公开(公告)号：US11836636B2

公开(公告)日：2023-12-05

申请号：US16443914

申请日：2019-06-18

Applicant: ABB Schweiz AG

Inventor： Andrew Cohen ,  Marcel Dix

IPC: G05B19/4155 ,  G05B23/02 ,  G06N5/02 ,  G06N7/01

CPC classification number: G06N5/02 ,  G05B19/4155 ,  G05B23/0254 ,  G06N7/01 ,  G05B2219/33139

Abstract: Disclosed is a computer-implemented method for generating a prediction model. The model can be for use in processing machine event data to predict behavior of a plurality of industrial machines under supervision. The prediction model can be configured to determine current and future states of the industrial machines. The method can include: extracting event features from event codes and structuring the event features into feature vectors; and generating the prediction model by clustering the feature vectors into a plurality of vector clusters, the vector clusters being assigned to respective machine states. The prediction model can be constructed based on event data from a first industrial machine and be applied to control an operating state of a second industrial machine.

公开(公告)号：US20230251635A1

公开(公告)日：2023-08-10

申请号：US18300782

申请日：2023-04-14

Applicant: ABB Schweiz AG

Inventor： Sten Gruener ,  Mario Hoernicke ,  Katharina Stark ,  Roland Braun ,  Michael Vach ,  Nicolai Schoch ,  Marcel Dix

IPC: G05B19/418

CPC classification number: G05B19/41845 ,  G05B19/41885 ,  G05B2219/32097 ,  G05B2219/32124

Abstract: A method of integrating modules into a hybrid modular plant comprising a discrete manufacturing part and a continuous manufacturing part includes integrating the discrete part into the continuous part, comprising constructing at least one module definition file mapping one or more discrete-part units of the discrete part to a continuous-part module and importing the module definition file into an orchestration layer of the continuous part. Alternatively, the method comprises integrating the continuous part into the discrete part, comprising constructing one or more interfaces representing each continuous-part module as one or more respective discrete-part units.

公开(公告)号：US20230034769A1

公开(公告)日：2023-02-02

申请号：US17966012

申请日：2022-10-14

Applicant: ABB Schweiz AG

Inventor： Moncef Chioua ,  Marcel Dix ,  Benjamin Kloepper ,  Ioannis Lymperopoulos ,  Dennis Janka ,  Pablo Rodriguez

IPC: G05B13/02 ,  G05B23/02

Abstract: A method and computer program product including training a machine learning model by means of input data and score data, wherein the machine learning model is an artificial neural net, ANN; running the trained machine learning model by applying the first time-series to the trained machine learning model; and outputting, by the trained machine learning model, an output value, comprising at least a second criticality value of the at least one predicted observable process-value indicative of the abnormal behaviour of the industrial process in a predefined temporal distance.

公开(公告)号：US20230014857A1

公开(公告)日：2023-01-19

申请号：US17952987

申请日：2022-09-26

Applicant: ABB Schweiz AG

Inventor： Fan Dai ,  Nuo Li ,  Marcel Dix ,  Dongliang Cao

IPC: B25J9/16 ,  G06F3/0486 ,  G06F3/04815

Abstract: A system and method for programming a robot includes providing a 3D representation of workpieces to be handled by the robot, and of a working environment; synthesizing and displaying a view of the working environment comprising an image of the workpieces at respective initial positions; identifying matching features of the selected workpiece and of the working environment which are able to cooperate to hold the workpiece in a final position in the working environment, and a skill by which the matching features can be brought to cooperate; identifying an intermediate position from where applying the skill to the workpiece moves the workpiece to the final position; and adding to a motion program for the robot a routine for moving the workpiece from its initial position to the intermediate position and for applying the skill to the workpiece at the intermediate position.

公开(公告)号：US20220343193A1

公开(公告)日：2022-10-27

申请号：US17724693

申请日：2022-04-20

Applicant: ABB Schweiz AG

Inventor： Divyasheel Sharma ,  Benjamin Kloepper ,  Marco Gaertler ,  Dawid Ziobro ,  Simon Linge ,  Pablo Rodriguez ,  Matthias Berning ,  Reuben Borrison ,  Marcel Dix ,  Benedikt Schmidt ,  Hadil Abukwaik ,  Arzam Muzaffar Kotriwala ,  Sylvia Maczey ,  Jens Doppelhamer ,  Chandrika K R ,  Gayathri Gopalakrishnan

IPC: G06N5/04

Abstract: A decision support system and method for an industrial plant is configured and operates to: obtain a causal graph modeling causal assumptions relating to conditional dependence between variables in the industrial plant; obtain observational data relating to operation of the industrial plant; and perform causal inference using the causal graph and the observational data to estimate at least one causal effect relevant for making decisions when operating the industrial plant.

