BE1029768B1 - Method for monitoring a load status of a flushing device, device and flushing device - Google Patents

Abstract

The invention relates to a method (100) for monitoring a load status of an interior (210) of a flushing device (200). A microphone signal (250) and a pump signal (230) are read in, with the microphone signal (250) representing a noise pattern and the pump signal (230) representing a pumping state of a pump (220) of the flushing device (200). At least one of the signals is processed using an artificial neural network (232) in order to classify a current load status of the flushing device (200) as correct or incorrect, and if the load status is incorrectly classified, a load signal (410) representing the incorrect load status is provided. A control signal (270) is provided to a device interface (275) and additionally or alternatively an information signal (280) is provided to a user interface (285) to a user in response to the loading signal (410), the control signal (270) indicating a change in a device setting for adapting the scavenging process to the current load status, and wherein the information signal (280) represents information about the current load status.

Description

Description

Method for monitoring a loading condition of a flushing device, device and

flushing device

The invention relates to a method for monitoring a load status of a s rinsing device, a device and a rinsing device.

Artificial intelligence (AI) devices can be integrated into household appliances to increase the usefulness of the appliances and the ease of use. Furthermore, various

Sensors, for example in camera systems, can be installed in the dishwasher, for example to provide information about the current load of items to be washed and its

get pollution.

The object of the invention is an improved method for monitoring a

Load status of a flushing device, an improved device and an improved

to create a dishwasher.

According to the invention, this object is achieved by a method for monitoring a loading condition of a flushing device, a device and a flushing device with the features of the main claims. Advantageous refinements and developments of the invention result from the following dependent claims.

The advantages that can be achieved with the invention consist not only in an optimization of the loading of the washing baskets but also in an increase in the operational reliability of the washing device. For example, the approach described is suitable for detecting items to be washed that have fallen over or for

Recognition of an accumulation of incorrectly loaded wash ware containing washing liquor.

A method for monitoring a loading condition of an interior of a vehicle is disclosed

Flushing device presented, the method having the following steps:

Reading in at least one microphone signal and at least one pump signal, wherein the

Microphone signal represents a noise pattern in the interior captured by a microphone unit and the pump signal represents a pumping state of a pump of the flushing device captured by a pump sensor,

Processing the microphone signal and additionally or alternatively the pump signal by means of an artificial neural network in order to classify a current load status of the flushing device as correct or incorrect, and in the case of an incorrectly classified one

Load condition representing the faulty load condition

provide loading signal, and

Outputting a control signal to a device interface and additionally or alternatively an information signal to a user interface to a user in response to the

Loading signal, wherein the control signal changes a device setting for

Adaptation of the flushing process to the current loading condition represents, and wherein the information signal represents information about the current loading condition.

The method described above for monitoring a loading condition of a

Rinsing device is in particular part of a method for cleaning items to be washed, in which items to be washed are rinsed in the rinsing device.

The washing device can be a dishwasher, for example, in the interior of which two or more washing baskets for arranging dishes can be arranged. In addition to standard recommendations for loading dishwashing baskets, which can also be referred to as crockery baskets, in practice there can often be a rather heterogeneous load pattern. On the one hand, the range of bowls, cups, pots and pans is very wide, which can result in a very household-specific range. ı5 On the other hand, the crockery baskets can usually be loaded more intuitively, especially when it comes to larger pieces of crockery such as pots and bowls. This can result in the problem that larger items of crockery can be placed in the wrong basket (larger bowls should ideally be in the lower basket), in a non-optimal position in the basket and even the wrong way around in the crockery basket. In addition, consists of light bowls, bowls and cups

Plastic poses the risk that they can be knocked over by the water pressure of the spray jets during rinsing and the washing liquor can then collect in a cavity, which can no longer be available to the water cycle. This can cause unstable

pump operation and also lead to a reduced cleaning and drying performance.

The pump can also be damaged if it delivers too little water.

It can also happen that dishes are placed in the baskets that are not in the

Dishwashers belong, such as spatulas made of wood or crockery

Aluminum. Furthermore, it can happen that crockery, for example coffee cups, tip over when the basket is pushed in if they were previously placed in the basket in a statically unstable manner. With the method presented here, an operator of the flushing device can advantageously have a monitored load management system, supported in particular by today's possibilities of digitization, in order to

to optimize the loading of the baskets and to increase the operational reliability of the device.

For this purpose, the flushing device can, for example, comprise a plurality of different sensors, or a sensor array, which can supply an artificial intelligence or a neural network with information. The neural

Network in turn can be connected to a device control, for example, a machine actuators and an output system, that is, an interface to a

user, can control. The output system can be, for example, an acoustic signal generator in the dishwasher, the display of the device or a smartphone with a corresponding app. With the method presented here, the

Combination of multiple sensor signals that support the neural network that

Load situation of the dishwasher recorded and based on this can be given to the user

Corrections and warnings are given.

