Compressor Controls, Sequencing & Multi-Compressor Optimisation

COMPRESSOR CONTROLS, SEQUENCING & MULTI-COMPRESSOR OPTIMISATION

Multiple compressors do not become an optimised system merely because they share a header

The objective is stable required pressure and flow with protected machine limits, credible standby capacity and minimum avoidable unloaded or inefficient operation—without sacrificing reliability.

1 — Control Design Basis

Record technology, corrected capacity, pressure range, fixed-speed/VSD control, minimum/maximum limits, unloaded power, start limits, controller capability, communications, maintenance state and intended role for every machine.

2 — Controlled Pressure Boundary

Choose a pressure measurement location representing the system being controlled. Compressor discharge and common dry-header sensing can drive different behaviour.

3 — Pressure-Sensor Integrity

Critical systems can justify redundant/validating sensors. Drift, isolation or failure can make a sequencer control the wrong pressure.

4 — Pressure Band

A wider band provides more storage/time but can raise average pressure; a narrow band improves consistency but needs storage and responsive capacity.

6 — Base / Trim / Standby

Base carries sustained demand, trim follows variation and standby protects continuity. Roles can change with efficiency, maintenance and demand.

7 — Fixed-Speed States

Loaded, unloaded, stopped and transition states have different power/airflow. Unloaded power must be visible.

8 — VSD Operating Envelope

VSD trim works within supported speed/pressure limits. Minimum output, maximum capacity and part-load efficiency are product-specific.

9 — Base Machine Selection

The largest compressor is not automatically the base machine. Use load duration, corrected capacity, specific energy, reliability and maintenance.

10 — VSD Role Selection

VSD is not automatically base. Fixed-speed base plus VSD trim can be strong in some plants; exact evidence governs.

11 — Cascade Control

Staggered local setpoints can stage machines, but overlap, wide bands and poor storage can cause hunting or excessive pressure.

12 — Central Sequencing

A master controller can select machine combinations using pressure, demand, availability, efficiency and programmed constraints.

13 — Equal-Hour Rotation

Balancing hours can support maintenance but is not automatically energy-optimal.

14 — Specific Energy

Compare electrical input per reconciled useful airflow at a defined pressure/boundary.

15 — Worked Specific-Energy Example

Illustrative only: 20 m3/min at 120 kW gives 6.0 kW/(m3/min); 16 m3/min at 104 kW gives 6.5 on the same basis. Real selection also needs part-load/system effects.

16 — Demand Duration Curve

Analyse time spent in demand bands to identify base load, trim variation and peaks.

17 — Average Demand Can Mislead

A 30 m3/min average can hide swings from 15 to 45 m3/min. Controls must follow the profile.

18 — Simultaneous Loading Overshoot

Several independent machines can load on one dip. Storage, delays and master control should prevent hunting.

20 — Start / Stop Limits

Minimum stop time, maximum starts, warm-up and other limits remain machine-specific.

21 — Minimum Run Time

Protect equipment while avoiding unnecessary unloaded operation.

22 — Unload-to-Stop Delay

Too short can create frequent starts; too long wastes unloaded power.

23 — Trim Headroom

A VSD at maximum speed cannot absorb an upward step. Stage additional capacity before trim authority is exhausted where needed.

24 — VSD Minimum Output

When demand falls below useful minimum output, change fleet combination rather than tolerating inefficient operation.

25 — Multiple VSDs

Independent VSD loops can compete or hunt. Use supported coordinated architecture.

26 — Mixed Compressor Sizes

Different capacities can improve demand matching when combinations are selected intelligently.

27 — Mixed Technologies

Screw, centrifugal, reciprocating and other technologies have different response/control constraints.

28 — Centrifugal Boundary

Respect exact minimum stable flow, surge/blow-off and inlet-control limits.

29 — Reciprocating Capacity Steps

Unloaders/cylinder control can create discrete steps that sequencing must recognise.

30 — Setpoint Governance

Approved source/machine setpoints are controlled engineering values, not informal operator adjustments.

32 — Remote Pressure Sensing

Representative downstream sensing can reduce excess source pressure, but communications/local protection must be reliable.

33 — Zone Control

Pressure/flow control with storage can decouple production zones from a higher source header.

35 — Drain Air Loss

Failed-open or poorly adjusted drains create artificial demand that should be repaired.

36 — Leakage Distorts Fleet Selection

High leakage can keep an extra machine online. Re-optimise after major leak repair.

37 — Production Changes

Cycle-time, machine and shift changes can alter the demand curve.

38 — Worked Two-Compressor Scenario

Illustrative only: a 20 m3/min fixed-speed machine plus 12 m3/min VSD serve 18–29 m3/min. Fixed-speed base plus VSD trim may fit if exact VSD range/performance supports it.

39 — Worked Low-Demand Scenario

Illustrative only: night demand is 6 m3/min while a large fixed-speed unit cycles. A smaller machine or VSD may reduce unloaded operation if reliability and limits permit.

40 — Worked Demand-Step Scenario

Illustrative only: demand rises 10 m3/min in five seconds. If storage cannot bridge the next-machine response, pressure falls. Storage, earlier staging, trim headroom or process sequencing may solve it.

41 — Standby Readiness

A faulted, isolated, cold or maintenance-disabled machine is not operational standby.

42 — N+1 Definition

N+1 means remaining available system can satisfy defined critical demand after loss of one required component.

