Methods and systems for monitoring lubricant oil condition using photoacoustic modelling

Patent metadata

Field	Value
Publication number	US-12529650-B2
Application number	US-202318355266-A
Country	US
Kind code	B2
Filing date	Jul 19, 2023
Priority date	Aug 12, 2022
Publication date	Jan 20, 2026
Grant date	Jan 20, 2026

Abstract

The disclosure relates generally to methods and systems for monitoring lubricant oil condition using a photoacoustic modelling. Conventional techniques in the art for checking the condition of the lubricant oil is laboratory based and thus time consuming, error prone and not efficient. The present disclosure discloses a photoacoustic simulation model which is developed utilizing a photonic model such as a Monte Carlo method-based optical simulation integrated with a finite element model such as a k-wave toolbox-based acoustic measurement. The photoacoustic simulation model of the present disclosure is used to obtain a photoacoustic signal of the lubricant oil sample and a set of statistical features are determined from the obtained photoacoustic signal. The determined set of statistical features are then used as a training data to develop a machine learning (ML) model which is used to classify a type of contamination of the test lubricating oil.

First claim

What is claimed is:

1 . A processor-implemented method comprising the steps of:
receiving, via one or more hardware processors, photoacoustic signals from a plurality of lubricant oil samples, and a classification label for each of the plurality of lubricant oil samples, wherein the classification label for each lubricant oil sample defines a contamination type of corresponding lubricant oil sample; simulating, via the one or more hardware processors, the photoacoustic signal from each lubricant oil sample, to obtain a plurality of simulated photoacoustic signals from the plurality of lubricant oil samples, using a photoacoustic simulation model, wherein the photoacoustic simulation model makes use of both one or more optical parameters and one or more acoustic parameters of the photoacoustic signal of the corresponding lubricant oil sample for simulation to obtain the corresponding simulated photoacoustic signal, wherein simulating each lubricant oil sample to obtain corresponding simulated photoacoustic signal using the photoacoustic simulation model, comprises:
determining the one or more optical parameters, a contaminant concentration, a temperature, and the one or more acoustic parameters, for the corresponding lubricant oil sample, wherein the one or more optical parameters of the photoacoustic signal of the lubricant oil sample comprises an optical absorption coefficient μ a , a scattering coefficient μ s and a scattering anisotropy g, wherein the one or more optical parameters of the photoacoustic signal of the corresponding lubricant oil sample are determined based on a predefined optical wavelength of the photoacoustic signal of the corresponding lubricant sample, the contaminant concentration of the lubricant oil sample is determined in the form of an effective absorption coefficient ( ) and the one or more acoustic parameters comprises a speed of sound, a density of the lubricant oil sample and a Grüneisen parameter;
determining a fluence for the corresponding lubricant oil sample, based on the one or more optical parameters and the contaminant concentration, using a photonic model;
generating an initial acoustic pressure signal for the lubricant oil sample, using a finite element technique based on the one or more acoustic parameters, the contaminant concentration, the temperature, and the fluence;
propagating the initial acoustic pressure signal to obtain a propagated acoustic pressure signal for the corresponding lubricant oil sample, using the finite element technique; and
obtaining the simulated photoacoustic signal for the corresponding lubricant oil sample, by correlating the initial acoustic pressure signal and the propagated acoustic pressure signal;
determining, via the one or more hardware processors, one or more statistical features, for each lubricant oil sample, from the corresponding simulated photoacoustic signal, using a signal processing technique, wherein the one or more statistical features for each lubricant oil sample are determined from a time-frequency signal of the corresponding simulated photoacoustic signal; and training, via the one or more hardware processors, a machine learning (ML) model with (i) the one or more statistical features for each lubricant oil sample of the plurality of lubricant oil samples and (ii) the classification label for each of the plurality of lubricant oil samples, to obtain a trained ML model; receiving, via the one or more hardware processors, a test photoacoustic signal from a test lubricant oil sample for which lubricant oil condition to be monitored; determining, via the one or more hardware processors, one or more statistical features of the test lubricant oil sample, from the test photoacoustic signal, using the signal processing technique; and passing, via the one or more hardware processors, the one or more statistical features of the test lubricant oil sample, to the trained ML model, to obtain classification label for the lubricant oil sample, wherein the classification label provides the lubricant oil condition of the test lubricant oil sample.

