Waste material analysis system

Patent metadata

Field	Value
Publication number	US-11262342-B2
Application number	US-201715648002-A
Country	US
Kind code	B2
Filing date	Jul 12, 2017
Priority date	Jul 12, 2017
Publication date	Mar 1, 2022
Grant date	Mar 1, 2022

Abstract

Methods, systems, and apparatus, including computer programs encoded on computer storage media, for molecular content analysis. One of the methods includes receiving molecular content data that indicates a molecular content of a portion of a batch of sludge; determining, by a machine learning module included in the system using the molecular content data and multiple parameters, a predicted value for a property of the portion of the batch of sludge; and providing the predicted value for the property of the portion of the batch of sludge.

First claim

What is claimed is:

1. A system comprising one or more computers and one or more storage devices on which are stored instructions that are operable, when executed by the one or more computers, to cause the one or more computers to perform operations comprising:
receiving molecular content data that indicates a molecular content of a portion of a batch of sludge;
determining, by a machine learning module included in the system using the molecular content data and multiple parameters, a predicted value for a property of the portion of the batch of sludge, the predicted value representing an expected value for the property if one or more actions associated with the batch of sludge were to occur;
providing the predicted value for the property of the portion of the batch of sludge; and
after determining the predicted value for the property of the portion of the batch of sludge:
receiving an actual value for the property of the portion of the batch of sludge after occurrence of the one or more actions; and
updating, by the machine learning module and using the actual value for the property of the portion of the batch of sludge, at least one of the multiple parameters used by the machine learning module to determine the predicted value for the property of the portion of the batch of sludge.

2. The system of claim 1 , wherein receiving the molecular content data that indicates the molecular content of the batch of sludge comprises receiving, from a near infrared spectrometer, near-infrared spectroscopy data.

3. The system of claim 2 , comprising the near infrared spectrometer.

4. The system of claim 3 , the operations comprising:
scanning, by the near infrared spectrometer, the batch of sludge to determine the molecular content data.

5. The system of claim 1 , wherein determining, by the machine learning module included in the system using the molecular content data and the multiple parameters, the predicted value for the property of the portion of the batch of sludge comprises determining, in parallel by the machine learning module for each of two or more properties of the batch of sludge, a predicted value for the respective property.

6. The system of claim 5 , wherein determining, in parallel by the machine learning module for each of the two or more properties of the batch of sludge, the predicted value for the respective property comprises determining, for each of the two or more properties of the batch of sludge using regression analysis, the predicted value for the respective property.

7. The system of claim 5 , wherein determining, in parallel by the machine learning module for each of the two or more properties of the batch of sludge, the predicted value for the respective property comprises determining, for each of the two or more properties of the batch of sludge using non-linear classification, the predicted value for the respective property.

8. The system of claim 1 , wherein determining, by the machine learning module included in the system using the molecular content data and the multiple parameters, the predicted value for the property of the portion of the batch of sludge comprises determining the predicted value for the property of the portion of the batch of sludge using the molecular content data, the multiple parameters, and one or more potential uses for the batch of sludge.

9. The system of claim 1 , wherein determining, by the machine learning module included in the system using the molecular content data and the multiple parameters, the predicted value for the property of the portion of the batch of sludge comprises determining, by the machine learning module, the predicted value using, as input to the machine learning module, characteristics of a potential purchaser, environmental parameters, or both.

10. The system of claim 1 , wherein determining, by the machine learning module included in the system using the molecular content data and the multiple parameters, the predicted value for the property of the portion of the batch of sludge comprises determining at least one of:
a predicted value for a percentage of dry matter of the batch of sludge;
a predicted value for a percentage of a primary material included in the batch of sludge;
a predicted value for a percentage of volatile solids included in the batch of sludge;
a predicted value for a percentage of biogas material included in the batch of sludge;
a predicted value for a percentage of organic matter included in the batch of sludge;
a predicted value for a percentage of phosphorus included in the batch of sludge;
a predicted value for a percentage of zinc included in the batch of sludge; or
a predicted value for a percentage of material included in the batch of sludge that can be reused.

11. The system of claim 1 , wherein:
providing the predicted value for the property of the portion of the batch of sludge comprises providing, to a treatment module, the predicted value for the property of the portion of the batch of sludge to cause the treatment module to determine a recommended treatment process for the batch of sludge given the predicted value for the property of the portion of the batch of sludge and one or more potential uses for the batch of sludge; and
receiving the actual value for the property of the portion of the batch of sludge comprises receiving the actual value for the property of the portion of the batch of sludge after a treatment system processes the batch of sludge using the recommended treatment process.

12. The system of claim 11 , the operations comprising:
determining, using the molecular content data and from two or more potentials uses for the batch of sludge, a recommended potential use for the batch of sludge, wherein:
providing, to the treatment module, the predicted value for the property of the portion of the batch of sludge comprises providing, to the treatment module, the recommended potential use and the predicted value for the property of the portion of the batch of sludge to cause the treatment module to determine a recommended treatment process for the batch of sludge given the predicted value for the property of the portion of the batch of sludge and the recommended potential use determined from the two or more potential uses for the batch of sludge.

13. The system of claim 12 , the operations comprising:
determining, by the treatment module, the recommended treatment process for the batch of sludge given the predicted value for the property of the portion of the batch of sludge and the recommended potential use determined from the two or more potential uses for the batch of sludge.

14. The system of claim 1 , wherein:
providing the predicted value for the property of the portion of the batch of sludge comprises generating instructions for presentation of the predicted value for the property of the portion of the batch of sludge in a user interface; and
receiving the actual value for the property of the portion of the batch of sludge comprises receiving data indicating user input of the actual value for the property of the portion of the batch of sludge.

15. The system of claim 14 , wherein generating the instructions for presentation of the predicted value for the property of the portion of the batch of sludge in a user interface comprises generating the instructions for presentation of a user interface that includes a filter option to enable a user to view details about multiple different batches of sludge, including the batch of sludge, and to filter details about batches of sludge using the predicted value for the property of the portion of the respective batch of sludge.

Assignees

• Accenture Global Solutions Ltd

Inventors

• Lamberti Corinna
• Goldklang Katherine
• Riley Robert

Classifications

• G06N20/00

  Machine learning · CPC title

• G01N33/18 Primary

  Water · CPC title

• G01N21/359

  using near infrared light · CPC title