公开(公告)号：US20210405631A1

公开(公告)日：2021-12-30

申请号：US17361990

申请日：2021-06-29

Applicant: Honeywell International Inc.

Inventor： Sanjay Kantilal Dave ,  Alanksha Jain ,  Viraj Srivastava ,  Vijoy Akavalappil

IPC: G05B23/02 ,  G06N3/04 ,  G06N3/08

Abstract: A trained machine learning algorithm processes time series production data. The time series production data are representative of a control process within a facility control loop. The machine learning training algorithm is trained using positive training data that are representative of a normal operation of components within the facility control loop and negative training data that are representative of an abnormal operation of components within the facility control loop. Output of the trained machine learning algorithm identifies abnormalities in the facility control loop.

公开(公告)号：US20240412079A1

公开(公告)日：2024-12-12

申请号：US18208360

申请日：2023-06-12

Applicant: HONEYWELL INTERNATIONAL INC.

Inventor： Viraj Srivastava ,  Minal Dani ,  Chinmaya Kar

IPC: G06N5/022

Abstract: Examples techniques to select training data to train Artificial Intelligence models to monitor industrial processes are described. From historical data relating to an industrial process, a range of values exhibited by operating parameters of the industrial process under normal operation is estimated. One or more steady time windows are identified for the operating parameters. A steady time window of an operating parameter is a duration of time where values of the operating parameter are within the estimated range of values. Based on the identified steady time windows, a composite steady time window is determined. The composite steady time window is a duration of time where a maximum of the identified steady time windows overlap. The data corresponding to the composite steady time window is provided as training data to the AI model.

公开(公告)号：US20240257276A1

公开(公告)日：2024-08-01

申请号：US18420783

申请日：2024-01-24

Applicant: HONEYWELL INTERNATIONAL INC.

Inventor： Viraj Srivastava ,  Praveen Sam ,  Niranjan Amrutur Subba Rao

IPC: G06Q50/04 ,  H01M10/04

CPC classification number: G06Q50/04 ,  H01M10/04

Abstract: Example methods, apparatuses, systems, and computer program products are provided. For example, an example computer-implemented method includes receiving a battery manufacturing operation parameter indicator, determining whether the battery manufacturing operation parameter indicator satisfies a battery manufacturing parameter threshold indicator, and in response to determining that the battery manufacturing operation parameter indicator does not satisfy the battery manufacturing parameter threshold indicator, the example computer-implemented method comprises: generating a battery manufacturing deviation event data object, and generating a battery manufacturing adjustment data object based at least in part on inputting the battery manufacturing deviation event data object to one or more machine learning models.

公开(公告)号：US20210382470A1

公开(公告)日：2021-12-09

申请号：US17241615

申请日：2021-04-27

Applicant: Honeywell International Inc.

Inventor： Prangya Priyadarsini ,  Suresh Thombare ,  Viraj Srivastava

IPC: G05B23/02 ,  G06N20/00

Abstract: A system processes historical facility data that relate to facility states and modes of operation. The historical facility data are clustered into groups representing the facility states and the modes of operation. The groups are used to determine a current state and mode of the facility. When the facility is in a normal state, the system determines whether an event in the facility is an abnormality. If an abnormality is identified, the system transmits a signal indicating the abnormality.

公开(公告)号：US10809704B2

公开(公告)日：2020-10-20

申请号：US16049372

申请日：2018-07-30

Applicant: HONEYWELL INTERNATIONAL INC.

Inventor： Michael Niemiec ,  Viraj Srivastava ,  Greg Stewart

IPC: G06F19/00 ,  G05B19/418 ,  G05B23/02 ,  G05B15/02

Abstract: A method of alarm notification includes providing data for an industrial process including stored alarm event data and stored Key Performance Indicator (KPI) data, and an alarm event-KPI correlation and operator notification system. Patterns of relationships are discovered between the stored alarm event and KPI data to provide a reference pattern database that identifies KPI violation events in the KPI data as a function of the alarm event data or alarm events in the alarm event data as a function of the KPI data. Pattern matching uses a real-time alarm event and real-time KPI data to determine a pattern similarity by comparing a current sequence spanning a predetermined time of the real-time alarm event and KPI data to the patterns. When the pattern matching identifies a current alarm event or KPI violation event, an operator is notified and a predicted KPI value or feature or a predicted alarm event.