A Pipeline Remaining Life Prediction Method Based on Inverse Gaussian Process and em-pf

Abstract

The invention discloses a pipeline remaining life predication method based on an inverse Gaussian process and EM-PF. The pipeline remaining life predication method comprises the following steps: acquiring corrosion data of an oil gas pipeline; according to the corrosion data of the oil gas pipeline, describing an erosion and degradation process of the oil gas pipeline through utilizing a random process-inverse Gaussian process; establishing a state equation and an observation equation of an oil gas pipeline state space model; establishing an inverse Gaussian-state erosion and degradation model according to the state equation and the observation equation of the oil gas pipeline state space model; and then predicating the remaining life of the oil gas pipeline according to the inverse Gaussian-state erosion and degradation model. The method disclosed by the invention can be used for accurately and rapidly predicating the remaining life of the oil gas pipeline.

Description

technical field [0001] The invention relates to a method for predicting the remaining life, in particular to a method for predicting the remaining life of a pipeline based on an inverse Gaussian process and EM-PF. Background technique [0002] As the main gathering and transportation equipment in oil and gas storage and transportation, oil and gas pipelines play an important role. As the oil and gas pipeline continues to extend to the seabed, the factors affecting its corrosion and aging and the environmental conditions become more and more complex, and the potential safety hazards increase accordingly. The remaining life prediction and maintenance of oil and gas pipelines is a key technology in its integrity management, which can reduce the probability of occurrence of risks and provide a basis for optimizing maintenance programs. [0003] Foreign J.L.Alamilla, A.Valor et al. studied the pipeline corrosion degradation model based on stochastic process; Gasperin et al. stud...

AI Technical Summary

Problems solved by technology

Peng W established a degradation model under the inverse Gaussian process model, and then analyzed its reliability; X.Wang studied the corrosion degradation model based on the inverse Gaussian process; Zhang Zenggang quantified the attenuation degree of the structural performance of corroded oil and gas pipelines, and introduced probability theory into its corrosion evaluation field; Zhao Xinwei et al. established a distribution model of pipeline pitting damage along the wall thickness direction, and proposed a prediction method for the remaining life of pipelines with pitting corrosion diffuse damage; Wang established a stochastic degradation process assuming that the model parameters obey the Wiener distribution; Fan Hongdong et al. established on the basis of literature Nonlinear function degradation model, and then estimate the remaining life; Ren Shuhong et al. proposed a remaining life prediction method based on performance degradation to improve the reliability of thermal aging equipment; Si Xiaosheng, etc. established a nonlinear degradation process model with measurement errors to estimate the remaining life; Zhang Xinghui and others researched equipment degradation state identification and remaining service life prediction based on the mixed Gaussian output Bayesian belief network model; Chen Liang established a corrosion degradation model based on the gamma process, and studied the influence of measurement errors on the estimated remaining life; Yuan Geng adopted strategies to improve The algorithm solves the Markov process, analyzes the pipeline maintenance measures and corresponding costs, and obtains the optimal pipeline maintenance strategy; Zhang Xinsheng et al. establish a remaining life prediction model considering the maintenance depth based on the gamma process, and finally realize the most economical and economical periodical maintenance of the oil and gas pipeline corrosion system. optimization, however, there is no method for accurately predicting the remaining life of oil and gas pipelines in the above technologies, so the reliability failure time data of oil and gas pipelines is difficult to obtain in a short time

Images