US 12,145,318 B2
Method and apparatus for monitoring a quality of an object of a 3D-print-job series of identical objects
Felix Buggenthin, Munich (DE); Siegmund Düll, Munich (DE); Mitchell Joblin, Munich (DE); Clemens Otte, Munich (DE); Axel Reitinger, Munich (DE); Victor Balanica, Ingolstadt (DE); Michael Caelers, Norrköping (SE); Jonas Eriksson, Finspong (SE); Jerry Fornander, Finspong (SE); Andreas Graichen, Finspang (SE); and Vincent Sidenvall, Örkelljunga (SE)
Assigned to Siemens Aktiengesellschaft, Munich (DE)
Appl. No. 16/650,754
Filed by Siemens Aktiengesellschaft, Munich (DE)
PCT Filed Sep. 19, 2018, PCT No. PCT/EP2018/075277
§ 371(c)(1), (2) Date Mar. 25, 2020,
PCT Pub. No. WO2019/063370, PCT Pub. Date Apr. 4, 2019.
Claims priority of application No. 17193155 (EP), filed on Sep. 26, 2017.
Prior Publication US 2020/0230884 A1, Jul. 23, 2020
Int. Cl. B29C 64/393 (2017.01); B22F 10/28 (2021.01); B22F 10/80 (2021.01); B22F 10/85 (2021.01); B22F 12/90 (2021.01); B29C 64/386 (2017.01); B33Y 50/00 (2015.01); G05B 19/418 (2006.01)
CPC B29C 64/393 (2017.08) [B22F 10/80 (2021.01); B22F 10/85 (2021.01); B22F 12/90 (2021.01); B29C 64/386 (2017.08); B33Y 50/00 (2014.12); G05B 19/41875 (2013.01); B22F 10/28 (2021.01); G05B 2219/31263 (2013.01); G05B 2219/32368 (2013.01); G05B 2219/35012 (2013.01)] 13 Claims
OG exemplary drawing
 
1. A method for monitoring a quality of an object of a 3D-print job series of identical objects, each object built from a multitude of stacked 2D-layers printed by a 3D-printer in an additive manufacturing process, the method comprising:
determining a layer quality indicator of a currently printed layer of the object, wherein the layer quality indicator is calculated under same conditions during former print jobs of identical objects having a same shape and a number of layers, for which a completely manufactured object was evaluated;
comparing the layer quality indicator of the currently printed layer with a predetermined lower confidence limit of a corresponding layer, the predetermined lower confidence limit being calculated depending on layer quality indicators of previously completely manufactured objects complying with predefined quality requirements;
performing a trend analysis taking into account the layer quality indicator of the currently printed layer and a subset of preceding layer quality indicators of preceding layers in a sequence of layer quality indicators for all layers of the object;
generating an early warning signal before a value of the layer quality indicator of the currently printed layer is equal to or lower than a lower quality limit, in response to the trend analysis showing that the subset of preceding layer quality indicators is trending toward the predetermined lower confidence limit;
in response to the early warning signal, automatically adjusting a setting of the 3D-printer to raise a layer quality of subsequent layers, or stopping the additive manufacturing process before completion of the object;
generating a warning signal, in response to the layer quality indicator of the currently printed layer having a value equal or lower than the lower quality limit or in response to layer quality indicators of subsequent layers showing a common trend towards the lower confidence limit; and
in response to the warning signal, automatically adjusting parameter settings of the additive manufacturing process to raise the layer quality of subsequent layers;
determining the layer quality indicator of the currently printed layer by machine learning means.