For this purpose, the AI or the neural network can be trained on how

Items of crockery are not standing in the basket or how they are not allowed to be put away. by means of

The neural network provides information about the number of wrongly admitted objects

The operator has the option of correcting the position of the cups via the dispensing system or the program with the announcement of a later automatic

rinsing parameter adjustment, for example an additional water change or a ı5 increase in the amount of water. If an object changes position during a wash cycle, such as a bowl that has fallen over, a

Information for the user can be provided and additionally or alternatively

Rinsing parameters can be adjusted to the changed load status.

According to one embodiment, in the reading step or in a reading step that is carried out again, at least one camera signal can be read that represents an image of the interior captured by a camera unit. In doing so,

Step of processing the camera signal are processed to provide the loading signal. For example, a camera can be both inside the dishwasher and outside the appliance, for example it can be the camera in a smart

Phone or a camera installed in the kitchen, for example under the ceiling.

With the camera unit or optionally with multiple cameras can advantageously

Load situation of the dishwasher baskets are recorded. With classic

Image processing can, for example, capture the opening side and additionally or alternatively the support side of critical crockery items, such as cups and bowls, by at least one camera. In addition, the shape and orientation of the handle can be used as a criterion for cups. The cup handles can, for example, be opened wider on the cup opening side than on the bottom side, which, in addition to the other criteria, can provide reliable information about the orientation of this piece of crockery in the basket.

Furthermore, some cups can have a manufacturer's stamp on the bottom, which can also be recorded by the cameras and used as a position criterion. If the stamp can be seen, then the cup was put in the right way around.

In the new step of reading in, the camera signal can be read in if the current load status using the microphone signal and/or the

Pump signal is classified as faulty. Thus, items to be washed that have been incorrectly loaded can be initially recognized, for example, using a pump pressure sensor and a microphone s. Only in the second place can an optional further evaluation be carried out by image evaluation. In this case, the image evaluation can be triggered by the fact that using the pump pressure sensor and the microphone, items to be washed that have fallen over, for example, or an accumulation of washing liquor in items to be washed that have been placed incorrectly, have been detected beforehand.

According to a further embodiment, in the step of reading in at least one

Weight signal are read, which represents a detected by a load sensor weight of a washing basket arranged in the interior. In the step of

Processing the weight signal are processed to provide the load signal. For example, the loading sensor or several loading sensors can be designed as strain gauges or acceleration sensors. by means of

Loading sensors, the neural network can advantageously receive additional information about the weight of the baskets and thus also about the weight of the basket load. The

The weight of the empty baskets and that of a standard and also an intensive load can be

Known in advance or easily ascertainable. If the measured weight deviates from

From experience, for example, an incomplete load, a load with mostly plastic crockery or a stocking with pots or

pans available. In conjunction with the image information from the camera unit, the neural network can check the loading situation for plausibility and, if necessary,

Provide correction recommendations to the user. For example, recognize the

If the load sensors in the upper basket are too heavy, the neural network detects pots or pans in the upper basket using the learned camera images

Loading status are classified as faulty and the display unit can

Indicate to put the heavy crockery in the lower basket. So can under

Use of the weight signal advantageously as real and comprehensive as possible

image of the loading condition can be created.

According to a further embodiment, the method can include a step of modifying the neural network using a data read in via the user interface

Have confirmation signal, wherein the confirmation signal requests a reclassification of the current load status. In practice, for example, it can happen that, for example, rather unusual items of crockery, in particular atypical items of crockery newly acquired by the user, or items to be washed outside of the crockery context in the

dishwashers that the neural network does not yet know. Such crockery can be inserted incorrectly by the neural network or identified as unknown and the

Load state are classified according to faulty. Using the

Information signal related information can be provided. The user can react to this by actively confirming that the load status can actually be classified as correct. This can be done with a smartphone program, for example

carried out at the end of the wash cycle. Advantageously, using the acknowledgment signal, the learning of the neural network can be triggered proactively and the identified object can be added to a database, for example, for future reference

to avoid false alarms.

