论文部分内容阅读
Increasingly serious leak problem in pipeline transportation has not only affected the operation of pipelines but also caused loss of precious resource and environmental damage. Based on the analysis of the occurrence of negative pressure waves and the unsupervised learning of pattern recognition, the Interactive Self-organizing Data Analysis Technique Algorithm (ISODATA) method was used to classify the negative pressure waves and then the states of pipelines could be determined. K_L transformation was used to eliminate the correlativity of feature parameters and to reduce the dimensionality of feature vector space to speed up calculation. Experimental results validated the accuracy and practical value of this method.
Increasingly serious leak problem in pipeline transportation has not only affected the operation of pipelines but also caused loss of precious resource and environmental damage. Based on the analysis of the occurrence of negative pressure waves and the unsupervised learning of pattern recognition, the Interactive Self-organizing Data Analysis Technique Algorithm (ISODATA) method was used to classify the negative pressure waves and then the states of pipelines could be determined. K_L transformation was used to eliminate the correlativity of feature parameters and to reduce the dimensionality of feature vector space to speed up calculation . Experimental results validated the accuracy and practical value of this method.