Fracturing Pump Valve Life Prediction Using Speed-Normalized Vibration
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Hydraulic fracturing rig pump valves experience wear and tear, leading to unpredictable useful life and increased downtime due to the difficulty in accurately predicting their condition based on operating conditions, resulting in unnecessary storage of spare valves and increased costs.
Innovation Solution
A rig management system that gathers vibration and pump speed data from sensors, normalizes it, and uses statistical analysis and machine learning models to determine the remaining useful life of the valve, scheduling maintenance or replacement operations as needed.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Measurement precision
If traditional monitoring methods are used to detect pump valve conditions, then the system structure remains simple, but the prediction accuracy of valve useful life is insufficient leading to unexpected failures
Solution Approach 1:
The monitoring system is segmented into multiple independent sensor modules (vibration sensors, temperature sensors, pressure sensors) that collect different types of data separately. This segmentation allows the system to achieve comprehensive monitoring accuracy while keeping each sensor module simple and modular, resolving the contradiction between prediction accuracy and system complexity.
Solution Approach 2:
A machine learning model serves as an intermediary between the raw sensor data and the useful life prediction. The model processes multiple sensor inputs and transforms them into accurate predictions, enabling high measurement precision without requiring complex manual analysis systems, thus resolving the contradiction between accuracy and system complexity.
2Reliability
If spare valves are stored at the worksite to prevent downtime, then the reliability of the hydraulic fracturing rig is improved, but the storage costs and space requirements increase
Solution Approach 1:
The system performs preliminary monitoring and prediction of valve conditions before actual failure occurs. By detecting early signs of degradation through sensor data analysis, the system allows for proactive maintenance scheduling, eliminating the need to store multiple spare valves while maintaining high operational reliability through timely interventions.
Solution Approach 2:
The monitoring system enables the valve to essentially monitor its own condition through integrated sensors that detect vibration, temperature, and pressure changes. This self-diagnostic capability allows the system to determine when maintenance is needed without external inspection, reducing the need for spare inventory while maintaining reliability.
3Measurement precision
If vibration data is collected without normalization to pump speed, then the data collection process is simple, but the trend identification accuracy is compromised
Solution Approach 1:
The system normalizes vibration data by changing the reference parameter from absolute vibration values to vibration values relative to pump speed. This parameter transformation (dividing vibration data by corresponding pump speed data) enables accurate trend identification across varying operating conditions while using straightforward mathematical processing rather than complex analysis methods.
Data Source
AI summary
A rig management system is disclosed. The rig management system may gather, from a set of sensors, data that includes: vibration data related to vibration of a valve associated with a pump of a hydraulic fracturing rig during operation of the pump, and pump speed data related to a speed of the pump during operation of the pump. The rig management system may process the data to identify a trend related to the vibration data, wherein the vibration data has been normalized based on the pump speed data. The rig management system may determine a remaining useful life of the valve based on the trend and threshold data that identifies a set of thresholds related to the vibration of the valve. The rig management system may determine that the remaining useful life satisfies a threshold. The rig management system may schedule an operation to be performed on the valve.