According to a further embodiment, in the step of processing the

Pump signal are compared with the camera signal when the pump signal represents an irregular pump pressure to incorrectly arranged objects in the

capturing the interior. The pump sensor can be used, for example, as a speed sensor,

Pump pressure sensor or armature current sensor can be designed and can be used during the

Monitor the stable pump operation during flushing operation. In addition or as an alternative to this, the switching behavior of the heating pressure monitor can also be evaluated. If the user after loading the baskets the program with the need for a later

Program customization starts, that is, when, for example, notice of wrongly granted

Dishes that are ignored can cause wash liquor to collect in the volumes of the cups or bowls, which can lead to instability in the pump pressure. The pump pressure, the

In such a case, the pump speed or the armature current of the pump motor can fluctuate. If all items of crockery are placed correctly, a relatively stable

pump pressure, a relatively constant speed and a relatively constant armature current result. If the rinsing water collects in the cavities of the vessels, the pump can deliver irregularly, which can lead to fluctuations in the signal. When such a fluctuating pump signal is read in, the neural network can, for example, compare the pump signal with the camera signal in order to compare the pump behavior with the images from the

Compare the loading situation derived from the camera. For example, it can be detected whether the unstable pump pressure was triggered by an object placed the wrong way round, in which water is now collecting, which is consequently missing from the water circuit of the pump. If such a condition or a similar condition is detected by comparing the pump signal with the camera signal, then advantageously the to be refilled

Amount of water, which would again lead to a stable pump behavior, determined by the neural network as a function of the detected number of incorrectly placed dishes and transferred to the device control, for example using the control signal. Advantageously, the amount of water can then be adjusted.

According to a further embodiment, the pump can be stopped in the output step using the control signal. For example, a

The pump can be stopped or its speed reduced when the read-in

Pump signal shows inconsistencies or fluctuations. Then, for example, the pump can be stopped or at least its speed can be reduced until the

The cause of the problem was found, for example, by comparing it with other sensor signals and corrected using the control signal or using the

Information signal the user has been informed, for example to change a setting.

Advantageously, a malfunction of the pump can be detected early and a

Damage to the flushing device can be avoided.

According to a further embodiment, in the step of outputting stopping the

Pump can be triggered when the microphone signal has an acoustic deviation from that

noise image represented. During the flushing process, the neural network can

Example continuously access the microphone unit. The microphone can, for example, be arranged waterproof or at least water-protected in the interior of the flushing device and can be configured additionally or alternatively. A characteristic noise pattern can arise when the machine is being washed. The smooth movement of

For example, spray arms can cause periodically recurring signals. be in

If dishes are tipped over or thrown over during washing operation, for example by the pressure of the spray jets, this can be seen in an acoustic deviation from the characteristic uniform noise profile, for example in a deviation in the form of a so-called signal peak. In such a case, to obtain a reference,

Pump stopped to allow recording of camera information through the neural network, triggered by the microphone signal. This has the advantage that the spray jets do not interfere with the image recording. The neural network can do that

Compare the camera recording, for example, with stored reference profiles and images of the loaded basket before the start of the washing program and can identify the error under

signal use of the information signal. Since it can be expected that washing up liquid can collect in an overturned bowl or a spilled cup, a water change can also be arranged to bring the washing process back to life

supply nominal water quantity. Furthermore, this can be located in the vessel

liquid to be diluted. Furthermore, the neural network in addition to the

Compare image information, for example the microphone signal with the pump signal. If the signal fluctuation shows up in both the microphone signal and the pump signal, the likely item that was thrown over is a bowl. Around Here

To gain plausibility, the neural network can reference the camera signal with the signals from the other sensors. Since the volume of a bowl increases slowly with

If the washing liquid fills up, the peak in the microphone and pump signal would not occur at the same time,

but at different times. The hydraulic signal image can therefore be made later in time, since the volume of the vessel only fills up. Over the duration of the delay, the neural network can advantageously determine the amount of water to be refilled in conjunction with the images of the loading situation. Additionally or alternatively, the

Users are informed about the incorrect loading status. If the user does not respond to the prompt within a specified time, the program adjustment can be carried out as described.

According to another embodiment, the method may include a step of temporarily

turn on the pump to detect a sustained acoustic deviation from the

to capture noise. For example, an acoustic deviation from the

Set the noise level when one or more spray arms are blocked or become blocked or one or more spray arms that can still turn by the

touch dishes that have slipped or slide along them. In order for the neural network to perform an image comparison or an image comparison using the camera unit in response to the microphone signal ı5 Unaffected by the spray jets

Can make a decision, for example, the circulation pump can be switched off.

This means that the rinsing operation can be interrupted. The circulation pump can then be released again for at least 5 seconds, for example, and then switched off again. If the position of one or more spray arms has not changed, there is a spray arm blockage and information about the faulty one

Operating status can be communicated to the user using the information signal. Advantageously, damage to the flushing device or to objects arranged in the interior can thereby be avoided.

According to a further embodiment, the processing step can be carried out again in response to the outputting step in order to classify the current load status of the flushing device again as correct or faulty. For example, as a result of the step of outputting, the load status can have been changed by user intervention or by changing a device setting. Accordingly, the processing step can be performed again to reflect the changed circumstances

TO examine and reclassify. Additionally or alternatively, a load state classified as faulty by the neural network can also have been confirmed by the user as correct and the neural network can have modified itself accordingly in order to

Load status must therefore be classified as correct. Advantageously, the current

Load status can be continuously and optimally monitored.

