# US 12,465,967 B2 · Steel pipe collapse strength prediction model generation method, steel pipe collapse strength prediction method, steel pipe manufacturing characteristics determination method, and steel pipe manufacturing method

> Agents: this data is available as JSON from the patentgrep API and MCP server — see https://patentgrep.com/llms.txt

- Number: US 12,465,967 B2
- Document ID: US12465967B2
- Status: Granted
- Kind: B2
- Application number: 17/925,762
- Filed: 2021-02-08
- Published: 2023-06-22
- Granted: 2025-11-11
- Earliest claimed: 2020-05-26
- Assignees: Jfe Steel Corporation
- Applicants: Jfe Steel Corporation
- Inventors: Takahiro Sakimoto; Hisakazu Tajika; Tsunehisa Handa
- Web page: https://patentgrep.com/patent/US12465967B2

## CPC

- B21C 37/08: Making tubes with welded or soldered seams {(with helically arranged seams B21C37/122)}
- B21C 37/06: of tubes or metal hoses; Combined procedures for making tubes, e.g. for making multi-wall tubes
- G01N 3/00: Investigating strength properties of solid materials by application of mechanical stress
- G01N 3/08: by applying steady tensile or compressive forces (G01N3/28 takes precedence)
- G01N 2203/0016: Tensile or compressive
- G01N 2203/0067: Fracture or rupture
- G01N 2203/0075: Strain-stress relations or elastic constants
- G01N 2203/0216: Finite elements
- G01N 2203/0274: Tubular or ring-shaped specimens
- G01N 2203/0298: Manufacturing or preparing specimens
- G05B 13/027: {using neural networks only}
- G05B 19/41875: {characterised by quality surveillance of production}
- G05B 2219/32193: Ann, neural base quality management
- G06F 30/10: Geometric CAD
- G06F 30/27: using machine learning, e.g. artificial intelligence, neural networks, support vector machines [SVM] or training a model
- G06N 3/04: Architecture, e.g. interconnection topology
- G06N 3/08: Learning methods
- G06N 20/00: Machine learning
- B21C: MANUFACTURE OF METAL SHEETS, WIRE, RODS, TUBES, PROFILES OR LIKE SEMI-MANUFACTURED PRODUCTS OTHERWISE THAN BY ROLLING; AUXILIARY OPERATIONS USED IN CONNECTION WITH METAL-WORKING WITHOUT ESSENTIALLY REMOVING MATERIAL
- G01N: INVESTIGATING OR ANALYSING MATERIALS BY DETERMINING THEIR CHEMICAL OR PHYSICAL PROPERTIES (measuring or testing apparatus or processes other than immunoassay, involving enzymes or microorganisms C12M, C12Q)
- G05B: CONTROL OR REGULATING SYSTEMS IN GENERAL; FUNCTIONAL ELEMENTS OF SUCH SYSTEMS; MONITORING OR TESTING ARRANGEMENTS FOR SUCH SYSTEMS OR ELEMENTS (systems for controlling or regulating non-electric variables G05D; systems for regulating electric or magnetic variables G05F; control devices or systems insofar as characterised by mechanical features only G05G)
- G06F: ELECTRIC DIGITAL DATA PROCESSING (computer systems based on specific computational models G06N)
- G06N: COMPUTING ARRANGEMENTS BASED ON SPECIFIC COMPUTATIONAL MODELS

## Abstract

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 of a steel pipe after forming or a coated steel pipe in consideration of the pipe-making strain during forming. Into a steel pipe collapse strength prediction model generated by the prediction model generation method, steel pipe manufacturing characteristics including the shape of a steel pipe to be predicted after forming, strength characteristics, and the pipe-making strain are input to predict the collapse strength after forming. Into a steel pipe collapse strength prediction model, steel pipe manufacturing characteristics including the shape of a coated steel pipe to be predicted after forming, strength characteristics, the pipe-making strain, and coating conditions are input to predict the collapse strength of the coated steel pipe.

## Claims

### Claim 1

A steel pipe manufacturing method comprising:
performing machine learning of a plurality of learning data that include, as an input datum, a previous steel pipe manufacturing characteristic including a steel pipe shape after steel pipe forming, a steel pipe strength characteristic after steel pipe forming, and a pipe-making strain during steel pipe forming and, as an output datum for the input datum, a previous collapse strength of a steel pipe after steel pipe forming, to generate a steel pipe collapse strength prediction model that predicts a steel pipe collapse strength after steel pipe forming;
forming a steel pipe;
predicting a collapse strength of the formed steel pipe by inputting, into the steel pipe collapse strength prediction model, a steel pipe manufacturing characteristic including a steel pipe shape of the formed steel pipe, a steel pipe strength characteristic of the formed steel pipe, and a pipe-making strain during steel pipe forming, to predict a steel pipe collapse strength of the formed steel pipe; and
assigning the predicted steel pipe collapse strength to the formed steel pipe.

