# US 2022/0036207 A1 · Apparatus and method for drive selection using machine learning

> Agents: this data is available as JSON from the patentgrep API and MCP server — see https://patentgrep.com/llms.txt

- Number: US 2022/0036207 A1
- Document ID: US20220036207A1
- Status: Published application
- Kind: A1
- Application number: 17/505,665
- Filed: 2021-10-20
- Published: 2022-02-03
- Earliest claimed: 2019-04-25
- Assignees: ABB Schweiz AG
- Applicants: ABB Schweiz AG
- Inventors: Pablo Rodriguez; Jens Doppelhamer; Katharina Stark; Prerna Juhlin
- Web page: https://patentgrep.com/patent/US20220036207A1

## CPC

- G06N 5/025: Extracting rules from data
- G06N 3/09: Supervised learning
- G06N 3/091: Active learning
- G06N 20/00: Machine learning
- G06N 3/04: Architecture, e.g. interconnection topology
- G06N 7/01: Probabilistic graphical models, e.g. probabilistic networks
- G06N: COMPUTING ARRANGEMENTS BASED ON SPECIFIC COMPUTATIONAL MODELS

## Abstract

A computer-implemented method of training an artificial intelligence module for selecting at least one component of an industrial drive system and an apparatus that includes a correspondingly trained AI module are described. The method includes providing an initial training data set that includes a set of rules, where each rule is indicative of a selection decision for selecting at least one component of the industrial drive system based on at least one usage requirement for the industrial drive system. Further, a modified training data set is generated based on supplementing the initial training data set with at least one further selection decision for selecting the at least one component of the drive system based on at least one of simulation data, empiric data, and operational data. Further, the AI module is trained based on the modified training data set.

## Claims

### Claim 1

A computer-implemented method of training an artificial intelligence module, AI module, for selecting at least one component of an industrial drive system, the method comprising:
providing, on data storage, an initial training data set comprising a set of rules, wherein each rule is indicative of a selection decision for selecting at least one component of the industrial drive system based on at least one usage requirement for the industrial drive system;
generating a modified training data set based on supplementing the initial training data set with at least one further selection decision for selecting the at least one component of the drive system based on at least one of simulation data relating to a simulation of at least a part of the drive system, empiric data relating to a selection of the at least one component, and operational data indicative of at least one operational characteristic of at least one installed industrial drive system; and
training the AI module based on the modified training data set.

### Claim 2 (depends on 1)

The method according to claim 1, wherein providing the initial training data set comprises:
generating the initial training data set using a software tool for deterministically selecting one or more components of the at least one component of the industrial drive system based on one or more usage requirements of the at least one usage requirement for the industrial drive system.

### Claim 3 (depends on 1)

The method according to claim 1,
wherein each rule of the set of rules of the initial training data set comprises at least one pair of data entries with a first data element indicative of the at least one usage requirement and a second data element indicative of the at least one component of the industrial drive system fulfilling the at least one usage requirement.

### Claim 4 (depends on 1)

The method according to claim 1,
wherein the at least one component of the industrial drive system is at least one of a drive, a motor, a load, a transformer, a gearbox, a pump, a ventilation device, a heating device, an air conditioning device, a controller, a motion control, and a machinery of the drive system.

### Claim 5 (depends on 1)

The method according to claim 1,
wherein the at least one usage requirement for the industrial drive system is indicative of a specification of at least one of a drive, a motor, a load, a transformer, a gearbox, a pump, a ventilation device, a heating device, an air conditioning device, a controller, a motion control, and a machinery of the drive system.

### Claim 6 (depends on 1)

The method according to claim 1, further comprising:
weighting the at least one further selection decision for selecting the at least one component of the industrial drive system relative to one or more rules of the set of rules of the initial training data set.

### Claim 7 (depends on 1)

The method according to claim 1, further comprising:
weighting at least a part of the simulation data, the empiric data and the operational data with at least partly differing weights and/or weighting factors.

### Claim 8 (depends on 1)

The method according to claim 1, wherein supplementing the initial training data set comprises:
adjusting at least one rule of the set of rules of the initial training data set based on the at least one further selection decision; and/or
adjusting at least one rule of the set of rules of the initial training data set based on at least one of the simulation data, the empiric data, and the operational data.

### Claim 9 (depends on 1)

The method according to claim 1, wherein supplementing the initial training data set comprises:
replacing at least one rule of the set of rules of the initial training data set based on the at least one further selection decision and/or based on at least one of the simulation data, the empiric data, and the operational data.

### Claim 10 (depends on 1)

The method according to claim 1, wherein supplementing the initial training data set comprises:
deriving the at least one further selection decision from at least one of the simulation data, the empiric data, and the operational data.

### Claim 11 (depends on 1)

The method according to claim 1,
wherein the operational data comprise at least one of maintenance data indicative of a maintenance procedure performed on the at least one installed drive system, sensor data of at least one senor of the at least one installed drive system, disturbance data indicative of a disturbance of the at least one installed drive system, time series data indicative of a temporal development of the at least one installed drive system, and operator data indicative of at least one operator action performed on the at least one installed drive system.

### Claim 12 (depends on 1)

The method according to claim 1, wherein supplementing the initial training data set comprises:
identifying an inappropriate selection decision for selecting the at least one component of the industrial drive system based on at least one of the simulation data, the empiric data, and the operational data; and
selecting the at least one further selection decision taking into account the identified inappropriate selection decision.

### Claim 13 (depends on 1)

The method according to claim 1, further comprising:
generating the simulation data to supplement the initial training data set.

### Claim 14 (depends on 1)

A method of using an artificial intelligence module, the method comprising selecting at least one component of the industrial drive system, wherein the artificial intelligence module is trained according to the method of claim 1.

### Claim 15 (depends on 1)

An apparatus for selecting at least one component of an industrial drive system, the apparatus comprising:
an artificial intelligence module, AI module, wherein the AI module is trained according to the method of claim 1.
