The presence of air bubbles inside an internal gears pump usually leads to fast pump damage and breaking. When bubbles enter the high pressure chamber they may implode releasing pressure waves in the fluid that ultimately detach material from the gears and other crucial pump parts. This paper recalls and describes the activities performed in order to develop a condition monitoring system able to indicate the presence of air bubbles in an internal gear pump that feeds the hydraulic system of a packaging machine: this pump experienced a breakdown after only 3 minutes of air bubbles flowing in it. Since the machine is supposed to work continuously, an unexpected stop has heavy consequences in terms of loss of production time: therefore the machine producer decided to implement the condition monitoring (CM) of the pump. The producer’s goal is to use the warnings/alarms generated by the CM algorithm as a supplemental machine control signal: this may eventually command the stop of the machine to preserve it (and consequently the machine functionality) from fast damaging. The main phases of the development had been: data recording and analysis, diagnostic parameters identification, algorithm development and final algorithm validation in field tests. The data recording campaign included the simultaneous acquisition of 14 quantities of 5 different kinds (vibration, pressure, temperature, electrical torque and angular speed). The data analysis highlighted one of the vibration signals as the most significant to be monitored. Several signal features were analysed from the in order to evaluate their diagnostic capability point of view (i.e.: the ability to detect the presence of air bubbles in the pump); five features are used by the CM algorithm for the bubbles warning/alarm generation, these being the RMS, the bandpass filtered signal RMS, the Signal Entropy, the Spectral Mean Square Error and the Spectral Cumulative Difference. Currently, the algorithm is being tested and validated on recorded data and a field test campaign is being scheduled
Developing of a monitoring system for air bubbles detection in an internal gear pump / Cotogno, Michele; Cocconcelli, Marco; Rubini, Riccardo. - ELETTRONICO. - (2013), pp. 1-9. (Intervento presentato al convegno The International Congress Surveillance 7 tenutosi a Chartres (France) nel 29-30 october 2013).
Developing of a monitoring system for air bubbles detection in an internal gear pump
COTOGNO, MICHELE;COCCONCELLI, Marco;RUBINI, Riccardo
2013
Abstract
The presence of air bubbles inside an internal gears pump usually leads to fast pump damage and breaking. When bubbles enter the high pressure chamber they may implode releasing pressure waves in the fluid that ultimately detach material from the gears and other crucial pump parts. This paper recalls and describes the activities performed in order to develop a condition monitoring system able to indicate the presence of air bubbles in an internal gear pump that feeds the hydraulic system of a packaging machine: this pump experienced a breakdown after only 3 minutes of air bubbles flowing in it. Since the machine is supposed to work continuously, an unexpected stop has heavy consequences in terms of loss of production time: therefore the machine producer decided to implement the condition monitoring (CM) of the pump. The producer’s goal is to use the warnings/alarms generated by the CM algorithm as a supplemental machine control signal: this may eventually command the stop of the machine to preserve it (and consequently the machine functionality) from fast damaging. The main phases of the development had been: data recording and analysis, diagnostic parameters identification, algorithm development and final algorithm validation in field tests. The data recording campaign included the simultaneous acquisition of 14 quantities of 5 different kinds (vibration, pressure, temperature, electrical torque and angular speed). The data analysis highlighted one of the vibration signals as the most significant to be monitored. Several signal features were analysed from the in order to evaluate their diagnostic capability point of view (i.e.: the ability to detect the presence of air bubbles in the pump); five features are used by the CM algorithm for the bubbles warning/alarm generation, these being the RMS, the bandpass filtered signal RMS, the Signal Entropy, the Spectral Mean Square Error and the Spectral Cumulative Difference. Currently, the algorithm is being tested and validated on recorded data and a field test campaign is being scheduled
Pubblicazioni consigliate
I metadati presenti in IRIS UNIMORE sono rilasciati con licenza Creative Commons CC0 1.0 Universal, mentre i file delle pubblicazioni sono rilasciati con licenza Attribuzione 4.0 Internazionale (CC BY 4.0), salvo diversa indicazione.
In caso di violazione di copyright, contattare Supporto Iris
