Classifications

• • G—PHYSICS
  • G05—CONTROLLING; REGULATING
  • G05B—CONTROL OR REGULATING SYSTEMS IN GENERAL; FUNCTIONAL ELEMENTS OF SUCH SYSTEMS; MONITORING OR TESTING ARRANGEMENTS FOR SUCH SYSTEMS OR ELEMENTS
  • G05B23/00—Testing or monitoring of control systems or parts thereof
  • G05B23/02—Electric testing or monitoring
  • G05B23/0205—Electric testing or monitoring by means of a monitoring system capable of detecting and responding to faults
  • G05B23/0218—Electric testing or monitoring by means of a monitoring system capable of detecting and responding to faults characterised by the fault detection method dealing with either existing or incipient faults
  • G05B23/0243—Electric testing or monitoring by means of a monitoring system capable of detecting and responding to faults characterised by the fault detection method dealing with either existing or incipient faults model based detection method, e.g. first-principles knowledge model
  • G05B23/0254—Electric testing or monitoring by means of a monitoring system capable of detecting and responding to faults characterised by the fault detection method dealing with either existing or incipient faults model based detection method, e.g. first-principles knowledge model based on a quantitative model, e.g. mathematical relationships between inputs and outputs; functions: observer, Kalman filter, residual calculation, Neural Networks
• • G—PHYSICS
  • G05—CONTROLLING; REGULATING
  • G05B—CONTROL OR REGULATING SYSTEMS IN GENERAL; FUNCTIONAL ELEMENTS OF SUCH SYSTEMS; MONITORING OR TESTING ARRANGEMENTS FOR SUCH SYSTEMS OR ELEMENTS
  • G05B2219/00—Program-control systems
  • G05B2219/20—Pc systems
  • G05B2219/24—Pc safety
  • G05B2219/24042—Signature analysis, compare recorded with current data, if error then alarm
• • G—PHYSICS
  • G05—CONTROLLING; REGULATING
  • G05B—CONTROL OR REGULATING SYSTEMS IN GENERAL; FUNCTIONAL ELEMENTS OF SUCH SYSTEMS; MONITORING OR TESTING ARRANGEMENTS FOR SUCH SYSTEMS OR ELEMENTS
  • G05B2219/00—Program-control systems
  • G05B2219/20—Pc systems
  • G05B2219/25—Pc structure of the system
  • G05B2219/25235—Associate a sequence function to each control element, event signature
• • G—PHYSICS
  • G05—CONTROLLING; REGULATING
  • G05B—CONTROL OR REGULATING SYSTEMS IN GENERAL; FUNCTIONAL ELEMENTS OF SUCH SYSTEMS; MONITORING OR TESTING ARRANGEMENTS FOR SUCH SYSTEMS OR ELEMENTS
  • G05B2219/00—Program-control systems
  • G05B2219/20—Pc systems
  • G05B2219/26—Pc applications
  • G05B2219/2613—Household appliance in general

