US 12,187,298 B2
Method and apparatus for monitoring operation of a technical object
Thomas Bierweiler, Karlsruhe (DE); Jana Kemnitz, Vienna (AT); Daniel Schall, Hollabrunn (AT); Stefan Von Dosky, Karlsruhe (DE); and Roman Khazankin, Vienna (AT)
Assigned to Siemens Aktiengesellschaft, Munich (DE)
Filed by Siemens Aktiengesellschaft, Munich (DE)
Filed on Jan. 19, 2022, as Appl. No. 17/578,563.
Claims priority of application No. 21152772 (EP), filed on Jan. 21, 2021.
Prior Publication US 2022/0227378 A1, Jul. 21, 2022
Int. Cl. B60W 50/02 (2012.01); B60W 50/00 (2006.01); G06F 18/22 (2023.01); G06N 20/00 (2019.01)
CPC B60W 50/0225 (2013.01) [B60W 50/0205 (2013.01); G06F 18/22 (2023.01); G06N 20/00 (2019.01); B60W 2050/0057 (2013.01); B60W 2050/021 (2013.01)] 15 Claims
OG exemplary drawing
 
1. A method for monitoring the operation of a technical object aided by a mathematical model based on machine learning after assembly of a three-axis replacement acceleration sensor, a three-axis first acceleration sensor being initially assembled on the object at a position, and during operation the object generating mechanical oscillations, the method comprising:
a) determining a first orientation of the three-axis first acceleration sensor on the object as a position vector relative to gravitational force;
b) generating and training the mathematical model for the operation of the object with respect to the position vector;
c) detaching the three-axis first acceleration sensor and assembling the three-axis replacement acceleration sensor with a new orientation on the object at the same position, at which the first acceleration sensor was assembled;
d) detecting acceleration values in an x-, y- and z-direction aided by respective individual sensors of the three-axis replacement acceleration sensor of the object;
e) calculating respective indicator values from a temporal course of the detected acceleration values of the three-axis replacement acceleration sensor;
f) determining a replacement vector from the calculated respective indicator values and determining a differential vector between the replacement vector and the position vector of the first acceleration sensor determined during said generating and training of b);
g) applying the model during operation of the object for the position vector in the orientation of the replacement vector by taking into account the differential vector.