According to a further embodiment, the method can include a step of providing the camera signal using the camera unit and the microphone signal

Using the microphone unit and the pump signal using the

Include pump sensor. In this case, for example, the neural network can be connected to the camera unit, the microphone unit and the pump sensor in a manner capable of transmitting signals, as well as additionally or alternatively to other sensors and, for example, can be designed to control them. Advantageously, this allows a

Signal transmission is facilitated and the monitoring of the loading status of the flushing device can be optimized.

The approach presented here also creates a device that is designed to

Carry out, control or implement steps of a variant of a method presented here in appropriate facilities. Also through this

Embodiment of the invention in the form of a device, the object on which the invention is based can be solved quickly and efficiently.

The device can be designed to read input signals and

Using the input signals to determine and provide output signals. A

The input signal can represent, for example, a sensor signal that can be read in via an input interface of the device. An output signal can be a control signal or a

Represent data signal that can be provided at an output interface of the device. The device can be designed to take the output signals

to determine the use of a processing rule implemented in hardware or software. For example, the device to a logic circuit, an integrated

Circuit or include a software module and for example as a discrete

Be realized component or be comprised of a discrete component.

In addition, a flushing device with a variant of the device presented above is presented.

This combination has the advantage that all the advantages mentioned above can be optimally implemented.

A computer program product or computer program is also advantageous

Program code on a machine-readable carrier or storage medium such as a

Semiconductor memory, a hard disk memory or an optical memory can be stored. If the program product or program runs on a computer or

Executed device, the program product or program for implementation,

Implementation and / or control of the steps of the method are used according to one of the embodiments described above.

An embodiment of the invention is shown purely schematically in the drawings and is described in more detail below. It shows

Figure 1 is a flow chart of a method for monitoring a

Load status of an interior of a flushing device according to a

embodiment;

FIG. 2 shows a schematic representation of an exemplary embodiment of a flushing device with a device;

Figure 3 is a flowchart of a method for monitoring a

Load status of an interior of a flushing device according to a

embodiment;

Figure 4 is a schematic representation of a device according to a

embodiment;

FIG. 5 shows a schematic representation of an exemplary embodiment of a flushing device with a device;

Figure 6 is a flow chart of a method for monitoring a

Load status of an interior of a flushing device according to a

embodiment;

Figure 7 is a schematic representation of a pump signal according to a

embodiment; and

Figure 8 is a schematic representation of a microphone signal according to a

example.

Figure 1 shows a flowchart of a method 100 for monitoring a

Loading status of an interior of a flushing device according to an embodiment.

The method 100 includes a step 105 of reading in a microphone signal and additionally or alternatively at least one pump signal. It represents that

Microphone signal a noise pattern in the interior captured by a microphone unit and the

Pump signal represents a pumping condition detected by a pump sensor

Flushing pump. Optionally, in this step or in a subsequent step 105 of reading in, a camera signal can also be read in, which can represent an image of the interior captured by a camera unit.

In a subsequent step 110 of the processing, in this exemplary embodiment,

Microphone signal and the pump signal, as well as optionally the camera signal, are processed by means of an artificial neural network in order to obtain a current loading status of the

to classify the flushing device as correct or faulty. In the case of an incorrectly classified

Load status becomes one representing the faulty load status

Load signal provided. For example, such a loading signal indicates that items to be washed that have fallen over or been placed incorrectly have been detected, for example to a

accumulation of washing-up liquid.

According to one exemplary embodiment, the camera signal is read in, triggered by an evaluation of the microphone signal and/or the pump signal, and in step 110 for

Providing the loading signal and/or for reclassifying the

loading condition used.

The method 100 further includes a step 115 of outputting a control signal to a device interface in response to the loading signal. In this

In this step 115, an additional information signal is sent to an exemplary embodiment

User interface issued to a user. The control signal represents a

Changing a device setting to adapt the rinsing process to the current one

Loading status and the information signal represents information about the current loading status.

Figure 2 shows a schematic representation of an embodiment of a washing machine 200 with a device 205 for monitoring a load status of an interior 210 of the washing machine 200. In this embodiment, the washing machine 200 is designed as a ı5 dishwasher and includes, by way of example, three washing baskets 211, 212, 213 to accommodate of dishes, the washing baskets 211, 212, 213 in this

Embodiment as a lower basket 211 in the lower area of the interior 210, a

Upper basket 212 in the central area of the interior 210 and as a cutlery drawer 213 upper area of the interior 210 are formed. Each wash basket 211, 212, 213 is

Can be supplied with washing liquor by means of a spray arm 215, 216, 217 in order to wash dishes that can be arranged in the washing baskets 211, 212, 213. For this purpose, the spray arms 215, 216, 217 are connected to a pump 220 for pumping water. The pump 220 is in this

Embodiment designed as a circulation pump and connected to a water switch 222 and a pump sensor 225. The pump sensor 225 is in this

Embodiment designed as a pump pressure sensor and monitored during a

rinsing operation ensures stable pump operation. In another embodiment, the

The pump sensor can also be designed as a speed sensor or armature current sensor, for example, and this can also be used in addition or as an alternative to monitoring the pump operation

Switching behavior of a heating pressure monitor can be evaluated. In this

Embodiment, the pump sensor 225 is also designed to

To detect the pump status of the pump 220 and to provide the pump status using a pump signal 230 to the device 205 .

