Solution of Pump Fault Prediction System
Problems In Pump equipment
In the pump rotating equipment industry, the industrial chain consists of pump manufacturers, OEMs, and end users. Generally, pump manufacturers hope to solve problems related to after-sales operation and maintenance efficiency and maintenance costs; to ensure the market application performance of the product; and the health of the pump at the customer's site. The main pump manufacturer hopes to reduce the maintenance cost of the core components of the main equipment, and to understand the operating health of the main equipment on the customer's site in time. End users hope to solve problems related to downtime and production, real-time operation and health of equipment, and personal safety hazards caused by abnormal equipment production.
According to the problems encountered in the pump industry,Witium launched the "pump fault prediction system solution", which carries out the health status analysis and early warning fault analysis of the pump and host equipment for the monitoring of the pump operation status on the user's site, forms a professional pump fault prediction and health diagnosis report, and gives maintenance suggestions. Witium's pump failure prediction solution brings great commercial value to users:
Improve operation and maintenance efficiency and reduce equipment maintenance cost;
Reduce abnormal faults and equipment damage cost;
Predict equipment deterioration and reduce the cost of spare parts;
Reduce failure probability and avoid production line shutdown loss;
Assess equipment risks and prevent potential personal safety hazards;
Apply data analysis to improve equipment quality and process;
The pump fault prediction solution uses the high-frequency vibration sensor of the Internet of things to collect the vibration waveform samples and surface temperature of the equipment, and generate the vibration characteristic samples of the pump equipment; The edge computing gateway collects the three-phase winding temperature and lubricating oil temperature of the pump through the platinum resistance to understand the operating load state of the pump. Combined with the vibration characteristics of the sensor, the intelligent mechanism model algorithm is used to analyze the sampled data, so as to predict the pump equipment faults such as imbalance, misalignment, bearing fault, cavitation, impeller imbalance, looseness and gear wear; Then upload the data samples and analysis learning results to the witcloud Internet of things data cloud platform through mqtt; The SaaS business cloud platform of the solution provides users with status monitoring, scene monitoring, fault management, operation and maintenance management, equipment management, customer management, authority management and other functions; It also provides data interface function to facilitate users to import equipment status into intelligent factory operation and maintenance platform system. At the same time, through big data analysis tools, the business cloud platform provides professional users with pump equipment operation status analysis, early warning fault analysis, and forms professional pump equipment health diagnosis reports and maintenance suggestions.
Witcloud Internet of things data cloud platform
The functions of witcloud cloud platform include: access configuration of vibration sensor and edge computing gateway on pump equipment (including subject configuration and channel configuration), remote information configuration and equipment firmware upgrade, alarm configuration, journal recording and viewing, multi-layer edge computing gateway upload, database backup and retrieval, management and control of influxdb database.
The powerful system architecture of the witcloud cloud platform has scalability, the ability to access a large number of devices and million level concurrency. It can easily access more than 1 million pump devices and support the access ability of more than 100000 pump devices every year.
SaaS service cloud platform for pump equipment
The SaaS service cloud platform for pump fault prediction provides users with platform functions such as status monitoring, fault analysis, scene monitoring, fault management, operation and maintenance management, equipment management, customer management and authority management, as well as platform data reading interface function. Through big data analysis tools, SaaS business cloud platform provides professional users with equipment operation status analysis, early warning fault analysis, and forms professional industrial equipment health diagnosis reports and maintenance suggestions.
Fault prediction algorithm model: firstly, the system collects the vibration characteristics and load temperature characteristics of pump equipment through high-frequency vibration sensor and platinum resistance of Internet of things, and forms a vibration characteristic sample library; Then the intelligent algorithm (machine learning algorithm library) is used to carry out machine self-learning on these original data samples, generate the corresponding fault prediction algorithm model, and form the fault prediction algorithm model library; The fault prediction algorithm model is used to analyze and process the real-time collected data to realize the fault prediction and diagnosis of pump equipment, such as imbalance, misalignment, bearing fault, cavitation, impeller imbalance, looseness, gear wear and so on.
Solution Web Side Software Diagram
High frequency vibration sensor
MEMS vibration sensor
Medium and high frequency vibration monitoring
The sampling frequency is up to 26khz
Magnetic suction easy installation
Edge computing gateway
Edge AI Calculation
Several vibrating channels
Multiple wireless Transmit
Embedded algorithm model
Number of vibration channels
1-6 Optional Channels
1 Pt100 Breakage Detection
Three winding temperature
3 Pt100 Breakage Detection
1 * power light, 1 * sensor communication light, 1 * network status light
Online update, remote update
Solution Hardware Diagram
Site Map Of Successful Cases