Definitions

• the invention relates to the detection of an anomaly in a household appliance.
• the invention relates to the automatic determination of an anomaly in the household appliance.
• a household appliance includes a control device that is set up to record device data of the household appliance during its operation.
• the household appliance can include one or more sensors, the readings of which are recorded by the control device.
• Additional device data can include circumstances under which the measured values were created, for example a function of the household appliance that is currently being carried out, a date or a time.
• the recorded device data can be stored in a local memory in the manner of a log file. For reasons of space, the stored device data can be overwritten cyclically, or only device data that meet a predetermined condition are stored.
• the stored appliance data are usually only analyzed if the household appliance shows a malfunction.
• a possible connection between the malfunction and previously stored device data is usually checked by a service technician. To do this, the household appliance usually has to be transported to the service technician or vice versa.
• there is usually no large-scale analysis of the stored device data so it can be difficult to draw conclusions about the behavior of a household device or its future behavior or that of another. As a result, detecting and, if necessary, rectifying an anomaly in a household appliance can be difficult and require a great deal of human experience.
• a method for detecting an anomaly in a household appliance comprises the steps of acquiring device data of the household appliance during its operation; determining that the device data does not correspond to a common device state; determining similarities of the captured device data to predetermined patterns of device data; and associating the captured device data with one of the templates.
• the recorded device data can be automatically assigned to one of the patterns.
• Each of the patterns can represent a predetermined behavior, an anomaly, a problem, or an aging state of the household appliance.
• the pattern to which the device data most closely resembles is preferably selected. More preferably, the assignment only takes place if the similarity of the device data to the next less similar pattern is below a predetermined threshold value.
• the home appliance can be better assessed based on the acquired device data, and a possible anomaly can be better determined.
• a description of a device status of the household appliance is preferably assigned to a pattern.
• the description can be in text form or in a machine-processable form, which can also be called binary form.
• the description may be provided when the device data is associated with the template.
• the device status can include an error, it being possible for an indication of the error to be output.
• the information can be directed to a user of the household appliance or to a technician.
• the description can have an explanatory effect, for example "pump defective”.
• the description can include a reference to an improvement in the condition of the device, for example “replace the pump, carry out a leak test”.
• the pattern to which the detected device data is assigned is then expanded to include the detected device data.
• the device data can be recognized even more easily, quickly or better by the pattern in the future.
• use can be made of machine pattern recognition and machine learning methods.
• recognition security can be improved in the long term by recording the device data.
• unsupervised learning methods can be used to automatically improve the present method.
• the pattern can be determined that the pattern has been extended by a predetermined number of detected device data.
• the predetermined number can be relative to the number of device data on which the pattern was originally created.
• the pattern may be originally created based on labeled device data. If the amount of device data on which it is based has been increased by around 20%, for example, then the pattern can be checked. The check can be automatic or manual.
• the automatic check can include checking whether the device data on which the pattern is based are consistent with one another. In other words, it can be checked whether the device data on which the pattern is based are sufficiently similar to one another and are sufficiently different from other patterns. Measures of similarity can be predetermined for this purpose. It may be possible to recreate the pattern based on a subset of device data. Discarded device data can be assigned to a different pattern.
• a new template is created if the collected device data is not sufficiently similar to any of the existing templates.
• device data that is similar to one another can be collected, even if it is not yet known which device status or which device behavior is associated with it.
• a broad database can be created in an improved manner, which allows the domestic appliance to be analyzed. For example, behavior of the household appliance that only occurs sporadically or only after a long period of operation can be tracked down in an improved manner in this way.
• the method can work particularly effectively if the patterns used initially recognize or represent the device states assigned to them in an improved manner.
• the initial patterns can be created on the basis of laboratory tests.
• a pattern is determined on the basis of appliance data that is recorded when the household appliance is in operation after a predetermined period of operation.
• the household appliance can be subjected to a long-term test in which operating times of different lengths can be generated or simulated.
• a pattern is determined on the basis of appliance data that is recorded when the household appliance is operated under various predetermined environmental conditions.
• the household appliance can be operated in an environmental laboratory.
• the environmental laboratory can include, for example, a climatic chamber or a vibrating table. In this way, different climatic or physical conditions can be created under which the household appliance can be systematically observed.
