Remote Mine Drainage Basin Level Control during Network Outages

A remote mine drainage basin cannot depend on a public or long-distance network for immediate pump control. Inflow may continue during a communications outage, so level-based sequencing, high-level action and safe fallback should execute in the local PLC or RTU. Cellular or other gateways remain valuable for trends, alarms and maintenance, but remote visibility must be clearly separated from the control functions needed to prevent flooding. The design must cover rapid inflow, sediment, turbulent pump zones, power loss and a failed duty pump as well as normal operation. Assign a maximum unattended duration and prove that local storage, controller power and alarm escalation support it. If backup power is provided, calculate runtime with the actual PLC, instruments, modem, relays and control loads, then define the behaviour before the battery is exhausted.

Define hydraulic and consequence cases

Record basin geometry, inflow range, maximum rate of rise, pump curves, suction limits, discharge constraints and retained storage. Include storm inflow, blocked screens, sediment accumulation and one or more pumps unavailable.

Set start, stop, low, high and high-high levels from those cases. Calculate how much water enters during pump start, valve movement and operator response.

Keep critical logic local

Implement lead-lag rotation, standby start, minimum run time, unavailable-pump handling and high-level response in the local controller. Confirm operation without SCADA, cloud service or cellular coverage.

Define manual controls and restoration after power or communications return. Avoid automatic restart where an unconfirmed downstream condition could make pumping unsafe.

Install a representative level measurement

Keep the radar beam away from inlet waterfalls, pump turbulence, ladders and supports. Verify the target at low water and maximum inflow and protect the mount from vibration and flooding.

Account for sediment changing the effective bottom. Use a stable surveyed datum and distinguish water level from calculated volume or pumpable capacity.

Make remote data trustworthy

Send level with timestamp, signal quality, lost echo, device status, pump state and controller health. A repeated value must be labelled stale rather than displayed as current.

Buffer records during network loss and forward them in time order after reconnection. Preserve event timestamps so delayed alarms are not mistaken for new conditions.

Secure and supervise the gateway

Use authenticated management, encrypted transport, least-privilege access and controlled configuration backups. Separate operational networks where required and record firmware and credential changes.

Monitor signal strength, data usage, heartbeat and recovery time. The gateway's failure must alarm without disabling the local pump sequence.

Test outage and failure scenarios

Commission known levels, pump starts and stops, high-high response, lost echo and failed pumps. Then isolate the network and prove that local control continues while records are buffered.

Restore communications and verify ordered data recovery, alarm meaning and operator displays. Retain cause-and-effect results and repeat tests after logic, modem or network changes. Include a timed outage long enough to expose storage, retry and power limitations rather than a brief cable-disconnection demonstration.

Engineering checklist

Calculate maximum inflow and retained storage.
Run critical pump logic in the local PLC or RTU.
Transmit timestamps and stale-data status.
Buffer records during network outages.
Secure and supervise the gateway.
Test outage, failed pump and recovery end to end.

Frequently asked questions

Should cloud software start emergency drainage pumps?

Immediate protection should normally remain local because network availability and delay cannot be guaranteed.

What happens to data during an outage?

A suitable RTU or gateway buffers timestamped records and forwards them in order after reconnection.

Does restored communication prove the pumps were safe?

No. Local control, alarms and failure scenarios require separate end-to-end tests.

Need a project-specific review? Send process data, drawings, photographs and acceptance criteria through our contact page.