# US 2025/0340004 A1 · Generative artificial intelligence based mold creation to minimize a rejection of a casting product due to a distortion

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

- Number: US 2025/0340004 A1
- Document ID: US20250340004A1
- Status: Published application
- Kind: A1
- Application number: 18/654,305
- Filed: 2024-05-03
- Published: 2025-11-06
- Earliest claimed: 2024-05-03
- Assignees: International Business Machines Corporation
- Applicants: International Business Machines Corporation
- Inventors: Sarbajit K. Rakshit; Sathya Santhar; Sridevi Kannan
- Web page: https://patentgrep.com/patent/US20250340004A1

## CPC

- B22C 9/00: Moulds or cores (uniquely adapted to particular casting processes B22D); Moulding processes
- B29C 45/768: {Detecting defective moulding conditions (B29C45/84 takes precedence)}
- B22C 9/06: Permanent moulds for shaped castings (moulds for ingots B22D7/06)
- B22D 45/00: Equipment for casting, not otherwise provided for
- B22D 46/00: Controlling, supervising, not restricted to casting covered by a single main group, e.g. for safety reasons (controlling or regulating in general G05)
- G06N 3/088: Non-supervised learning, e.g. competitive learning
- B29C 2945/76949: using a learning system, i.e. the system accumulates experience from previous occurrences, e.g. adaptive control
- B33Y 50/02: for controlling or regulating additive manufacturing processes
- B22C: FOUNDRY MOULDING (moulding refractory materials in general B28B)
- B29C: SHAPING OR JOINING OF PLASTICS; SHAPING OF MATERIAL IN A PLASTIC STATE, NOT OTHERWISE PROVIDED FOR; AFTER-TREATMENT OF THE SHAPED PRODUCTS, e.g. REPAIRING (making preforms B29B11/00; making laminated products by combining previously unconnected layers which become one product whose layers will remain together B32B37/00 - B32B41/00)
- B22D: CASTING OF METALS; CASTING OF OTHER SUBSTANCES BY THE SAME PROCESSES OR DEVICES (shaping of plastics or substances in a plastic state B29C; metallurgical processing, selection of substances to be added to metal C21, C22)
- G06N: COMPUTING ARRANGEMENTS BASED ON SPECIFIC COMPUTATIONAL MODELS
- B33Y: ADDITIVE MANUFACTURING, i.e. MANUFACTURING OF THREE-DIMENSIONAL [3D] OBJECTS BY ADDITIVE DEPOSITION, ADDITIVE AGGLOMERATION OR ADDITIVE LAYERING, e.g. BY 3D PRINTING, STEREOLITHOGRAPHY OR SELECTIVE LASER SINTERING

## Abstract

One or more computer processors predicting a distortion in a cast product created from a mold, utilizing a trained generative model. The one or more computer processors modify the mold associated with the predicted distortion. The one or more computer processors generate a corrective mold design for one or more material and shape conditions to remediate the predicted distortion. The one or more computer processors create a corrective mold with an appropriate specification based on the corrective mold design using a robotic system to minimize post-processing. The one or more computer processors produce a final cast product with the created corrective mold.

## Claims

### Claim 1

A computer-implemented method comprising:
predicting, by one or more computer processors, a distortion in a cast product created from a mold, utilizing a trained generative model;
modifying, by one or more computer processors, the mold associated with the predicted distortion, further comprising:
generating, by one or more computer processors, a corrective mold design for one or more material and shape conditions to remediate the predicted distortion; and
creating, by one or more computer processors, a corrective mold with an appropriate specification based on the corrective mold design using a robotic system to minimize post-processing; and
producing, by one or more computer processors, a final cast product with the created corrective mold.

### Claim 2 (depends on 1)

The computer-implemented method of claim 1, wherein training the generative model comprises:
gathering a set of historical data, wherein the set of historical data includes a visual analysis of distorted cast products manufactured with a casting process, one or more capabilities of one or more corrective methods, one or more limits of one or more corrective methods, an amount of material removed during a post-processing process, and a shape of the mold; and
training the generative model with the set of historical data.

### Claim 3 (depends on 1)

The computer-implemented method of claim 1, wherein modifying the mold associated with the predicted distortion, further comprises:
receiving a three-dimensional (3D) design of the cast product to be manufactured with casting; and
identifying an amount of allowance on the corrective mold so that the predicted distortion is minimized with minimal post-processing.

### Claim 4 (depends on 1)

The computer-implemented method of claim 1, further comprising:
monitoring at least one process parameter selected from the group consisting of a type of material, a shape of a mold, a temperature, and a cooling rate.

### Claim 5 (depends on 1)

The computer-implemented method of claim 1, wherein the generative model is a conditional generative adversarial network.