公开(公告)号：US20190165989A1

公开(公告)日：2019-05-30

申请号：US15822784

申请日：2017-11-27

Applicant: ABB Schweiz AG

Inventor： Sam Ade Jacobs ,  Mithun P. Acharya ,  Veronika Domova ,  Marcel Dix ,  Aldo Dagnino ,  Jinendra K. Gugaliya ,  Kaushik Ghosh

IPC: H04L12/24 ,  G08B29/18 ,  G05B23/02

Abstract: Unique systems, methods, techniques and apparatuses of an alarm management system are disclosed herein. One exemplary embodiment is a method for monitoring an industrial plant comprising determining a sequence of alarm events for each of a plurality of time intervals including a first alarm event of a plurality of alarm events and a second alarm event of the plurality of alarm events; determining a count of the first alarm events and second alarm events; determining the alarm events exceed a support threshold value; determining a third count of a sub-sequence of the sequences of alarm events including the first alarm event followed by the second alarm event in response to determining the first count and the second count exceeds the support threshold value; determining a ratio using the first count, the second count, and the third count exceeds a display threshold value; and displaying the sub-sequence.

公开(公告)号：US20250053885A1

公开(公告)日：2025-02-13

申请号：US18928402

申请日：2024-10-28

Applicant: ABB Schweiz AG

Inventor： Benedikt Schmidt ,  Benjamin Kloepper ,  Arzam Muzaffar Kotriwala ,  Yemao Man ,  Dawid Ziobro ,  Gayathri Gopalakrishnan ,  Joakim Astrom ,  Marcel Dix ,  Divyasheel Sharma

IPC: G06N20/20

Abstract: A method for explanation of machine learning results based on using a model collection includes training at least two machine learning models with at least two competing strategies for the at least one dataset; and using the least two machine learning models to yield at least two different predictions and/or at least two explanations for the at least one dataset.

公开(公告)号：US12141707B2

公开(公告)日：2024-11-12

申请号：US18498413

申请日：2023-10-31

Applicant: ABB Schweiz AG

Inventor： Andrew Cohen ,  Marcel Dix

IPC: G05B23/02 ,  G05B19/4155 ,  G06N5/02 ,  G06N5/04 ,  G06N7/01 ,  G06N20/00

Abstract: Disclosed is a method for generating a prediction model. The model can be used in processing machine event data to predict behavior of a plurality of industrial machines under supervision. The prediction model can be configured to determine current and future states of the industrial machines. The method can include extracting event features from event codes and structuring the event features into feature vectors. A first dimension of a first feature vector corresponds to a first event feature, and a second dimension of the first feature vector corresponds to a second event feature. The method can also include generating the prediction model by clustering the feature vectors into a plurality of vector clusters, the vector clusters assigned to respective machine states. The prediction model can be constructed based on event data from a first industrial machine and be applied to control an operating state of a second industrial machine.

公开(公告)号：US20240302832A1

公开(公告)日：2024-09-12

申请号：US18668370

申请日：2024-05-20

Applicant: ABB Schweiz AG

Inventor： Hadil Abukwaik ,  Divyasheel Sharma ,  Benjamin Kloepper ,  Arzam Muzaffar Kotriwala ,  Pablo Rodriguez ,  Benedikt Schmidt ,  Ruomu Tan ,  Chandrika K R ,  Reuben Borrison ,  Marcel Dix ,  Jens Doppelhamer

IPC: G05B23/02

CPC classification number: G05B23/0254 ,  G05B23/027 ,  G05B23/0286

Abstract: A method for training a prediction model includes obtaining training samples representing states of the process that do not cause the undesired event; obtaining based on a process model and a set of predetermined rules that stipulate states having an increased likelihood of the undesired event occurring; training samples representing states with an increased likelihood to cause the undesired event; providing samples to the to-be-trained prediction model to obtain a prediction of the likelihood for occurrence of the undesired event in a state of the process represented by the respective sample; rating a difference between the prediction and the label of the respective sample using a predetermined loss function; and optimizing parameters such that, when predictions are made, the rating by the loss function improves.

公开(公告)号：US20240062077A1

公开(公告)日：2024-02-22

申请号：US18498413

申请日：2023-10-31

Applicant: ABB Schweiz AG

Inventor： Andrew Cohen ,  Marcel Dix

IPC: G06N5/02 ,  G05B19/4155 ,  G05B23/02 ,  G06N7/01

CPC classification number: G06N5/02 ,  G05B19/4155 ,  G05B23/0254 ,  G06N7/01 ,  G05B2219/33139

Abstract: Disclosed is a method for generating a prediction model. The model can be used in processing machine event data to predict behavior of a plurality of industrial machines under supervision. The prediction model can be configured to determine current and future states of the industrial machines. The method can include extracting event features from event codes and structuring the event features into feature vectors. A first dimension of a first feature vector corresponds to a first event feature, and a second dimension of the first feature vector corresponds to a second event feature. The method can also include generating the prediction model by clustering the feature vectors into a plurality of vector clusters, the vector clusters assigned to respective machine states. The prediction model can be constructed based on event data from a first industrial machine and be applied to control an operating state of a second industrial machine.