Coated Steel Pipe Collapse Strength Prediction Using Neural Networks
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Current methods for predicting the collapse strength of coated steel pipes under external pressure bending, such as those used in submarine pipelines, are inaccurate due to neglecting pipe-making strain and coating conditions, leading to potential structural damage or failure.
Innovation Solution
A steel pipe manufacturing method that utilizes a neural network-based prediction model incorporating pipe shape, strength characteristics, pipe-making strain, coating conditions, and bending strain during construction to accurately forecast collapse strength under external pressure bending.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Measurement precision
If the estimation equation from NPL 1 is used to predict collapse strength, then the prediction can be performed with simple parameters, but the prediction accuracy is insufficient because pipe-making strain and coating conditions are not considered
Solution Approach 1:
The patent transforms the prediction model from a simple analytical equation to a neural network model that processes multiple input parameters including pipe-making strain, coating conditions, steel pipe shape, and strength characteristics. This parameter expansion enables accurate prediction of coated steel pipe collapse strength while maintaining computational efficiency through the neural network architecture.
Solution Approach 2:
The patent replaces the traditional mechanical/analytical prediction approach with a computational neural network model. This substitution allows the system to learn complex non-linear relationships between multiple input parameters and collapse strength from training data, achieving high accuracy without requiring complex analytical derivations.
2Reliability
If the prediction does not consider pipe-making strain and coating conditions, then the manufacturing process is simpler, but the collapse strength prediction fails to match actual measured values
Solution Approach 1:
The patent incorporates pipe-making strain and coating conditions as input parameters into the neural network model before prediction. By preliminarily capturing these critical manufacturing parameters, the model can accurately predict collapse strength without requiring complex post-manufacturing adjustments or additional testing procedures.
3Productivity
If simple estimation equation is used, then the calculation is faster and easier, but the predicted collapse strength may lead to excessively safe design or collapse at lower pressure
Solution Approach 1:
The patent replaces slow and imprecise analytical calculations with a neural network model that provides rapid predictions. The neural network, once trained, can instantly compute collapse strength for new inputs, maintaining calculation efficiency while dramatically improving prediction accuracy and design safety.
Data Source
Figure 1
Figure 2
Figure 3
AI summary
Provided are a steel pipe collapse strength prediction model generation method, a steel pipe collapse strength prediction method, a steel pipe manufacturing characteristics determination method, and a steel pipe manufacturing method capable of highly accurately predicting the collapse strength under external pressure bending of a coated steel pipe coated after steel pipe forming in consideration of the pipe-making strain during steel pipe forming and coating conditions as well as the bending strain during construction. Into a steel pipe collapse strength prediction model generated by the steel pipe collapse strength prediction model generation method, steel pipe manufacturing characteristics including the steel pipe shape of a coated steel pipe to be predicted after steel pipe forming, steel pipe strength characteristics after steel pipe forming, the pipe-making strain during steel pipe forming, coating conditions, and the bending strain during construction are input to predict the collapse strength under pressure bending of the coated steel pipe (step S1 to step S5).