begin correcting for the same utility disturbance — this is exactly what the single-loop comparison experiment isolates.
No. Both controllers use conditional integration that halts once their output saturates (flow request at 60 units, valve at 100% open). Once the valve is fully open and utility flow still cannot meet demand, as in the insufficient-utility experiment at reduced supply factor and a 95°C setpoint, no controller tuning can create missing physical capacity.
It uses normalized utility-flow units, lumped first-order thermal and flow dynamics, and PI-only control (no derivative action or transport delay), with no real thermodynamic steam properties. Manual mode tracks the manual operating point in its integrators, so mode switches can still produce transients that should be compared against identical disturbances and tuning.