In this exemplary embodiment, the device 205 comprises an artificial neural network 232 and is designed to implement a method for monitoring the loading state, as is the case

Described in the previous figure to execute. For this is the neural

Net 232 trained in this embodiment how crockery should not stand in the basket or not be allowed to be put away. Furthermore, the neural

Network 232 trained only as an example, reflecting or shiny metallic

Dishes such as stainless steel pots can be seen. A distinguishing feature is only an example on the one hand the shiny surface and on the other hand the pot handles or the characteristic handle of a casserole. Furthermore, the size of these dishes is a distinguishing feature or the shape in general. Furthermore, the neural learns

Network 232 to recognize crockery made of wood in this embodiment, such as

Example spatula and bread boards. In principle, crockery made of wood may

Dishwashers can be cleaned, but rinsing with detergent leads to a

leaching of the surface. Both spatulas and lunch boards have a characteristic shape, which is identified by neural network 232 in this embodiment as a

identifier is used. The learning process can be carried out as supervised learning. In supervised learning, for example, the neural network 232 is

Samples or image samples of possible basket loads or also ı5 individual crockery items created in the laboratory, in particular including additional annotation information on the

Patterns handed over, from which the loading condition of a washing basket 211, 212, 213, i.e. the shape or shape of the dishes and also orientation in the basket, in the

Within the scope of accuracy can be determined or predicted. In this

Embodiment, this is done only by way of example on the size, the color, the characteristic shape and the surface finish or the

surface gloss. The neural network 232 is designed to derive from the combination of all

Characteristics to develop a prognosis about the present piece of crockery. Furthermore, the position of the crockery is only an example of learning the opening that

Orientation and the shape of cup handles or also recognizable by the printing on the bottom. For this purpose, the neural network 232 is designed to the pump signal 230, and provided by a camera unit 235 and a camera signal 240 from a

Microphone unit 245 to read provided microphone signal 250.

The camera unit 235, which can also be referred to as a camera, is in this

Embodiment arranged on a device door 260 and designed to visualize the interior 210 and using the camera signal 240 to the device 205 to provide. In another embodiment, the camera unit can also be arranged outside of the device, for example on a smartphone or under a

kitchen ceiling. With the camera unit 235 is a loading situation or a

Loading status of the washing baskets 211, 212, 213, which can also be referred to as dishwasher baskets, can be recorded. With classic image processing, the opening side or the contact side of at least one camera are more critical

Dishes, such as cups and bowls, recognizable. Optionally, the handle shape and orientation can be used as a criterion for cups. This is possible

if, for example, the cup handles are more open on the cup opening side than on the bottom side, which, in addition to the other criteria, provides reliable information about the

Orientation of this piece of crockery can be in the basket. Furthermore, many cups have a manufacturer's stamp on the bottom, which can also be recognized by the camera and used as a location criterion. If the stamp can be seen, then the cup has been put in the right way round.

In this exemplary embodiment, the microphone unit 245 is arranged in the interior space 210 and is designed to be watertight, merely by way of example. In another exemplary embodiment, the microphone unit, which can also be referred to as a microphone, can be designed to be at least water-proof and, additionally or alternatively, can be water-proof or water-proof. A characteristic noise pattern is produced when the machine is being washed. The uniform movement of the spray arms 215, 216, 217 only causes periodically recurring signals as an example. In the event of a tipping over or

If a piece of crockery is knocked over during the washing operation, the resulting noise ı5 can be detected by the microphone unit 245 as an acoustic deviation from the noise pattern and can be made available to the device 205 or to the neural network 232 using the microphone signal 250 .

The device 205 is designed to read in the pump signal 225, the camera signal 240 and the microphone signal 250 and to process them by means of the artificial neural network 232 in order to classify a current load status of the flushing device 200 as correct or faulty. In the case of an incorrectly classified loading condition, the

Device 205 configured in this exemplary embodiment, on the one hand, to output a control signal 270 to a device interface 275 in order to bring about a change in a device setting and thus adapt a rinsing process to the current load status.

In addition to the control signal 270, the device 205 is designed in this exemplary embodiment to output an information signal 280 to a user interface 285 to a user, with the information signal 280 representing information about the current load status. The is merely an example

User interface 285 formed as a display on the appliance door 260 of the dishwasher 200.

