This invention is directed to a method of efficiently improving a relative density of a shaped object using an evaluation criterion having a higher correlation with a density of an object to be shaped. The method according to this invention includes acquiring three-dimensional point group data of a surface of a shaping object, calculating at least one of three-dimensional surface texture parameters extended to a plane region using the three-dimensional point group data, and evaluating a quality of the object to be shaped using the at least one of the three-dimensional surface texture parameters.

This invention is directed to a method of efficiently improving a relative density of a shaped object using an evaluation criterion having a higher correlation with a density of an object to be shaped. The method according to this invention includes acquiring three-dimensional point group data of a surface of a shaping object, calculating at least one of three-dimensional surface texture parameters extended to a plane region using the three-dimensional point group data, and evaluating a quality of the object to be shaped using the at least one of the three-dimensional surface texture parameters.

The invention relates to a material feeding device. The material feeding device according to the invention to be used in a material processing device has a material feeding channel with an output end facing a processing site during operation of the material feeding device, and is characterized in that the material feeding device has at least one microchannel.

The invention relates to a material feeding device. The material feeding device according to the invention to be used in a material processing device has a material feeding channel with an output end facing a processing site during operation of the material feeding device, and is characterized in that the material feeding device has at least one microchannel.

Embodiments relate to in-situ process monitoring of a part being made via additive manufacturing. The process can involve capturing computed tomography (CT) scans of a post-built part. A neural network (NN) can be used during the build of a new part to process multi-modal sensor data. Spatial and temporal registration techniques can be used to align the data to x,y,z coordinates on the build plate. During the build of the part, the multi-modal sensor data can be superimposed on the build plate. Machine learning can be used to train the NN to correlate the sensor data to a defect label or a non-defect label by looking to certain patterns in the sensor data at the x,y,z location to identify a defect in the CT scan at x,y,z. The NN can then be used to predict where defects are or will occur during an actual build of a part.

Embodiments relate to in-situ process monitoring of a part being made via additive manufacturing. The process can involve capturing computed tomography (CT) scans of a post-built part. A neural network (NN) can be used during the build of a new part to process multi-modal sensor data. Spatial and temporal registration techniques can be used to align the data to x,y,z coordinates on the build plate. During the build of the part, the multi-modal sensor data can be superimposed on the build plate. Machine learning can be used to train the NN to correlate the sensor data to a defect label or a non-defect label by looking to certain patterns in the sensor data at the x,y,z location to identify a defect in the CT scan at x,y,z. The NN can then be used to predict where defects are or will occur during an actual build of a part.

Embodiments relate to in-situ process monitoring of a part being made via additive manufacturing. The process can involve capturing computed tomography (CT) scans of a post-built part. A neural network (NN) can be used during the build of a new part to process multi-modal sensor data. Spatial and temporal registration techniques can be used to align the data to x,y,z coordinates on the build plate. During the build of the part, the multi-modal sensor data can be superimposed on the build plate. Machine learning can be used to train the NN to correlate the sensor data to a defect label or a non-defect label by looking to certain patterns in the sensor data at the x,y,z location to identify a defect in the CT scan at x,y,z. The NN can then be used to predict where defects are or will occur during an actual build of a part.

A method for forming a three-dimensional article through successive fusion of parts of a powder bed comprising: providing a model of the three dimensional article, applying a first powder layer on a work table, directing an energy beam over the work table causing the first powder layer to fuse in selected locations according to the model to form a first cross section of the three-dimensional article, applying a second powder layer on the work table, directing the energy beam over the work table causing the second powder layer to fuse in selected locations according to the model to form a second cross section of the three-dimensional article, wherein the second layer is bonded to the first layer, detecting a local thickness in at least two positions in at least the second powder layer, varying an energy beam parameter depending on the detected local thickness of the second powder layer.

A method and apparatus for the generative manufacture of a three-dimensional component in a processing chamber, in which the steps “providing a metallic starting material in the processing chamber” and “melting the starting material by means of energy input” are repeated multiple times, wherein a process gas is provided in the processing chamber are disclosed. The method is characterized by the steps: 1) the hydrogen content of the process gas or a sample of the process gas is determined; 2) the oxygen content of the process gas or a sample of the process gas is determined by means of an oxygen sensor and/or the dew point of the process gas or a sample of the process gas is determined; and 3) the values for the oxygen content and/or the dew point determined in step 2 are corrected by means of the value for the hydrogen content determined in step 1.

A method and apparatus for the generative manufacture of a three-dimensional component in a processing chamber, in which the steps “providing a metallic starting material in the processing chamber” and “melting the starting material by means of energy input” are repeated multiple times, wherein a process gas is provided in the processing chamber are disclosed. The method is characterized by the steps: 1) the hydrogen content of the process gas or a sample of the process gas is determined; 2) the oxygen content of the process gas or a sample of the process gas is determined by means of an oxygen sensor and/or the dew point of the process gas or a sample of the process gas is determined; and 3) the values for the oxygen content and/or the dew point determined in step 2 are corrected by means of the value for the hydrogen content determined in step 1.