The Jereh Pump Condition Monitoring System (JPCM) is specifically designed for frac pumps.  It enables real-time monitoring of pump performance and health, while its predictive analytics drastically reduces NPT.

System Composition

The JPCM system enables on-site engineers monitor in real-time, view and analyze equipment status and early warnings in the data vas through the wired or wireless local area network at the wellsite. It also supports remote data transmission to customer’s office for real-time review, complex diagnosis and data management.

Hardware

Sensors (e.g. vibration, temperature, keyphasor)

Data acquisition device

Edge computer

Gateway

Data server

Software

Configuration software modules

Data processing and communication modules

Condition monitoring and fault early warning modules

Customized chart modules

JPCM Key Features and Benefits

Real-Time Data Collection

Sensors and data acquisition devices allow high-frequency, multi-channel, real-time synchronized data collection from the plunger pump's gearbox, crankcase, and fluid end. This provides robust data support for condition monitoring and early fault warning.

User Data Security

The results are displayed in Data Van via local collection, local analysis and processing.

Condition Monitoring and Early Fault Warning

Multi-dimensional symptom parameters are established by integrating fault mechanism models with deep learning models, enabling equipment condition grading, common fault diagnosis and precise fault localization.

Real-time monitoring of the gearbox, crankshaft, and fluid end to assess pump condition.

Diagnosis of minor faults in gearbox, power end and crosshead; detection of leakage faults in valve and packing, and the location of faults in cylinders. This tracking of fault trends enables on-site engineers to take necessary measures to prevent significant damage to critical components.

Analytics

A variety of charts are available, including time domain plots, spectrograms, demodulation spectra, angular domain charts, and multi-trend charts. These provide powerful analytics tools for complex fault diagnosis, through which the fault diagnosis engineers can conduct in-depth analysis of potential issues for accurate diagnostic and maintenance decisions.

Visualization of Pump Condition and Fault Early Warning

An intuitive visual interface allows real-time monitoring of the pump's operating condition. This feature not only helps to easily understand the pump's condition, but also allows quick access to fault diagnosis and location through an alarm statistics interface.

Predictive Maintenance Mode

Upgrade from planned to predictive maintenance, shifting from periodic to predicted fault-based replacement

This reduces spare part waste, labor loss, and unexpected downtime, enhancing maintenance and production efficiency

Precision Monitoring and Rapid Positioning

Pump condition is classified into normal, minor, moderate, and serious for real-time tracking

Fast fault diagnosis improves maintenance efficiency and accuracy.

Hundreds of Millions of Parameter Models

Transformer deep learning processes tens of millions of data points to train a powerful hundreds-of-millions-parameter model.

With over 97% fault recognition accuracy, the system keeps improving through continuous training.