In another exemplary embodiment, the user interface can also be arranged in a smartphone or tablet with a corresponding app, for example. by means of a

The neural network 232 provides the user with information about the number of incorrectly allocated cups, by way of information via the output system, which is merely exemplary

to correct the position of the cup or to start the program with the announcement of a later automatic adjustment of the washing parameters.

Figure 3 shows a flow chart of a method 100 for monitoring a

Loading status of an interior of a flushing device according to an embodiment.

The method 100 presented here corresponds or is similar to that in the previous one

Figure 1 described method. In step 105 of reading in this embodiment s, in addition to the camera signal, the microphone signal and the

Pump signal read in at least one weight signal. The weight signal represents, merely by way of example, a weight of a washing basket arranged in the interior of the washing device, which weight is detected by a load sensor. In doing so, in step 110 of the processing in this embodiment, the weight signal is processed to provide the load signal. Because the weight of the empty baskets and that of a norm and also that of a

Intensive loading known in this embodiment or in the neuronal

net are stored, the weight of the basket load is also determined via the weight signal. If the measured weight deviates from empirical values, this means, for example, that the load is incomplete, that the load consists mainly of plastic crockery or that pots or pans have been fitted. In connection with the image information from the camera unit, the neural network in this exemplary embodiment checks the plausibility

Loading situation and are in step 115 of output using the

information signal correction recommendations to the user.

In addition, the method 100 in this exemplary embodiment includes an additional step 300 of modification. In this step 300, the neural network is modified using a confirmation signal read in via the user interface, the

Confirmation signal requests a reclassification of the current load status. In other words, the neural network learns from the user's request. For example, in practice it can happen that, for example, rather unusual

Crockery items, in particular atypical crockery items newly purchased by the user or

Items to be washed are placed in the dishwasher outside of the crockery context, which the neural network is not yet aware of. In such a situation, the user is notified via the

User interface, which is designed as part of a smartphone program only by way of example, provides the confirmation signal and thus confirms that the items of crockery detected as faulty are to be detected as the correct load. Under

Using the confirmation signal, the neural network is modified accordingly.

Figure 4 shows a schematic representation of a device 205 according to a

example. The device 205 shown here corresponds to or is similar to the devices described in the previous figure 2, with the difference that the

Device 205 in this embodiment both the neural network 232 and a

Controller 400 for controlling device settings includes as separate units.

In this case, the neural network 232 is designed to read in sensor signals which, for example, are a camera signal 240, a microphone signal 250, a

Pump signal 230 and a weight signal 405 is. It represents that

Camera signal 240 captured by a camera unit image of an interior of a

Flushing device, as described in the previous figure 2. Similarly s, the microphone signal 250 represents a noise image im captured by a microphone unit

Interior, the pump signal represents one detected by a pump sensor

Pumping state of a pump of the rinsing device and the weight signal represented in this

Exemplary embodiment, a weight of a washing basket arranged in the interior, in particular loaded with items to be washed, is detected by a load sensor. Neural network 232 is designed to process signals 230, 240, 250, 405 that are read in, in order to classify a current load status of the flushing device as correct or faulty, and to provide a load signal 410 representing the faulty load status if the load status is incorrectly classified. In this exemplary embodiment, the loading signal 410 is provided to the control device 400 of the device 205 . The ı5 control device 400 is only designed as an example to respond to the

Loading signal 410 output a control signal 270 to the device interface 275, which can also be referred to as machine actuators. In addition, in this

In the exemplary embodiment, the information signal 280 can likewise be output to the user interface 285 by the control device 400 .

Figure 5 shows a schematic representation of an exemplary embodiment of a rinsing device 200 with a device 205 for monitoring a load status of an interior 210 of the rinsing device 200. The rinsing device 200 shown here corresponds or is similar to the rinsing device described in the previous Figure 2 and the device 205 shown here corresponds to or is similar to that described in Figures 2 and 4 above

Contraption. In contrast to that described in the preceding FIG

Flushing device includes the flushing device 200 in this embodiment, an additional

Camera unit 500, which is on the opposite side of the camera unit 235 of the

Interior 210 is arranged. The additional camera unit 500 is designed only by way of example in order to capture the interior space 210 like the camera unit 235 and to provide an image of the interior space to the device 205 .

In this exemplary embodiment, the washing device 200 also has a load sensor 505, 506, 507 on each washing rack 211, 212, 213. The load sensors 505, 506, 507 are only designed by way of example to detect a weight of the respective washing rack 211, 212, 213 and to provide it to the device 205 using a weight signal 405. The loading sensors 505, 506, 507 are merely exemplary for this

Strain gauges formed. In another embodiment, a

Loading sensor can also be designed as an acceleration sensor, for example. The

In this exemplary embodiment, device 205 is designed to read in and process the weight signal 405 in order to provide the loading signal described in the previous FIGS. In doing so, both the weight of the empty

Flushing baskets 211, 212, 213 as well as the weight of a standard and an intensive load. As a result, the device is designed to take the weight of a current

To determine basket loading 510. If the measured weight deviates from empirical values, this means that the load is incomplete, that is, a load with mostly crockery

plastic or also equipped with pots or pans. In connection with the

The loading situation can be checked using image information from the cameras. If incorrect loading is detected, a corresponding correction instruction can be output via the user interface 285 .

