Monitoring of the electrical and mechanical performance of a high-pressure pump of a desalination plant

Aingura insights

Preprocessing Preprocessing
Transformation Transformation
Data Acquisition Data Acquisition

Sectors

Utilities Utilities

Applications

Consumption Efficiency Consumption Efficiency
Maintenance Maintenance

impact

  • Generation of an explainable model that represents the dynamic behavior of the electrical and mechanical system of the pump.
  • Use of the estimated modal parameters to generate a probabilistic machine learning baseline that produces a KPI of the healthy state of the pump’s operation.
  • Use of high-speed measurement of power consumption (8 kHz) to analyze both mechanical and electrical behavior, minimizing the number of sensors required and increasing their robustness to the environment

Developed AI Technologies

  • Machine Learning-based electromechanical baseline.
  • High-speed KPI generation for highly dynamic systems.

Challenges

  • Avoidance of unscheduled shutdowns, mainly to avoid interruptions in the supply of drinking water.
  • Improvement of energy consumption.
  • Development of maintenance procedures based on behavioral deviation analysis to avoid corrective maintenance and minimize preventive ones

results

  • Reduction of unscheduled shutdowns (>85%), allowing the reduction of the redundancy pumps.
  • Reduction of preventive maintenance cost (50%)
  • Reduction of energy consumptions (8%).
  • PRI: 2 years

Contact us