Kobe Steel, Ltd.

Kabushiki Kaisha Kobe Seiko Sho

Shuo Huang

Masatoshi Hida

Tatsuya Fujii

Shinji Sato

Takeshi Yamada

Yasuto Kataoka

Eisuke Kurosawa

Satoshi Chikaguchi

Akinori Yoshikawa

Shota Tsubaki

A machine learning device that performs machine learning of a welding condition for manufacturing an additively-manufactured object by welding a filler metal and depositing weld beads, the machine learning device includes: at least one hardware processor configured to perform a learning process for generating a learned model using a welding condition of a weld bead and a block pattern formed by the weld bead as input data and shape data of the weld bead as output data.

Machine learning device, additive manufacturing system, machine learning method for welding condition, method for determining welding condition, and a non-transitory computer readable medium

A method for manufacturing an additively-manufactured object, in which a plurality of weld beads obtained by melting and solidifying a filler metal are deposited on a base portion to build a built-up object, includes: a support bead forming step of forming a support bead on the base portion; and a depositing step of depositing a weld bead on the support bead. When the support bead is formed to be inclined from a vertical direction in the support bead forming step, a ratio H/W of a height H to a width W of the support bead is set to 0.35 or more.

Manufacturing method for additive manufactured article and additive manufactured article

In preparing a built-up object by depositing beads, in a step of dividing into the bead model, a trapezoidal bead model a cross section of which is a trapezoidal shape is applied to a position where the bead is formed in a portion not adjacent to an existing bead, and a parallelogram bead model a cross section of which is a parallelogram is applied to a position where the bead is formed adjacent to a bead that is already formed, in the parallelogram bead model opposite sides in the deposition direction of the bead being parallel to each other, and opposite sides in the bead arrangement direction being parallel to a side of another bead mode that is adjacent.

Additive manufacturing method, additive manufacturing device, and controller

An excess metal amount setting method includes: a thermal shrinkage prediction step of predicting a thermal shrinkage amount in the deposited body after manufacturing; a thermal shrinkage modifying step of obtaining a thermal deformation modifying profile by expanding a target profile according to the thermal shrinkage amount; a release strain prediction step of predicting an elastic deformation amount due to release strain of the deposited body after machining; an elastic deformation modifying step of obtaining an elastic deformation modifying profile by deforming the thermal deformation modifying profile according to the elastic deformation amount in a direction opposite to a deformation direction due to the release strain; and an excess metal amount setting step of adjusting an outer edge shape of the deposited body so that an excess metal amount from the elastic deformation modifying profile to an outer edge of the deposited body falls within a predetermined reference range.

Method for setting excess thickness, device for setting excess thickness, method for producing shaped object, and program

A method for manufacturing an additively-manufactured object, includes: an additively-manufacturing step of building a layered body by depositing a weld bead obtained by melting and solidifying a filler metal, the layered body having an opening along a forming direction of the weld bead and an internal space surrounded by the weld bead; and a closing step of forming a closing wall portion connecting an edge portion of the opening with the weld bead for closing. In the additively-manufacturing step, the opening is formed with a width dimension larger than a bead width of the weld bead, and in the closing step, the closing wall portion having a width dimension larger than the bead width is formed by the weld bead to close the opening.

Method for manufacturing laminated molding, and laminated molding

A building time for building an additively-manufactured object is calculated on the basis of the inter-pass time and the welding pass time and is compared with a preset upper limit value, and welding conditions in a depositing plan are repeatedly modified until the building time is equal to or less than the upper limit value. Alternatively, corrections are repeatedly performed until the shape difference between a building shape of built-up object shape data relating to the additively-manufactured object created on the basis of the inter-pass time and the inter-pass temperature, and a building shape of three-dimensional shape data, is smaller than a near net value.

Growing community of inventors

Hyogo, Japan

Shuo Huang