If the current load status is classified as faulty, although a user classifies it as correct, then the neural network 232 of the device 205 is additionally designed in this exemplary embodiment to be modified. For this purpose, a confirmation signal 515 can be read in via the user interface 285, with the confirmation signal 515 having a

Requests reclassification of the current load status. Using the

Confirmation signal 515 can be used to learn the current load status as correct and the neural network 232 can be modified accordingly.

Figure 6 shows a flow chart of a method 100 for monitoring a

Loading status of an interior of a flushing device according to an embodiment.

The method 100 presented here corresponds or is similar to those in the previous one

Figures 1 and the method described, with the difference that the one shown here

Method 100 has additional steps. In this exemplary embodiment, a step 600 of providing the camera signal takes place before step 105 of reading in

Using the camera unit and microphone signal using the

Microphone unit and the pump signal using the pump sensor. Only as a result of step 600 of providing are the signals read in and processed in this exemplary embodiment. In this case, only by way of example, in step 115 of outputting, the pump is caused to stop if the microphone signal represents an acoustic deviation from the noise pattern.

In this exemplary embodiment, the neural network accesses the microphone of the rinsing device during the rinsing process. A characteristic noise pattern is produced when the machine is being washed. The even movement of the spray arms causes periodically recurring signals. Are in flushing by the pressure of

If items of crockery are tipped over or knocked over by spray jets, this is reflected in an acoustic deviation in the form of a so-called signal peak from the characteristic, uniform noise profile. To get a reference go to

Recording of the camera information triggered by the microphone signal, the pump stopped. This has the particular advantage that the spray jets do not interfere with the image recording. The neural network compares them with stored reference profiles and images of the loaded basket before the start of the washing program and signals the error via the user interface. Since it is to be expected that washing up liquid collects in an overturned bowl or cup, in this exemplary embodiment a

Water change initiated in order to supply the nominal amount of water to the rinsing process again.

Furthermore, the liquid in the vessel is thereby diluted.

Accordingly, in this exemplary embodiment, step 110 of processing is carried out again in response to step 115 of outputting in order to compare the camera signal with the microphone signal and to reclassify the current load status of the flushing device as correct or faulty. In other words, in this exemplary embodiment, the neural network compares the signal ı5 of the microphone with that of the pump sensor in addition to the image information. If the signal fluctuation both in

microphone signal as well as in the pump signal, it will be the overturned

part of the crockery is probably a bowl. In order to achieve plausibility here, the neural network references the camera images with the signals from the other sensors.

Since the volume of a bowl only slowly fills with washing liquid, the peak in

Microphone and pump signals do not occur at the same time, but are offset in time. The hydraulic signal image follows later in time, since the volume of the vessel first has to fill up. The neural network in this determines the duration of the delay

Embodiment in connection with the images of the loading situation to be refilled

amount of water. Again, the user is using the information signal

Informed. If the user does not respond to the prompt within a specified time, the program adaptation is carried out as described.

A similar acoustic picture occurs when one or more spray arms are or become blocked or one or more spray arms, which can still rotate, touch crockery that has slipped through the basket or slide along them. So that

When the camera or the neural network can carry out an image comparison and make a decision based on the microphone signal and is unaffected by the spray jets, the circulation pump is switched off, i.e. the flushing operation is interrupted. In the following step 605 of temporarily switching on, the pump is released again for at least 5 seconds, only by way of example, in order to then be switched off again. If the position of one or more spray arms has not changed, there is one

Spray arm blockage before and the information about the disrupted operating status is dem

communicated to users.

Figure 7 shows a schematic representation of a pump signal 230 according to a

example. The pump signal 230 shown here corresponds to or is similar to the pump signal described in the preceding FIGS. 2, 4 and 5. While that is

Pump signal 230 is shown as a diagram in the figure shown here, with the s course over time being shown on the x-axis and the course of pump signal 230 being shown on the y-axis. The pump signal 230 can be provided by a pump sensor as was described in the previous FIG. The pump sensor monitors the stable pump operation during the flushing operation. Alternatively, this can also be done

Switching behavior of the heating pressure monitor can be evaluated. If the user after

Loading the washing baskets the program with the need for a later

program adjustment (indication of incorrectly loaded dishes is ignored) starts, washing liquid will collect in the volumes of the cups or bowls, which leads to instability in the pump pressure. The pump pressure, the pump speed or the armature current of the pump motor fluctuates. If all items of crockery are loaded correctly, the result is a relatively stable pump pressure, a relatively constant speed and a relatively constant armature current. This can be mapped in a constant partial signal 700 . If, on the other hand, rinsing water collects in vessel cavities, the pump delivers irregularly, which leads to

Fluctuations 705 in the signal. The amount of water to be refilled, which would again lead to a stable pump behavior, can only be determined by way of example from a neural network, as was described in the preceding FIGS. Then the

Water amount adjustable. In particular, the neural network compares that

Pump behavior with the loading situation derived from the images from the cameras.

