CPC G05B 19/042 (2013.01) [G06N 20/00 (2019.01); G05B 2219/2633 (2013.01)] | 16 Claims |
9. A method carried out by a system comprising:
an acoustic sensor data interface; and
a machine learning-based processing system;
wherein the method comprises:
receiving, by the acoustic sensor data interface, digital acoustic signal data corresponding to sensed sound from components of a laundry machine during operation of the laundry machine, wherein the acoustic sensor includes at least a microphone configured to render a transduced electronic signal of sound waves sensed by the microphone during operation of the laundry machine; and
rendering, by the machine learning-based processing system, a reason code indicative of a current operational status of the laundry machine,
wherein the processing system comprises a processor and a non-transitory computer readable medium including computer-executable instructions that, when executed by the processor, facilitate carrying out a method during the rendering that comprises:
receiving an acoustic data set rendered from the transduced electronic signal;
rendering functional metric parameter values indicative of an operational status of the laundry machine by applying machine learning models to the acoustic data set;
identifying, by applying a set of conditions to a set of predictive maintenance indicators derived from the functional metric parameter values, the reason code corresponding to a degraded operational status of the laundry machine; and
issuing, in accordance with the identifying, an electronic maintenance alert relating to a remedial operation for the laundry machine,
wherein the machine learning models define acoustic signatures for corresponding normal functions performed by the laundry machine, and
wherein the machine learning models define a percentage of a total operational time for processing a laundry load by the laundry machine where an identified normal function is acoustically sensed and identifiable using a corresponding acoustic signature of the machine learning models.
|