{"title":"Novel Valve Condition Prognostic System for Digitally Enabled High-Pressure Pump Maintenance","authors":"David Gerber, J. Thaduri","doi":"10.2523/iptc-22220-ea","DOIUrl":null,"url":null,"abstract":"\n The valve condition prognostics (VCP) system detects anomalies on high-pressure pump fluid-end valves and seats during fracturing before a total functional failure occurs. The VCP enables condition-based fluid-end replacement instead of time-based maintenance intervals, thereby minimizing downtime and maintenance cost and increasing asset utilization while eliminating permanent fluid-end damage due to operating with leaky valves. This is a step change when compared to the fixed-interval maintenance system.\n The VCP includes the programmable logic controller (PLC), analog and digital modules, a rotation monitoring encoder connected to the power end, and pressure transducers to monitor fluid-end discharge and suction pressures. The intelligent algorithms feature robust failure prediction criteria based on machine learning [1], pattern classification [2], and adaptive algorithms, applicable to various equipment and field conditions. By obtaining ongoing and accurate pressure signatures, the VCP detects warnings and alarms that are sent to the operators in real time for action. Based on the alarms and operating parameters, the pumps can be shut down automatically to prevent damage.\n During multiple field tests, the VCP successfully extended usable valve life by at least 45% when compared to our current fixed-interval maintenance method. The VCP is 100% accurate in detecting a catastrophic valve failure and avoided fluid-end damage. The VCP kits are easy to install onto existing pumps using existing discharge and suction pressure sensors. The data are sent to the cloud, and high-frequency data are recorded in the PLC for detailed analysis as needed.\n In contrast to the common replacement approach that is based on either a scheduled time interval or when an operational failure happens, the VCP can detect and notify when an anomaly occurs and performs maintenance only when necessary. The fixed-schedule maintenance approach replaces the valves and seats in a conservative fashion regardless of their condition, often leading to waste. In the failure-based maintenance situation, equipment damage often results, leading to devastating pump shut down and expensive fluid end replacement. The VCP addresses both challenges. It not only prevents prolonged and costly equipment failures, but also reduces downtime, valve and seat parts cost, and maintenance time.","PeriodicalId":11027,"journal":{"name":"Day 3 Wed, February 23, 2022","volume":"25 1","pages":""},"PeriodicalIF":0.0000,"publicationDate":"2022-02-21","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":"0","resultStr":null,"platform":"Semanticscholar","paperid":null,"PeriodicalName":"Day 3 Wed, February 23, 2022","FirstCategoryId":"1085","ListUrlMain":"https://doi.org/10.2523/iptc-22220-ea","RegionNum":0,"RegionCategory":null,"ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":null,"EPubDate":"","PubModel":"","JCR":"","JCRName":"","Score":null,"Total":0}
引用次数: 0
Abstract
The valve condition prognostics (VCP) system detects anomalies on high-pressure pump fluid-end valves and seats during fracturing before a total functional failure occurs. The VCP enables condition-based fluid-end replacement instead of time-based maintenance intervals, thereby minimizing downtime and maintenance cost and increasing asset utilization while eliminating permanent fluid-end damage due to operating with leaky valves. This is a step change when compared to the fixed-interval maintenance system.
The VCP includes the programmable logic controller (PLC), analog and digital modules, a rotation monitoring encoder connected to the power end, and pressure transducers to monitor fluid-end discharge and suction pressures. The intelligent algorithms feature robust failure prediction criteria based on machine learning [1], pattern classification [2], and adaptive algorithms, applicable to various equipment and field conditions. By obtaining ongoing and accurate pressure signatures, the VCP detects warnings and alarms that are sent to the operators in real time for action. Based on the alarms and operating parameters, the pumps can be shut down automatically to prevent damage.
During multiple field tests, the VCP successfully extended usable valve life by at least 45% when compared to our current fixed-interval maintenance method. The VCP is 100% accurate in detecting a catastrophic valve failure and avoided fluid-end damage. The VCP kits are easy to install onto existing pumps using existing discharge and suction pressure sensors. The data are sent to the cloud, and high-frequency data are recorded in the PLC for detailed analysis as needed.
In contrast to the common replacement approach that is based on either a scheduled time interval or when an operational failure happens, the VCP can detect and notify when an anomaly occurs and performs maintenance only when necessary. The fixed-schedule maintenance approach replaces the valves and seats in a conservative fashion regardless of their condition, often leading to waste. In the failure-based maintenance situation, equipment damage often results, leading to devastating pump shut down and expensive fluid end replacement. The VCP addresses both challenges. It not only prevents prolonged and costly equipment failures, but also reduces downtime, valve and seat parts cost, and maintenance time.