Construction of intelligent monitoring and leakage warning model of oil and gas pipeline
In view of the low accuracy and poor real-time of oil and gas pipeline monitoring and leakage warning,an intelligent oil and gas pipeline monitoring and leakage model is proposed based on support vector machine algorithm,and the model was constructed and the results were analyzed.The results showed that the performance test based on the least squares double support vector machine algo-rithm in the pressure signal test,the accuracy of the test set was 95％,and the classification speed was 0.049 8 seconds;in the flow sig-nal test,the accuracy rate was 92.5％,and the classification speed was 0.046 2 s.In the original signal,the pipeline pressure fluctuated from 0.083 to 0.159 MPa;after noise elimination,the pressure fluctuation range of the model was reduced to0.113-0.122 MPa.The results show that the research model can monitor the pressure change of oil and gas pipeline in real-time,and reduce the pressure fluctu-ation range and improve the stability of the pipeline through noise reduction signal processing technology.This research result provides a strong guarantee for the safe operation of oil and gas pipelines and has important practical significance.
oil and gas pipelineintelligent monitoringleakage warningsupport vector machine
NETL
NSTL
万方数据