### Claim 6 (depends on 1)

The computer-implemented method of claim 1, wherein distortions are selected from the group consisting of a thermal gradient, a shrinkage, a residual stress, a mold constraint, a core shift, an inadequate support or anchoring, and a machining operation.

### Claim 7 (depends on 1)

The computer-implemented method of claim 1, wherein post-production steps are selected from the group consisting of a removal of gating and risers, a surface cleaning and finishing, a machining and precision operation, a heat treatment, a surface coating, a non-destructive testing, and a quality inspection.

### Claim 8

A computer program product comprising:
one or more computer readable storage media having computer-readable program instructions stored on the one or more computer readable storage media, said program instructions executes a computer-implemented method comprising steps of:
predicting a distortion in a cast product created from a mold, utilizing a trained generative model;
modifying the mold associated with the predicted distortion, further comprising:
generating a corrective mold design for one or more material and shape conditions to remediate the predicted distortion; and
creating a corrective mold with an appropriate specification based on the corrective mold design using a robotic system to minimize post-processing; and
producing a final cast product with the created corrective mold.

### Claim 9 (depends on 8)

The computer program product of claim 8, wherein the program instructions to train the generative model, stored on the one or more computer readable storage media, comprise the steps of:
gathering a set of historical data, wherein the set of historical data includes a visual analysis of distorted cast products manufactured with a casting process, one or more capabilities of one or more corrective methods, one or more limits of one or more corrective methods, an amount of material removed during a post-processing process, and a shape of the mold; and
training the generative model with the set of historical data.

### Claim 10 (depends on 8)

The computer program product of claim 8, wherein the program instructions to modify the mold associated with the predicted distortion, stored on the one or more computer readable storage media, comprise the steps of:
receiving a three-dimensional (3D) design of the cast product to be manufactured with casting; and
identifying an amount of allowance on the corrective mold so that the predicted distortion is minimized with minimal post-processing.

### Claim 11 (depends on 8)

The computer program product of claim 8, wherein the program instructions, stored on the one or more computer readable storage media, further comprise the steps of:
monitoring at least one process parameter selected from the group consisting of a type of material, a shape of a mold, a temperature, and a cooling rate.

### Claim 12 (depends on 8)

The computer program product of claim 8, wherein the generative model is a conditional generative adversarial network.

### Claim 13 (depends on 8)

The computer program product of claim 8, wherein distortions are selected from the group consisting of a thermal gradient, a shrinkage, a residual stress, a mold constraint, a core shift, an inadequate support or anchoring, and a machining operation.

### Claim 14 (depends on 8)

The computer program product of claim 8, wherein post-production steps are selected from the group consisting of a removal of gating and risers, a surface cleaning and finishing, a machining and precision operation, a heat treatment, a surface coating, a non-destructive testing, and a quality inspection.

### Claim 15

A computer system comprising:
one or more computer processors;
one or more computer readable storage media having computer readable program instructions stored on the one or more computer readable storage media for execution by at least one of the one or more processors, the stored program instructions execute a computer-implemented method comprising steps of:
predicting a distortion in a cast product created from a mold, utilizing a trained generative model;
modifying the mold associated with the predicted distortion, further comprising:
generating a corrective mold design for one or more material and shape conditions to remediate the predicted distortion; and
creating a corrective mold with an appropriate specification based on the corrective mold design using a robotic system to minimize post-processing; and
producing a final cast product with the created corrective mold.

### Claim 16 (depends on 15)

The computer system of claim 15, wherein the program instructions to train the generative model, stored on the one or more computer readable storage media, comprise the steps of:
gathering a set of historical data, wherein the set of historical data includes a visual analysis of distorted cast products manufactured with a casting process, one or more capabilities of one or more corrective methods, one or more limits of one or more corrective methods, an amount of material removed during a post-processing process, and a shape of the mold; and
training the generative model with the set of historical data.

### Claim 17 (depends on 15)

The computer system of claim 15, wherein the program instructions to modify the mold associated with the predicted distortion, stored on the one or more computer readable storage media, comprise the steps of:
receiving a three-dimensional (3D) design of the cast product to be manufactured with casting; and
identifying an amount of allowance on the corrective mold so that the predicted distortion is minimized with minimal post-processing.

### Claim 18 (depends on 15)

The computer system of claim 15, wherein the program instructions, stored on the one or more computer readable storage media, further comprise the steps of:
monitoring at least one process parameter selected from the group consisting of a type of material, a shape of a mold, a temperature, and a cooling rate.

### Claim 19 (depends on 15)

The computer system of claim 15, wherein the generative model is a conditional generative adversarial network.

### Claim 20 (depends on 15)

The computer system of claim 15, wherein distortions are selected from the group consisting of a thermal gradient, a shrinkage, a residual stress, a mold constraint, a core shift, an inadequate support or anchoring, and a machining operation.
