Method and device for predicting corrosion growth in pipeline based on generalized additive model

Abstract

The invention provides an in-pipeline corrosion growth prediction method and device based on a generalized additive model. The method comprises the steps: screening independent variables, obtained inadvance, of an in-pipeline corrosion growth prediction model through a Lasso compression estimation algorithm; based on a generalized linear additive model, establishing an in-pipeline corrosion growth prediction model according to the screened independent variables and the in-pipeline corrosion growth rate; and predicting the in-pipeline corrosion growth rate according to the in-pipeline corrosion growth prediction model. According to the method, modeling can be carried out on the basis of existing corrosion data in the pipeline, and the future corrosion depth is accurately estimated. Therefore, the internal detection period can be determined, a maintenance plan can be made, and safe pipeline operation and cost saving are facilitated.

Description

technical field [0001] The invention relates to the technical field of oil and gas gathering and transportation, in particular to the field of oil and gas field development and oil-gas-water mixed transportation pipeline flow safety assurance, and in particular to a method and device for predicting internal corrosion growth in pipelines based on a generalized additive model. Background technique [0002] According to the Canadian National Energy Board, there are approximately 21,636 kilometers of oil pipelines and 55,982 kilometers of natural gas pipelines in the world. NACE (National Association of Corrosion Engineers International) conservatively estimates based on available data that as of 2015, the global investment related to pipeline corrosion has reached 2.5 trillion US dollars. Accidents caused by pipeline corrosion not only have an economic impact, but also threaten human safety and cause environmental damage. For example, on July 31, 2014, a leakage accident cause...

AI Technical Summary

Problems solved by technology

Typically, operators conduct a full inspection of pipelines every five years, so data on all of the different stages of corrosion are not available;

[0007] 2) The American Petroleum Institute has recorded more than 60 pipeline damage mechanisms, including carbon dioxide corrosion, sulfur corrosion and amino acid corrosion, and this type of model can only be established for one of the mechanisms at a time, so in order to accurately predict the corrosion growth of the entire pipeline, at least 60 models need to be built, not to mention that there are coupling effects between different mechanisms, thus increasing the complexity of the model and reducing the predictive efficiency of such models in industrial applications;

[0008] 3) Models based on stochastic processes require probability distribution sampling of unknown variables, which means that the modeler needs to have prior knowledge of the corrosion mechanism, so it is easy to introduce human error

Images