Pump

Sensors on the pump continuously measure vibration, temperature, pressure or flow and send the values via radio, IO-Link or cellular network to an IoT platform. There, software compares the readings with limit values or with a learned normal state. If the pump deviates, maintenance receives a warning before damage leads to downtime. In most cases this does not require any change to the plant's control system.

Vibration analysis reveals mechanical changes in a pump before they can be heard or seen. Imbalance, misalignment and bearing damage change the vibration pattern. Sensors on the bearing housing measure axial and radial vibration for this purpose. At an aquaculture centre in Singapore, Honeywell's system raised an ISO 10816-3 alarm because a seawater pump was far above the damage limit. Analysis of the frequency spectrum showed mechanical bearing play.

With IoT, pump maintenance becomes condition-based: maintenance teams plan work according to measured data rather than fixed intervals. CURRENTA used to service the pumps of its river water plant at regular intervals regardless of their condition, and failures still occurred unexpectedly. Today, LoRaWAN sensors measure temperature and vibration, and imbalance or bearing play is detected early. Root causes can be analysed using real operating data.

IoT shows how much energy a pump consumes in actual operation and provides the data for more efficient control. BESI Marine Systems controls pumps on board ships, which often run for hours, with a data-driven system and documents the CO₂ savings. In machining, the in.hub solution detects idle times and automatically switches machines, including their pumps, to standby.