Have a personal or library account? Click to login
Research on Fault Diagnosis System of IOT for Oil Well Pump Based on Machine Learning Cover

Research on Fault Diagnosis System of IOT for Oil Well Pump Based on Machine Learning

Open Access
|Mar 2024

Figures & Tables

Figure 1.

Organization structure of the system based on internet of things technology
Organization structure of the system based on internet of things technology

Figure 2.

Hardware components theory block diagram of the fuel pump sensor instrument node
Hardware components theory block diagram of the fuel pump sensor instrument node

Figure 3.

Hardware components theory block diagram of the IOT aggregation instrument node
Hardware components theory block diagram of the IOT aggregation instrument node

Figure 4.

MCU control program flow chart of the IOT aggregation instrument node
MCU control program flow chart of the IOT aggregation instrument node

Figure 5.

Module structure diagram of the IOT center computer monitor system software
Module structure diagram of the IOT center computer monitor system software

Figure 6.

The structure of the expert system for oil well pump LSTM machine learning fault diagnosis
The structure of the expert system for oil well pump LSTM machine learning fault diagnosis

Figure 7.

Neural network module structure diagram
Neural network module structure diagram

Figure 8.

LSTM machine learning model structure principle
LSTM machine learning model structure principle

Figure 9.

Fault diagnosis and prediction scatter diagram of oil well pump oil delivery pressure training set
Fault diagnosis and prediction scatter diagram of oil well pump oil delivery pressure training set

Figure 10.

Oil well pump oil delivery pressure test set fault diagnosis and prediction scatter diagram
Oil well pump oil delivery pressure test set fault diagnosis and prediction scatter diagram

Figure 11.

Comparison chart of machine learning actual value and predicted value of oil delivery pressure of oil well delivery pump
Comparison chart of machine learning actual value and predicted value of oil delivery pressure of oil well delivery pump

Evaluation index data obtained by machine learning operations

TypeTrainTest
MAE1.915.59
MSE9.3311.35
RMSE5.427.41
nRMSE (%)4.038.85

Top 20 records in the data set used by oil well pump machine learning

CodeDateTimeOil Pressure (Mpa)
16/10:001.69
26/11:001.81
36/12.002
46/13001.97
56/14:001.84
66/15:002.01
76/16:001.62
86/17:001.89
96/18:002.14
106/19.002
116/110:002.06
126/111001.8
136/112:001.65
146/113:001.67
156/114:002.11
166/115.001.88
176/116:002.12
186/117:001.54
196/118:002.08
206/119:002.02
Language: English
Page range: 10 - 25
Published on: Mar 15, 2024
Published by: Xi’an Technological University
In partnership with: Paradigm Publishing Services
Publication frequency: 4 issues per year

© 2024 Bozhi Xiao, Yangbing Lu, Miao Wang, Haojie Li, Hengyan Zhu, Xuan Cao, Shengquan Yang, published by Xi’an Technological University
This work is licensed under the Creative Commons Attribution-NonCommercial-NoDerivatives 4.0 License.