Analyzing the operation of a heat recovery boiler
Target

To develop a predictive analysis model for a waste heat boiler, to determine the parameters characterizing the waste heat boiler’s state, and to detect abnormal conditions and work defects.

Solution

The use of Clover PMM — a system for diagnosing and predicting the technical condition of equipment.

Initial data: the certificate and the operating instructions for the waste heat boiler, the log of defects and technological violations, and the telemetry data from the operation of the waste heat boiler for one year.

Sequence of actions

  • decided to look for potential leaks or fumes in high, medium, and low-pressure circuits
  • selected the parameters of interest from the sample provided
  • performed an analysis of the change in the values of the waste heat boiler’s parameters
  • estimated noise in the parameter values
  • searched for potential leaks or fumes using mathematical algorithms for linear regression and data integration
  • compared the obtained results with the fault log records and the power unit’s technological violations

Result
147 anomalous deviations in substance balance were detected; 18 of them were registered as stops and start-ups of the waste heat boiler. The remaining 129 were anomalies in the operation of the waste heat boiler’s high-pressure circuit, of which 15 corresponded to defects registered in the log of defects and technological violations.
37 abnormal balance deviations were detected on the low-pressure circuit, of which 18 were registered as stops and start-ups of the recovery boiler. The rest were anomalies in the operation of the waste heat boiler’s low-pressure circuit.
The model constructed determined the anomalous states in the form of possible leaks and fumes in the high-pressure circuit during operation of the waste heat boiler, while the process control system did not identify these anomalies.
Do you want to improve the efficiency of your company?
Learn more about the possibilities of intelligent analytics and predictive analysis
* Marked fields are required
Thank you!
A Clover Group specialist will contact you during business hours.