43 — Common-Mode Dependencies

Ventilation, feeder, cooling water, treatment or controls can remove several machines simultaneously.

44 — Maintenance Mode

The master should know unavailable machines and re-optimise remaining capacity. Temporary overrides must be removed.

45 — Manual Mode

Manual control can aid troubleshooting but should not become permanent optimisation.

46 — Communication Failure

Define safe local fallback if master communication is lost.

47 — Sensor Failure

Use plausibility logic or redundant sensing where consequence warrants it.

48 — Controller Failure

Critical plants may require redundant master control, local fallback or documented manual recovery.

49 — Time Synchronisation

Align compressor, pressure, flow, power and production timestamps.

50 — Trend Machine States

Log loaded/unloaded/stopped state, speed, power, flow, pressure, alarms and master commands where available.

51 — Power Boundary

Measure package power consistently and separate auxiliaries when assessing system energy.

53 — Loaded vs Unloaded Hours

High unloaded hours can indicate poor fleet matching or delay settings.

54 — Starts and Transitions

Track starts, load/unload cycles and VSD operating range.

55 — Pressure Stability KPI

Track minimum, maximum and time outside accepted pressure band.

56 — Specific-Energy KPI

Trend kW per reconciled airflow at defined pressure conditions.

57 — Standby KPI

Track availability, faults, isolation and maintenance state.

58 — Alarm Rationalisation

Classify alarms by consequence and assign ownership/action.

59 — Analytics Need Data Quality

Advanced analytics cannot fix bad sensors, inconsistent flow bases or missing states.

60 — Remote Optimisation Change Control

Remote setpoint/sequencing changes require approval, audit trail and rollback.

61 — OT / Cyber Interface

Network segmentation, accounts, remote access, updates and backups follow site IT/OT governance.

62 — Configuration Backup

Preserve controller configurations, setpoints, logic, network settings and versions before changes.

64 — Black-Start Sequencing

After power restoration, ventilation, dryers, drains, controls and compressors may need ordered restart.

65 — Planned Shutdown Sequencing

Define unloading, stopping, isolation and depressurisation while protecting critical air.

66 — Emergency Stop Boundary

Emergency-stop functions belong to safety design and are not routine sequencing controls.

67 — Independent Pressure Protection

Software does not replace pressure-vessel relief/protection.

68 — High-Pressure Trips

Repeated trips can indicate poor sequencing, failed unloading, sensor error or blocked flow.

69 — Low-Pressure Trips

Diagnose demand, source response, storage, distribution and local restriction before adding capacity.

70 — Honest Availability

A machine with recurring trips or overdue service should not be counted as fully reliable capacity.

71 — Maintenance Windows

Use load profiles to plan service during naturally low-demand periods where possible.

72 — Seasonal Demand

Weather, production seasonality and dryer loads can change fleet roles.

73 — Ambient / Altitude Correction

Use exact corrected performance rather than nameplate capacity.

74 — Heat-Recovery Economics

Include real recovered-heat value in total-energy analysis without running unnecessary compressors solely to make heat.

75 — Tariff-Aware Scheduling

Electricity tariffs can influence scheduling while preserving reliability and machine efficiency.

76 — Demand Response

Any compressor reduction for electrical demand management needs a defined compressed-air risk boundary.

77 — Scheduled Pressure Reduction

Lower pressure during low-demand periods only after validating critical users and fallback.

78 — Night / Weekend Scheduling

Low demand may justify a smaller machine combination; verify leakage, regeneration and standby.

79 — Maintenance-Based Priority

Deprioritising a machine nearing service can be valid when controlled and documented.

80 — Maintain Efficiency Maps

Measured capacity/power can change with fouling, ageing or overhaul. Update optimisation models.

81 — Sequencer Acceptance

Verify pressure stability, transitions, start limits, VSD range, standby response, communication failure and representative demand.

82 — Baseline Before Optimisation

Record representative pressure, flow, power and machine states before change.

83 — Post-Change Verification

Compare equivalent KPIs/production conditions and confirm savings did not increase pressure faults or cycling.

84 — Worked Before/After Example

Illustrative only: before optimisation, two fixed-speed machines spend long periods unloaded while pressure varies. After controlled sequencing, one carries base load and a trim machine follows demand. Prove any benefit with measured power, flow, pressure and production data.

85 — Continuous Governance

Demand, machines and maintenance change. Review sequencing periodically rather than treating one commissioning setting as permanent.

87 — Premium Multi-Compressor Control Register

88 — Sequencing Failure-Mode Matrix

89 — Technical / Controls Verification Gate

Control pressure boundary is explicitly defined.

Machine capacities and limits use corrected/evidenced data.

Base/trim/standby roles are justified.

Storage and pressure-band interaction is included.

Start/stop/minimum-run constraints are protected.

VSD minimum/maximum boundaries are controlled.

Mixed-technology constraints are respected.

Dryer regeneration, drains and leakage are included where material.

Standby readiness and N+1 boundary are defined.

Sensor/communication/controller failure behaviour is defined.

Pressure, flow, power and state data use reconciled boundaries.

Configuration and firmware backups are preserved.

Remote changes follow approval/rollback governance.

Sequencer commissioning covers representative demand cases.

Post-change performance is verified against baseline.

Periodic re-optimisation/change-control triggers are defined.