​​

Smart Compressor monitoring system



Challenge Summary

An intelligent system to continuously monitor air compressors in oxygen plants, predict system failure, provide alerts and preventive measures there by reducing the production loss and down time of machine.

Challenge Scenario

Oxygen is one of the key components used in combustion, oxygen plants are used in the mining industry to produce high purity oxygen . Air compressors are one of the major units used in production of air at required pressure. The production of air at required pressure and efficiency is dependent on parameters such as power, compression method and various other parameters which are an integral part of the system. The variance in these parameters lead to low efficiency, failure of the compressor, high power consumption, downtime and production loss. An intelligent system to find anomalies in parameters, predict failure points and suggest suitable maintenance measures is required.

Profile of the End-User

Operations and maintenance team of Oxygen plant

Existing method : Currently vibration monitoring is done for Compressor through CBM

Gaps Currently vibration monitoring is done for Compressor through CBM

Functional Requirements of the End-User

  • Predict anomalies in the compressor parameters such as power, vibration, heat, filters and other parameters relative to the compressor.
  • Schedule maintenance and replacement of parts.

Functional & Operational Capabilities

  • Constant monitoring of parameters such as power, vibration, heat, filter status, lubrication, air flow.
  • Predictive Analytics Algorithms to predict the failure points and the time of failure of the compressor from assessing the necessary parameters.
  • Suggest suitable maintenance and replacement measures
  • Display all data to the maintenance team in a structured format.
  • Alert maintenance team incase of any kind of anomaly detected in the functioning of the machine

Operational constraints

  • Type of compressor
  • Setting of threshold for each parameters
  • Availability of clean data for each parameters
  • Provision to integrate sensors

Expected Tangible Benefits and Measurable Gains

  • Reduced downtime of machinery due to early prediction of failure points and time of failure.
  • Real time monitoring and data driven decision making.
  • Reduced production losses due to prior maintenance scheduling and suggestion of maintenance measures.

Performance Metrics or Outcomes

  • Accuracy of the system
  • Sensitivity of the sensors
  • Time lost in down time of machinery

​​​