2 . The processor-implemented method of claim 1 , wherein determining the test photoacoustic signal from the test lubricant oil sample, using the experimental model, comprising:
obtaining a modulated signal for the test lubricant oil sample, using an arbitrary waveform generator; obtaining an intensity modulated laser signal for the test lubricant oil sample, by passing the modulated signal to a continuous wave laser source; obtaining an ultrasound signal for the test lubricant oil sample, by irradiating the intensity modulated laser signal on the test lubricant oil sample using an ultrasound sensor; and obtaining the test photoacoustic signal for the test lubricant oil sample, by correlating the modulated signal and the ultrasound signal.

3 . The processor-implemented method of claim 1 , wherein the one or more statistical features of the test lubricant oil sample, are determined from a time-frequency signal of the test photoacoustic signal.

4 . The processor implemented claim 1 , wherein the photoacoustic simulation model is used as a digital twin, wherein one copy of the photoacoustic simulation model is used to develop a machine learning model and another copy of the photoacoustic simulation model is used to monitor condition of each lubricant oil sample using the trained ML model.

5 . The processor implemented claim 1 , wherein the trained ML model is used for monitoring the condition of the lubricant oil and thus monitoring corresponding health and performance of a machine and determining a fault type of the machine.

6 . A system comprising:
a memory storing instructions; one or more input/output (I/O) interfaces; and one or more hardware processors coupled to the memory via the one or more I/O interfaces, wherein the one or more hardware processors are configured by the instructions to: receive photoacoustic signals from a plurality of lubricant oil samples, and a classification label for each of the plurality of lubricant oil samples; simulate photoacoustic signal from each lubricant oil sample, to obtain a plurality of simulated photoacoustic signals from the plurality of lubricant oil samples, using a photoacoustic simulation model, wherein the one or more hardware processors are configured by the instructions to simulate the photoacoustic signal from each lubricant oil sample to obtain corresponding simulated photoacoustic signal using the photoacoustic simulation model by:
determining one or more optical parameters, a contaminant concentration, a temperature, and one or more acoustic parameters, for the corresponding lubricant oil sample, wherein the one or more optical parameters of the lubricant oil sample comprises an optical absorption coefficient μ a , a scattering coefficient μ s and a scattering anisotropy g, wherein the one or more optical parameters of the lubricant oil sample are determined based on a predefined optical wavelength of the photoacoustic signal of the corresponding lubricant sample, the contaminant concentration of the lubricant oil sample is determined in the form of an effective absorption coefficient ( ) and the one or more acoustic parameters comprises a speed of sound, a density of the lubricant oil sample and a Grüneisen parameter;
determining a fluence for the corresponding lubricant oil sample, based on the one or more optical parameters and the contaminant concentration, using a photonic model;
generating an initial acoustic pressure signal for the lubricant oil sample, using a finite element technique based on the one or more acoustic parameters, the contaminant concentration, the temperature, and the fluence;
propagating the initial acoustic pressure signal to obtain a propagated acoustic pressure signal f

Assignees

• Tata Consultancy Services Ltd

Inventors

• Chatterjee Subhasri
• Gorey Abhijit
• Sinharay Arijit
• Bhaumik Chirabrata
• Chakravarty Tapas
• Gain Supriya
• Pal Arpan

Classifications

• G01N2201/06113

  Coherent sources; lasers · CPC title

• G01N33/2888

  Lubricating oil characteristics, e.g. deterioration (lubricating properties G01N33/30) · CPC title

• G06N20/10

  using kernel methods, e.g. support vector machines [SVM] · CPC title

• G06N7/01

  Probabilistic graphical models, e.g. probabilistic networks · CPC title

• G06N5/01

  Dynamic search techniques; Heuristics; Dynamic trees; Branch-and-bound · CPC title