# US 2026/0216812 A1 · Welded portion characteristic prediction method, prediction model generation method, and prediction model

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

- Number: US 2026/0216812 A1
- Document ID: US20260216812A1
- Status: Published application
- Kind: A1
- Application number: 19/146,218
- Filed: 2024-02-27
- Published: 2026-07-30
- Earliest claimed: 2023-02-28
- Assignees: Jfe Steel Corporation
- Applicants: Jfe Steel Corporation
- Inventors: Yuki UMEDA; Koichi TANIGUCHI; Naoaki MUNEMURA; Nao KAWABE
- Web page: https://patentgrep.com/patent/US20260216812A1

## CPC

- B23K 11/256: {the measured parameter being the inter-electrode electrical resistance}
- B23K 11/115: {by means of two electrodes placed opposite one another on both sides of the welded parts}
- B23K 11/257: {the measured parameter being an electrical current}
- B23K 11/258: {the measured parameter being a voltage}
- B23K 31/006: {relating to using of neural networks}
- B23K 31/125: {Weld quality monitoring}
- B23K 2101/006: {Vehicles}
- B23K 2101/18: Sheet panels
- B23K 2103/04: Steel {or steel} alloys
- B23K: SOLDERING OR UNSOLDERING; WELDING; CLADDING OR PLATING BY SOLDERING OR WELDING; CUTTING BY APPLYING HEAT LOCALLY, e.g. FLAME CUTTING; WORKING BY LASER BEAM

## Abstract

A welded portion characteristic prediction method is a method of predicting characteristics of a welded portion (5) obtained by resistance spot welding to pass a current between two electrodes (3, 4) clamping a sheet combination of two or more overlapping metal sheets, the method including acquiring electrical information outputted during resistance spot welding, the electrical information including at least one of the current flowing between the two electrodes (3, 4), the voltage between the two electrodes (3, 4), the resistance between the two electrodes (3, 4), and the amount of heat generated 10 between the two electrodes (3, 4) during the resistance spot welding, and predicting characteristics of the welded portion (5) based on the electrical information.

## Claims

### Claim 1

A welded portion characteristic prediction method of predicting characteristics of a welded portion of a sheet combination of two or more overlapping metal sheets, the welded portion being obtained by resistance spot welding to pass a current between two electrodes clamping the sheet combination of two or more overlapping metal sheets, the method comprising:
acquiring electrical information outputted during resistance spot welding, the electrical information including at least one of the current flowing between the two electrodes, a voltage between the two electrodes, a resistance between the two electrodes, and an amount of heat generated between the two electrodes during the resistance spot welding; and
predicting characteristics of the welded portion of the sheet combination of two or more overlapping metal sheets based on the electrical information.

### Claim 2 (depends on 1)

The welded portion characteristic prediction method according to claim 1, wherein the predicting of the characteristics of the welded portion includes inputting the electrical information to a prediction model of the characteristics of the welded portion and acquiring a prediction result of the characteristics of the welded portion outputted from the prediction model.

### Claim 3 (depends on 2)

The welded portion characteristic prediction method according to claim 2, further comprising generating the prediction model by performing learning using the electrical information and an analysis result of the characteristics of the welded portion as learning data.

### Claim 4 (depends on 1)

The welded portion characteristic prediction method according to claim 1, wherein the acquiring of a prediction result of the characteristics of the welded portion includes acquiring a prediction result of at least one of a nugget diameter, a tensile shear strength, a cross tensile strength, and a fracture strength by mechanical peel test of the welded portion as the characteristics of the welded portion.

### Claim 5 (depends on 1)

The welded portion characteristic prediction method according to claim 1, wherein the acquiring of the electrical information includes acquiring a total amount of heat generated from a start of current passage to an end of current passage in the resistance spot welding as the electrical information.

### Claim 6 (depends on 5)

The welded portion characteristic prediction method according to claim 5, wherein the acquiring of the electrical information includes acquiring a resistance between the two electrodes at the end of current passage as the electrical information.

### Claim 7 (depends on 5)

The welded portion characteristic prediction method according to claim 5, wherein the acquiring of the electrical information includes acquiring an amount of heat generated during a partial period from the start of current passage, within a period from the start of current passage to the end of current passage, in the resistance spot welding as the electrical information.

### Claim 8 (depends on 1)

The welded portion characteristic prediction method according to claim 1, wherein the resistance spot welding is performed on a sheet combination of three or more overlapping metal sheets including at least a first metal sheet, a second metal sheet, and a third metal sheet, and
a sheet thickness ratio obtained by dividing a total sheet thickness of the sheet combination by a sheet thickness of a thinnest metal sheet in the sheet combination is 5 or more.

### Claim 9 (depends on 8)

The welded portion characteristic prediction method according to claim 8, wherein the acquiring of the electrical information includes acquiring a total amount of heat generated from a start of current passage to an end of current passage in the resistance spot welding as the electrical information.

### Claim 10 (depends on 9)

The welded portion characteristic prediction method according to claim 9, wherein the acquiring of the electrical information includes acquiring a resistance between the two electrodes at the end of current passage as the electrical information.

### Claim 11 (depends on 9)

The welded portion characteristic prediction method according to claim 9, wherein the resistance spot welding includes a first stage of passing a first current and a second stage of passing a second current, and
the acquiring of the electrical information includes acquiring an amount of heat generated during the first stage within a period from the start of current passage to the end of current passage in the resistance spot welding as the electrical information.

### Claim 12

A prediction model generation method of generating a model to predict characteristics of a welded portion of a sheet combination of two or more overlapping metal sheets, the welded portion being obtained by resistance spot welding to pass a current between two electrodes clamping the sheet combination of two or more overlapping metal sheets, the method comprising:
acquiring electrical information outputted during resistance spot welding, the electrical information including at least one of the current flowing between the two electrodes, a voltage between the two electrodes, a resistance between the two electrodes, and an amount of heat generated between the two electrodes during the resistance spot welding;
acquiring an analysis result of characteristics of the welded portion when resistance spot welding is performed on the sheet combination of two or more overlapping metal sheets; and
generating a model to predict the characteristics of the welded portion of the sheet combination of two or more overlapping metal sheets by performing machine learning based on learning data associating the electrical information with the analysis result of the characteristics of the welded portion when resistance spot welding is performed on the sheet combination of two or more overlapping metal sheets.

### Claim 13 (depends on 12)

A prediction model to predict characteristics of a welded portion obtained by resistance spot welding, the prediction model being generated by the prediction model generation method according to claim 12.