• the climate chamber can generate different temperatures or different relative humidity levels.
• Other possible environmental factors that can be generated in the climate chamber include, for example, infrared radiation or ultraviolet radiation.
• Other conditions that may be varied include different supplies or fluctuations in a supply voltage. Other operating conditions can also be established accordingly.
• a device for controlling a household appliance includes a device for acquiring device data of the household appliance during its operation; a device for determining similarities of the detected device data with predetermined patterns; and a device for assigning the recorded device data to one of the patterns.
• the device can include a control device for the household appliance.
• the control device can be included in the household appliance.
• the detection and, if necessary, correction of an anomaly can thus be carried out in an improved manner using local means.
• measures for predictive maintenance or for correcting simple errors can be carried out directly by a user of the household appliance. Intervention by a trained person or the use of special tools or techniques may not be required.
• One or more of the devices can be implemented by a processing device which, in one embodiment, is set up to partially or completely carry out a method described herein.
• the processing device may comprise a programmable microcomputer or microcontroller and the method may be in the form of a computer program product having program code means.
• the computer program product can be stored on a computer-readable data carrier. Additional features or advantages of the method can be transferred to the device and vice versa.
• the device for determining similarities of the recorded device data with predetermined patterns comprises an artificial neural network.
• the neural network can be trained on the basis of predetermined device data to recognize initial patterns.
• the pattern can be adapted by the device itself, or new device data can be sent to a remote device, which trains an artificial neural network based on them and can transmit them back to the household device.
• the household appliance can replace or update the existing neural network with a received one.
• a household appliance comprises a device as described herein.
• a central point comprises a device for receiving appliance data which were recorded on various household appliances which are comparable to one another while they were in operation; a device for determining similarities of the detected device data with predetermined patterns; and a device for assigning the recorded device data to one of the patterns.
• the household appliances can in particular be examples of the same series of household appliances. Samples of different but similar series of household appliances can also be used.
• the central point can be implemented in particular in the form of a server or, for example, as a service in a cloud.
• the device data can in particular be transmitted wirelessly from the household appliance to the central location.
• the central office can carry out the technology described with regard to different household appliances. Parallels between the different household appliances can then be better determined. This determination can be made manually or automatically. The transferability of results from one household appliance to another can be checked accordingly.
• the central location can access a disproportionately enlarged base of device data. This can enable an improved determination of relevant patterns.
• FIG. 1 shows a system with a household appliance and a central location
• FIG. 2 shows a flow chart of a method.
• FIG. 1 shows a system 100 with a household appliance 105 and a central location 110.
• the household appliance 105 is shown as a coffee machine by way of example, but it can also include any other appliance that is preferably set up for use in a household.
• the household appliance 105 can be used in a household kitchen, can be used for laundry care or household cleaning.
• the household appliance 105 includes a control device 115 which can be set up to control a function of the household appliance 105 .
• a dedicated control device can be provided for device control.
• the control device 115 includes a processing device 120 which is connected to one or more sensors 125 .
• a first sensor 125 which is set up to determine a pressure of water when brewing coffee, and a second sensor 125, which is set up to detect an operating noise of the household appliance 105, are shown as examples.
• any number of further sensors 125 can be provided.
• an input device for use by a person can also be regarded as sensor 125 .
• activation of an actuator of the household appliance 105 for example a heater or a pump, can be detected, which can also be viewed as a sensor 125.
• an actuator of the household appliance 105 for example a heater or a pump
• a sensor 125 can also be viewed as a sensor 125.
• Device data of the household appliance 105 recorded by means of the sensors 125 can be processed by the processing device 120 and optionally also stored.
• the processing can include a comparison with predetermined patterns, which are described in more detail below with reference to FIG.
• a pattern can be created or improved on the basis of acquired device data. Both the recognition and the adaptation of a pattern can take place locally on the part of the control device 115 or remotely on the part of the central location 110 . It is preferred that the pattern recognition is done locally and device data that is compared to the pattern is sent to the central location 110 where a pattern can be improved based on the device data. A changed or new pattern can be sent back to the household appliance 105 .
• the household appliance 105 For communication with the central location 110, the household appliance 105 preferably includes a communication device 130, which is shown as a wireless interface by way of example.
