Cell control system

• US 10,642,247 B2
• Filed: 08/24/2017
• Issued: 05/05/2020
• Est. Priority Date: 08/25/2016
• Status: Active Grant

• Current Assignee: Fanuc Corporation
• Original Assignee: Fanuc Corporation
• Inventors: Saitou, Manabu
• Primary Examiner(s): Lee, Thomas C
• Assistant Examiner(s): Tang, Michael
• Application Number:
  US15/685,129
• Publication Number:
  US 20180059639A1
• Time in Patent Office: 985 Days
• CPC Class Codes: G05B 19/406 characterised by monitoring or safety G05B19/19 takes precedence
• CPC Class Codes: G05B 2219/33296 ANN for diagnostic, monitoring
• CPC Class Codes: G05B 2219/36487 Record position, motion and sound
• CPC Class Codes: G05B 2219/37337 Noise, acoustic emission, sound
• CPC Class Codes: G06N 20/00 Machine learning
• CPC Class Codes: G06N 3/02 Neural networks
• CPC Class Codes: Y02P 90/02 Total factory control, e.g. smart factories, flexible manufacturing systems [FMS] or integrated manufacturing systems [IMS]

Abstract

A cell control system capable of estimating a cause of an alarm by estimating an influence of noise in a plurality of machines includes a machine operation instruction unit for transmitting an operation instruction to a managed manufacturing machine, a noise value collection unit for collecting detected noise information, an operation information collection unit for collecting operation information of a manufacturing machine, a learning unit for creating a learning model by performing machine learning using the collected operation information collected as an input signal and the detected noise information as an instruction signal, an estimation unit for analyzing the learning model to estimate operation information corresponding to a noise factor, and an operation instruction change unit for instructing the machine operation instruction unit to change instruction content based on the operation information corresponding to the noise factor.

Claims

1. A cell control system, comprising:

at least one manufacturing cell including a first manufacturing machine and a second manufacturing machine; and

a cell controller for transmitting an operation instruction to the first manufacturing machine and the second manufacturing machine based on a manufacturing plan received from a production planning device,

wherein the cell controller includes

a machine operation instruction unit for transmitting the operation instruction to the first manufacturing machine and the second manufacturing machine based on the manufacturing plan,

a noise value collection unit for collecting a first noise value related to noise detected by the first manufacturing machine and a second noise value related to noise detected by the second manufacturing machine,

an operation information collection unit for collecting first operation information of the first manufacturing machine and second operation information of the second manufacturing machine,

a learning unit for creating a learning model by

performing machine learning using (i) the first operation information and the second operation information collected by the operation information collection unit as an input signal and (ii) the first noise value collected by the noise value collection unit as an instruction signal, and

performing machine learning using (a) the first operation information and the second operation information collected by the operation information collection unit as an input signal and (b) the second noise value collected by the noise value collection unit as an instruction signal,

an estimation unit for analyzing the learning model to estimate operation information corresponding to a cause of the noise detected by the first manufacturing machine and operation information corresponding to a cause of the noise detected by the second manufacturing machine, the estimation unit being configured to estimate a trend in a change of the noise value due to a change of the operation information, and

an operation instruction change unit for instructing the machine operation instruction unit to change instruction content of the operation instruction sent to at least one of the first manufacturing machine or the second manufacturing machine, based on the operation information corresponding to the cause of the noise estimated by the estimation unit.

2. The cell control system according to claim 1, wherein

the estimation unit is configured to perform a prediction using the learning model created by the learning unit to estimate a manufacturing machine having a low noise immunity between the first manufacturing machine and the second manufacturing machine, and

the operation instruction change unit is configured to instruct the machine operation instruction unit to change the instruction content of the operation instruction based on information about the manufacturing machine having the low noise immunity estimated by the estimation unit.

3. The cell control system according to claim 2, wherein

the operation instruction change unit is configured to instruct the machine operation instruction unit to change the instruction content of the operation instruction estimated to affect the manufacturing machine having the low noise immunity estimated by the estimation unit.

4. A cell control system, comprising:

at least one manufacturing cell including at least one manufacturing machine; and

a cell controller for transmitting an operation instruction to the manufacturing machine based on a manufacturing plan received from a production planning device,

wherein

the cell controller includes;

a machine operation instruction unit for transmitting the operation instruction to the manufacturing machine based on the manufacturing plan,

a noise value collection unit for collecting a noise value related to noise detected by the manufacturing machine,

an operation information collection unit for collecting operation information of the manufacturing machine,

a learning unit for creating a learning model by performing machine learning using the operation information collected by the operation information collection unit as an input signal and the noise value collected by the noise value collection unit as an instruction signal,

an estimation unit for analyzing the learning model created by the learning unit to estimate operation information corresponding to a cause of the noise detected by the manufacturing machine, the estimation unit being configured to estimate a trend in a change of the noise value due to a change of the operation information, and

an operation instruction change unit for instructing the machine operation instruction unit to change instruction content of the operation instruction, based on the operation information corresponding to the cause of the noise estimated by the estimation unit,

the at least one manufacturing machine includes a plurality of manufacturing machines, and

the estimation unit is configured to estimate a manufacturing machine having a low noise immunity among the plurality of manufacturing machines by;

selecting one piece of the operation information collected by the operation information collection unit,

inputting the selected piece of the operation information as an input value to the learning model,

calculating a slope of the input value at which each output value increases,

varying each input value in a direction of the calculated slope, the varying comprising increasing or decreasing,

repeating the inputting, the calculating and the varying until one of output values exceeds a predetermined threshold value, and

estimating the manufacturing machine corresponding to the output value exceeding the predetermined threshold value as the manufacturing machine having the low noise immunity.