### Claim 2 (depends on 1)

The steel pipe manufacturing method according to claim 1, wherein the steel pipe collapse strength prediction model is constructed by a neural network.

### Claim 3

A steel pipe manufacturing method comprising:
performing machine learning of a plurality of learning data that include, as an input datum, a previous steel pipe manufacturing characteristic including a steel pipe shape after steel pipe forming, a steel pipe strength characteristic after steel pipe forming, and a pipe-making strain during steel pipe forming and, as an output datum for the input datum, a previous collapse strength of a steel pipe after steel pipe forming, to generate a steel pipe collapse strength prediction model that predicts a steel pipe collapse strength after steel pipe forming;
inputting, into the steel pipe collapse strength prediction model, a steel pipe manufacturing characteristic including a steel pipe shape of a steel pipe to be predicted after steel pipe forming, a steel pipe strength characteristic after steel pipe forming, and a pipe-making strain during steel pipe forming, to predict a steel pipe collapse strength after steel pipe forming;
sequentially changing at least one of the steel pipe shape after steel pipe forming, the steel pipe strength characteristic after steel pipe forming, and the pipe-making strain during steel pipe forming included in the steel pipe manufacturing characteristic such that the predicted steel pipe collapse strength after steel pipe forming asymptotically approaches a requested collapse strength of an intended steel pipe after steel pipe forming, to determine an optimum steel pipe manufacturing characteristic;
determining a steel pipe manufacturing condition in accordance with the determined optimum steel pipe manufacturing characteristic; and
manufacturing a steel pipe under the determined steel pipe manufacturing condition.

### Claim 4 (depends on 3)

The steel pipe manufacturing method according to claim 3, wherein the steel pipe collapse strength prediction model is constructed by a neural network.

### Claim 5

A steel pipe manufacturing method comprising:
performing machine learning of a plurality of learning data that include, as an input datum, a previous steel pipe manufacturing characteristic including a steel pipe shape after steel pipe forming, a steel pipe strength characteristic after steel pipe forming, a pipe-making strain during steel pipe forming, and a coating condition and, as an output datum for the input datum, a previous collapse strength of a coated steel pipe coated after steel pipe forming, to generate a steel pipe collapse strength prediction model that predicts a collapse strength of a coated steel pipe coated after steel pipe forming;
forming a steel pipe and coating the formed steel pipe to form a coated steel pipe;
predicting a collapse strength of the coated steel pipe by inputting, into the steel pipe collapse strength prediction model, a steel pipe manufacturing characteristic including a steel pipe shape of the coated steel pipe, a steel pipe strength characteristic of the coated steel pipe, a pipe-making strain during steel pipe forming, and a coating condition, to predict a collapse strength of the coated steel pipe; and
assigning the predicted coated steel pipe collapse strength to the coated steel pipe.

### Claim 6 (depends on 5)

The steel pipe manufacturing method according to claim 5, wherein the steel pipe collapse strength prediction model is constructed by a neural network.

### Claim 7

A steel pipe manufacturing method comprising:
performing machine learning of a plurality of learning data that include, as an input datum, a previous steel pipe manufacturing characteristic including a steel pipe shape after steel pipe forming, a steel pipe strength characteristic after steel pipe forming, a pipe-making strain during steel pipe forming, and a coating condition and, as an output datum for the input datum, a previous collapse strength of a coated steel pipe coated after steel pipe forming, to generate a steel pipe collapse strength prediction model that predicts a collapse strength of a coated steel pipe coated after steel pipe forming;
inputting, into the steel pipe collapse strength prediction model, a steel pipe manufacturing characteristic including a steel pipe shape of a coated steel pipe to be predicted after steel pipe forming, a steel pipe strength characteristic after steel pipe forming, a pipe-making strain during steel pipe forming, and a coating condition, to predict a collapse strength of a coated steel pipe coated after steel pipe forming;
sequentially changing at least one of the steel pipe shape after steel pipe forming, the steel pipe strength characteristic after steel pipe forming, the pipe-making strain during steel pipe forming, and the coating condition included in the steel pipe manufacturing characteristic such that the predicted collapse strength of the coated steel pipe asymptotically approaches a requested collapse strength of an intended coated steel pipe, to determine an optimum steel pipe manufacturing characteristic;
determining a coated steel pipe manufacturing condition in accordance with the determined optimum steel pipe manufacturing characteristic; and
manufacturing a coated steel pipe under the determined coated steel pipe manufacturing condition.

### Claim 8 (depends on 7)

The steel pipe manufacturing method according to claim 7, wherein the steel pipe collapse strength prediction model is constructed by a neural network.