Figure 8 shows a schematic representation of a microphone signal 250 according to a

example. The microphone signal 250 shown here corresponds to or is similar to the microphone signal described in the previous FIGS. 2, 4 and 5. While that is

Microphone signal 250 shown as a diagram in the figure shown here, with the temporal

Course is shown on the x-axis and the course of the microphone signal 250 on the y-axis. During the rinsing process, a neural network of a rinsing device, as described in the previous FIGS. 2, 4 and 5, accesses the microphone of the rinsing device. A characteristic noise pattern is produced when the machine is being washed. The uniform movement of the spray arms causes periodically recurring partial signals 800.

If items of crockery are tipped over or knocked over during washing operation due to the pressure of the spray jets, this is reflected in an acoustic deviation 805 in the form of a so-called signal peak from the characteristic, uniform noise profile.

Claims (14)

patent claims 1. Method (100) for monitoring a load status of an interior (210) of a flushing device (200), the method (100) having the following steps: reading in (105) at least one microphone signal (250) and at least one pump signal (230), wherein the microphone signal (250) represents a noise pattern in the interior (210) detected by a microphone unit (245) and the pump signal (230) represents a pumping state of a pump (220) of the flushing device (200) detected by a pump sensor (225); Processing (110) the microphone signal (250) and/or the pump signal (230) by means of an artificial neural network (232) in order to classify a current load status of the flushing device (200) as correct or faulty, and if the load status is classified as faulty, the faulty one to provide a load signal (410) representing the load state; and outputting (115) a control signal (270) to a device interface (275) and/or an information signal (280) to a user interface (285) to a user in response to the loading signal (410), the control signal (270) indicating a change in a Device setting for adapting the flushing process to the current loading status represents, and wherein the information signal (280) represents information about the current loading status. 2. The method (100) according to claim 1, wherein in the step (105) of reading or in a new step (105) of reading in at least one camera signal (240) is read in, which is a camera unit (235) captured image of the interior ( 210), wherein in the processing step (110) the camera signal (240) is processed to provide the loading signal (410). 3. The method (100) according to claim 2, wherein in the renewed step (105) of reading in, the camera signal (240) is read in if the current load status is classified as faulty using the microphone signal (250) and/or the pump signal (230). . 4. The method (100) according to any one of the preceding claims, wherein in step (105) of reading in at least one weight signal (405) is read in, which represents a load sensor (506) detected weight of an interior (210) arranged washing rack, wherein in the step (110) of processing the weight signal (405) is processed to provide the load signal (410). 5. The method (100) according to any one of the preceding claims, with a step (300) of modifying the neural network (232) using a confirmation signal (515) read in via the user interface (285), wherein the confirmation signal (515) involves a reclassification of the current load status requests. 6. The method (100) according to any one of claims 3 to 5, wherein in the step (110) of processing the pump signal (230) is compared with the camera signal (240) if the pump signal (230) represents an irregular pump pressure to erroneously arranged to detect objects in the interior (210). 7. The method (100) according to any one of the preceding claims, wherein in the step (115) of outputting using the control signal (270) a stop of the pump (220) is caused. 8. The method (100) according to claim 7, wherein in step (115) of outputting the stopping of the pump (220) is caused if the microphone signal (250) represents an acoustic deviation (805) from the noise pattern. 9. The method (100) according to claim 7 or 8, including a step (605) of temporarily turning on the pump (220) to detect a sustained acoustic deviation from the noise pattern. 10. The method (100) according to any one of the preceding claims, wherein the step (110) of processing is carried out again in response to the step (115) of outputting in order to reclassify the current load status of the flushing device (200) as correct or incorrect. 11. The method (100) according to any one of claims 2 to 10, with a step (600) of providing the camera signal (240) using the camera unit (235) and the microphone signal (250) using the microphone unit (245) and the pump signal (230) using the pump sensor (225). 12. Device (205) which is designed to carry out the steps (105, 110, 115) of the method (100) according to one of the preceding claims in corresponding units. 13. Rinsing device (200) with a device (205) according to claim 12. 14. Computer program product with program code for carrying out the method (100) according to one of the preceding claims, when the computer program product is executed on a device (205).