• the central point 110 includes a corresponding communication communication device 135.
• the communication devices 130, 135 can also be set up for wired communication.
• the communication usually takes place via a predetermined network, for example a mobile radio network or the Internet.
• the central point 110 further comprises a processing device 140 and an optional storage device 145.
• An interface 150 can also be provided.
• the processing device 140 is preferably set up to execute an artificial neural network 155 .
• Device data of the household appliance 105 which was received via the communication device 135, can be checked by the processing device 140 for similarity to one or more predetermined patterns.
• a pattern can be trained on the basis of received device data.
• this may include training the network 155 on the received device data.
• Received device data can be stored in the storage device 145 .
• Training an artificial neural network 155 may require a large variety of device data, and the central office 110 is preferably set up to receive and store device data from a large number of household appliances 105 .
• Received device data may also be accumulated over time to create a pattern or improve recognition performance related to a pattern.
• information that is associated with a pattern can be stored in the storage device 145 . This information can include, for example, a designation, the frequency of an occurrence, a note on rectifying or improving the device status, or service information. In particular, this information can be created by a person skilled in the art on the basis of an analysis of device data that is assigned to a pattern.
• An indication of a pattern matching received device data may be provided via the interface 150 .
• the notice may include the information mentioned in part or in full.
• functions of the processing device 140 of the central location 110 can also be performed by the processing device 120 of the control device 115 of the household appliance 105 .
• the neural network 155 it is preferable for the neural network 155 to be local in the processing Device 120 of the household appliance 105 takes place, and a creation or further development of the neural network 155 by the central office 110.
• FIG. 2 shows a flowchart of a method 200 that can be executed on system 100 in particular.
• the method 200 can be executed entirely or partially on the control device 115 of the household appliance 105 or on the central location 110 .
• a mixed version is also possible.
• appliance data of the household appliance 105 can be scanned.
• the device data can come from one or more of the sensors 125 in particular.
• an abnormal device condition can be determined based on the device data. In particular, it can be determined that the device data does not match a pattern that indicates a normal device state.
• the collected device data may be compared to one or more templates, each relating to an abnormal device condition.
• the patterns can, for example, have been created on the basis of laboratory tests or manually labeled data sets using methods of artificial learning.
• a step 220 it can be determined whether a pattern could be found to which the detected device data are sufficiently similar. In particular, it can be checked whether an unambiguous assignment of the device data to one of the patterns is possible. If this is not the case, then in a step 225 it can be determined that the device data are sufficiently different from all known patterns. In addition, a new template can be created based on the acquired device data. In a step 230, the scanned device data can be assigned to the new pattern or to a previously found sufficiently similar pattern. Optionally, in a step 235, a description of an associated device state can be provided. This description can be provided in particular to an operator or a service person for the household appliance 105 .
• the pattern to which the device data could be assigned can be expanded to include the device data. In the case of the newly generated pattern, this may already have taken place. In the case of an already existing pattern, the device data can first be saved for a later update of the pattern. In a step 245 it can be determined whether there is enough new device data for a pattern present. It can also be checked whether device data that is to be assigned to a template has reached a higher than a predetermined age. Other criteria are also possible. In a step 250, the pattern can be updated as a function of the criteria determined. In particular, it can be checked whether there is device data that is associated with the pattern but should actually be better associated with another pattern.
• the pattern or its recognition can be trained in step 250 .
• the recognition of all existing patterns is trained together.
• This may also include a pattern associated with a device condition that indicates normal, non-abnormal operation of the home appliance 105 .
• step 250 it can be checked whether descriptive information associated with a pattern needs to be updated. For this purpose, a signal can be provided to an operator, who can then carry out the check.

Landscapes

• Physics & Mathematics (AREA)
• Engineering & Computer Science (AREA)
• Artificial Intelligence (AREA)
• Evolutionary Computation (AREA)
• Mathematical Physics (AREA)
• General Physics & Mathematics (AREA)
• Automation & Control Theory (AREA)
• Testing And Monitoring For Control Systems (AREA)
• Selective Calling Equipment (AREA)
• Management, Administration, Business Operations System, And Electronic Commerce (AREA)

Priority Applications (2)

Applications Claiming Priority (2)

Publications (1)

Family

ID=82898969

Family Applications (1)

Country Status (4)

Citations (3)

Family Cites Families (4)

• 2021
  • 2021-08-04 DE DE102021208443.5A patent/DE102021208443A1/de active Pending
• 2022
  • 2022-07-19 EP EP22754344.4A patent/EP4381360A1/de active Pending
  • 2022-07-19 WO PCT/EP2022/070124 patent/WO2023011909A1/de active Application Filing
  • 2022-07-19 CN CN202280053252.3A patent/CN117751334A/zh active Pending

Patent Citations (3)

Also Published As

Similar Documents

Legal Events