Back to results
US 11,899,440 B2Granted
Remote monitoring
AssigneeABB Schweiz AG
InventorsOlli Alkkiomäki, Teemu Tanila, Joni Siimesjärvi
Life of this patent
Estimating the term…
Abstract
To provide a status information also to one or more industrial devices for which no process data is available, a machine learning model is trained by using process data of a subset of industrial devices, corresponding product data, and statuses obtained by performing remote monitoring analysis to the process data. When user input including first information, which at least indicate at least one industrial device type is received, product data of one or more industrial automation devices, which are of the same indicated industrial device type is retrieved and inputted to the trained model, which outputs one or more estimated statuses.
Claims15
A computer-implemented method comprising:
- receiving a user input comprising industrial device type information, which at least indicates at least one industrial device type;
- retrieving, using the industrial device type information, from a product database comprising product data of a plurality of industrial devices, product data of one or more industrial devices, which are of the same industrial device type as the at least one industrial device type indicated;
- determining, for each industrial device whose product data is retrieved, whether or not process data is available for the industrial device;
- in response to process data being available, performing a remote monitoring analysis to the process data to obtain and output one or more predictions of the at least one status;
- in response to no process data being available, performing the following:
- inputting the retrieved product data to a trained model, wherein the trained model utilizes machine learning and wherein the trained model has been trained using process data collected by a subset of the plurality of the industrial devices, predictions of at least one status generated by a remote monitoring analysis tool from the process data, and product data of the subset of the plurality of the industrial devices;
- processing, by the trained model, the inputted product data to one or more estimated predictions of the at least one status;
- and
- outputting the one or more estimated predictions of the at least one status.
6 more elements
A system for obtaining estimated predictions of at least one status, the system comprising at least:
- one or more industrial devices;
- a product database comprising product data at least on the one or more industrial devices;
- and
- one or more apparatuses, wherein an apparatus comprises at least one processor;
- and at least one memory including computer program code, the at least one memory and computer program code configured to, with the at least one processor, cause the apparatus at least to perform:
- retrieving, in response to a user input comprising industrial device type information, which at least indicates at least one industrial device type, using the industrial device type information, from a product database comprising product data of a plurality of industrial devices, product data of one or more industrial devices, which are of the same industrial device type as the at least one industrial device type indicated;
- determining, after retrieving the product data, for each industrial device whose product data is retrieved, whether or not process data is available for the industrial device;
- in response to process data being available, performing a remote monitoring analysis to the process data to obtain and output one or more predictions of the at least one status;
- in response to no process data being available for the industrial device, performing:
- inputting the retrieved product data to a trained model, wherein the trained model utilizes machine learning and wherein the trained model has been trained using process data collected by a subset of the plurality of the industrial devices, predictions of at least one status generated by a remote monitoring analysis tool from the process data, and product data of the subset of the plurality of the industrial devices, and wherein the trained model processes the inputted product data to one or more estimated predictions of the at least one status;
- and
- outputting the one or more estimated predictions of the at least one status.
9 more elements
An apparatus comprising:
- at least one processor;
- and
- at least one memory including computer program code, the at least one memory and computer program code configured to, with the at least one processor, cause the apparatus at least to perform:
- retrieving, in response to a user input comprising industrial device type information, which at least indicates at least one industrial device type, using the industrial device type information, from a product database comprising product data of a plurality of industrial devices, product data of one or more industrial devices, which are of the same industrial device type as the at least one industrial device type indicated;
- determining, after retrieving the product data, for each industrial device whose product data is retrieved, whether or not process data is available for the industrial device;
- in response to process data being available, performing a remote monitoring analysis to the process data to obtain and output one or more predictions of the at least one status;
- in response to no process data being available for the industrial device, performing:
- inputting the retrieved product data to a trained model, wherein the trained model utilizes machine learning and wherein the trained model has been trained using process data collected by a subset of the plurality of the industrial devices, predictions of at least one status generated by a remote monitoring analysis tool from the process data, and product data of the subset of the plurality of the industrial devices, and wherein the trained model processes the inputted product data to one or more estimated predictions of the at least one status;
- and
- outputting the one or more estimated predictions of the at least one status.
7 more elements
A non-transitory computer readable storage medium storing one or more instructions which, when executed by one or more processors, cause the one or more processors to carry out at least:
- retrieving, in response to a user input comprising industrial device type information, which at least indicates at least one industrial device type, using the industrial device type information, from a product database comprising product data of a plurality of industrial devices, product data of one or more industrial devices, which are of the same industrial device type as the at least one industrial device type indicated;
- determining, after retrieving the product data, for each industrial device whose product data is retrieved, whether or not process data is available for the industrial device;
- in response to process data being available, performing a remote monitoring analysis to the process data to obtain and output one or more predictions of the at least one status;
- in response to no process data being available, performing:
- inputting the retrieved product data to a trained model, wherein the trained model utilizes machine learning and wherein the trained model has been trained using process data collected by a subset of the plurality of the industrial devices, predictions of at least one status generated by a remote monitoring analysis tool from the process data, and product data of the subset of the plurality of the industrial devices and wherein the trained model processes the inputted product data to one or more estimated predictions of the at least one status;
- and
- outputting the one or more estimated predictions of the at least one status.
4 more elements
Dependencies are inferred from the claim text.