US 10,890,499 B2
System and method for predicting strain power spectral densities of light machine structure
Benjamin John Hodel, Dunlap, IL (US); and David Andres Villero, Champaign, IL (US)
Assigned to Caterpillar Inc., Peoria, IL (US)
Filed by Caterpillar Inc., Peoria, IL (US)
Filed on Dec. 21, 2017, as Appl. No. 15/850,217.
Prior Publication US 2019/0195712 A1, Jun. 27, 2019
Int. Cl. G01L 5/00 (2006.01); G01P 15/08 (2006.01); G01M 7/02 (2006.01)
CPC G01L 5/0061 (2013.01) [G01M 7/025 (2013.01); G01P 15/0802 (2013.01)] 20 Claims
OG exemplary drawing
 
1. A machine, comprising:
a frame;
an engine supported on the frame;
ground-engaging elements supported on the frame and driven by the engine to propel the machine;
a machine light structure supported on the frame;
a multiple-axis accelerometer positioned to generate an acceleration signal relative to a base of the machine light structure;
a controller to:
receive the acceleration signal;
partition the acceleration signal into overlapped window periods;
transform the acceleration signal of each overlapped window period into a respective power spectral density; and
pass the power spectral densities through a strain power spectral densities prediction model to generate predicted strain power spectral densities of the machine light structure,
wherein the strain power spectral densities prediction model includes a regression model;
a health monitor to:
receive the predicted strain power spectral densities; and
generate a notification responsive to the predicted strain power spectral densities,
wherein the predicted strain power spectral densities are used to optimize the machine by one or more of:
adding strength to the machine light structure, or
reducing weight from the machine light structure.