Work vehicle sprayer system and method with nozzle monitoring

Patent metadata

Field	Value
Publication number	US-12403494-B2
Application number	US-202117383050-A
Country	US
Kind code	B2
Filing date	Jul 22, 2021
Priority date	Aug 26, 2020
Publication date	Sep 2, 2025
Grant date	Sep 2, 2025

Abstract

A nozzle monitoring system is provided for a sprayer system of a work vehicle. The system includes a first sensor configured to generate signals associated with vibrations of a first nozzle apparatus on the work vehicle that disperses a primary fluid from the sprayer system during operation; and a controller having a processor receiving the signals generated by the first sensor and having a memory coupled to the processor and storing instructions. The processor executes the stored instructions to: convert the vibrations into a frequency domain representation; generate an image from the frequency domain representation; classify the image to generate a clog condition probability; and generate, based on the clog condition probability, a command to initiate a cleaning event of the first nozzle apparatus.

First claim

What is claimed is:

1. A nozzle monitoring system for a sprayer system of a work vehicle comprising:
a first sensor configured to generate signals associated with vibrations of a first nozzle apparatus on the work vehicle that disperses a primary fluid from the sprayer system during operation; and
a controller having a processor receiving the signals generated by the first sensor and having a memory coupled to the processor and storing instructions, the processor executing the stored instructions to:
convert the vibrations into a frequency domain representation;
generate an image from the frequency domain representation;
execute image recognition to classify the image with regard to a clog condition;
generate, based on the clog condition of the classified image, a clog condition probability;
generate, based on the clog condition probability, a cleaning event command; and
initiate, based on the cleaning event command, a cleaning event of the first nozzle apparatus.

2. The nozzle monitoring system of claim 1 , wherein the controller is further configured to convert the vibrations into time domain frames and to subsequently convert the time domain frames into the frequency domain representation.

3. The nozzle monitoring system of claim 2 , wherein the controller is further configured to convert the time domain frames into the frequency domain representation with a Fourier transform.

4. The nozzle monitoring system of claim 3 , wherein the controller is further configured to, prior to generating the image, wrap the frequency domain representation onto a Mel-frequency cepstrum scale (MFC) spectrogram.

5. The nozzle monitoring system of claim 4 , wherein the controller is further configured to generate the image by performing a cosine transformation on the MFC spectrogram.

6. The nozzle monitoring system of claim 5 , wherein the controller is further configured to generate the image as an MFC coefficient image by performing the cosine transformation on the MFC spectrogram, identifying amplitudes of the cosine transformation, and expressing the amplitudes as MFC coefficients over time.

7. The nozzle monitoring system of claim 6 , wherein the controller is further configured to classify the MFC coefficient image with a neural network.

8. The nozzle monitoring system of claim 7 ,
wherein the MFC coefficient image generated based on the signals from the first sensor is a nozzle MFC coefficient image,
wherein the nozzle monitoring system further comprises a second sensor positioned at a distance from the first nozzle apparatus and configured to generate signals associated with vibrations of the work vehicle during operation; and
wherein the controller is further configured to receive the signals from the second sensor, to generate a baseline MFC coefficient image based on the signals from the second sensor, and to classify the nozzle MFC coefficient image with the baseline MFC coefficient image to generate the clog condition probability.

9. The nozzle monitoring system of claim 8 , wherein the controller is further configured to classify the MFC coefficient image with the neural network as a convolution neural network.

10. The nozzle monitoring system of claim 1 , wherein the first sensor is a piezoelectric sensor.

11. A sprayer system for dispersing a primary fluid on a work vehicle, comprising:
at least a first nozzle apparatus configured to disperse the primary fluid and to execute a cleaning event;
a first sensor configured to generate signals associated with vibrations of the first nozzle apparatus during operation; and
a controller having a processor receiving the signals generated by the first sensor and having a memory coupled to the processor and storing instructions, the processor executing the stored instructions to:
convert the vibrations into a frequency domain representation;
generate an image from the frequency domain representation;
execute image recognition with a neural network to classify the image with regard to a clog condition;
generate, based on the clog condition of the classified image, a clog condition probability;
compare the clog condition probability to a clog condition threshold;
generate, when the clog condition probability meets or exceeds the clog condition threshold, a cleaning event command; and
initiate, based on the cleaning event command, a cleaning event of the first nozzle apparatus.

12. The sprayer system of claim 11 , wherein the controller is further configured to convert the vibrations into time domain frames, to subsequently convert the time domain frames into the frequency domain representation, and to convert the time domain frames into the frequency domain representation with a Fourier transform.

13. The sprayer system of claim 12 , wherein the controller is further configured to, prior to generating the image, wrap the frequency domain representation onto a Mel-frequency cepstrum scale (MFC) spectrogram, and wherein the controller is further configured to generate the image by performing a cosine transformation on the MFC spectrogram.

14. The sprayer system of claim 13 , wherein the controller is further configured to generate the image as an MFC coefficient image by performing the cosine transformation on the MFC spectrogram, identifying amplitudes of the cosine transformation, and expressing the amplitudes as MFC coefficients over time.

15. The sprayer system of claim 14 , wherein the controller is further configured to classify the MFC coefficient image with the neural network as a convolutional neural network.

16. The sprayer system of claim 15 ,
wherein the MFC coefficient image generated based on the signals from the first sensor is a nozzle MFC coefficient image, and
wherein the sprayer system further comprises a second sensor positioned at a distance from the first nozzle apparatus and configured to generate signals associated with vibrations of the work vehicle during operation; and
wherein the controller is further configured to receive the signals from the second sensor, to generate a baseline MFC coefficient image based on the signals from the second sensor, and to classify the nozzle MFC coefficient image with the baseline MFC coefficient image to generate the clog condition probability.

17. The sprayer system of claim 16 , wherein the first sensor is a piezoelectric sensor.

18. The sprayer system of claim 17 , wherein the second sensor is a piezoelectric sensor.

19. The sprayer system of claim 11 , wherein the first nozzle apparatus comprises:
a manifold defining a plurality of manifold faces and a nozzle cavity within an interior of the manifold, wherein the manifold defines a fluid inlet passage extending between a first face of the manifold faces and the nozzle cavity, a fluid outlet passage extending between a second face of the manifold faces and the nozzle cavity, an air outlet passage extending between a third face of the manifold faces and the nozzle cavity, and an air inlet passage extending between an air inlet on at least one of the manifold faces and the nozzle cavity, and wherein the fluid inlet passage is configured to selectively receive the primary fluid and the air inlet passage is configured to selectively receive a flow of air;
a nozzle holder arranged within the nozzle cavity; and
at least one nozzle element mounted to or within the nozzle holder and defining a nozzle element passage with a nozzle element inlet and a nozzle element outlet, and
wherein the nozzle holder is selectively pivotable within the nozzle cavity, including betwe

Assignees

• Deere & Co

Inventors

• Ferren Bran
• Lariviere Donald
• Lichtenwalner Houstin
• Peraza Omar
• Liebersbach Robert

Classifications

• G06F2218/12

  Classification; Matching · CPC title

• G06F18/21

  Design or setup of recognition systems or techniques; Extraction of features in feature space; Blind source separation · CPC title

• G06T5/10

  using non-spatial domain filtering · CPC title

• G01H1/00

  Measuring {characteristics of} vibrations in solids by using direct conduction to the detector (G01H9/00, G01H11/00 take precedence) · CPC title

• B05B15/55

  using cleaning fluids · CPC title