Training method of dehydration eccentric prediction model, washing apparatus and control method thereof

Abstract

The present application relates to the technical field of household appliances, and particularly provides a training method of a dehydration eccentricity prediction model, a washing device and a control method thereof. The present application aims to solve the problem that the washing device in the prior art is difficult to predict the clothes distribution state in advance during the dehydration process. To this end, the present application provides a training method of a dehydration eccentricity prediction model, which comprises: obtaining dehydration characteristic data required for predicting the eccentricity of the washing device during the dehydration process; bringing the obtained dehydration characteristic data into an initial model for training; and obtaining the trained dehydration eccentricity prediction model. Therefore, the present application can predict the eccentricity state of the clothes in real time according to the real-time data of the washing device and the dehydration eccentricity prediction model, and output control instructions for the dehydration program of the washing device, thereby predicting the clothes distribution state in advance during the dehydration process, and avoiding problems such as noise, vibration, displacement and collision of the washing drum caused by the eccentricity of the clothes.

Description

TECHNICAL FIELD

[0001] The present application relates to the technical field of household appliances, and particularly provides a training method of a dehydration eccentricity prediction model, a washing device and a control method thereof. BACKGROUND

[0002] In the dehydration process of a conventional washing device, uneven distribution of laundry in the drum often causes large eccentricity during dehydration. If the laundry distribution is not adjusted in time after this situation occurs, it will cause unbalanced dehydration, and further cause serious problems such as noise, vibration, displacement, and drum collision. At the same time, it will also make the dehydration speed unable to rise to high speed, prolong the laundry dehydration time, and cause incomplete dehydration, which seriously interferes with user use.

[0003] In the prior art, the washing device judges the laundry distribution state by weight before dehydration, and it is difficult to predict and judge in advance during the dehydration process. When the determination timing is early, it will cause the dehydration speed to decrease and the dehydration time to be too long. When the determination timing is late, it will cause the washing drum to displace and collide in a critical state.

[0004] Correspondingly, there is a need in the art for a training method of a dehydration eccentricity prediction model, a washing device and a control method thereof to solve the above technical problems. SUMMARY

[0005] The present application aims to solve the above technical problems, i.e., the problem that the washing device in the prior art is difficult to predict the laundry distribution state in advance during the dehydration process.

[0006] In a first aspect, the present application provides a training method of a dehydration eccentricity prediction model, which comprises:

[0007] obtaining dehydration characteristic data required for predicting eccentricity of a washing device during a dehydration process;

[0008] training the obtained dehydration characteristic data in an initial model;

[0009] obtaining a trained dehydration eccentricity prediction model.

[0010] In the specific embodiment of the above training method of a dehydration eccentricity prediction model, the dehydration characteristic data required for predicting eccentricity of a washing device during a dehydration process comprises:

[0011] initial distribution positions of laundry to be dehydrated,

[0012] and a corresponding relationship of dehydration time, rotation speed, and eccentricity distance.

[0013] In the specific embodiment of the above training method of a dehydration eccentricity prediction model, the dehydration characteristic data comprises experimentally obtained dehydration characteristic data and user historical dehydration characteristic data.

[0014] In the specific embodiment of the training method of the dehydration eccentricity prediction model, the step of "bringing the obtained dehydration feature data into the initial model for training" further comprises:

[0015] bringing the obtained experimental dehydration feature data into the initial model for training to obtain a transition model;

[0016] bringing the obtained user historical dehydration feature data into the transition model for correction of the prediction model to obtain the dehydration eccentricity prediction model.

[0017] In the specific embodiment of the training method of the dehydration eccentricity prediction model, the training method further comprises:

[0018] acquiring new user historical dehydration feature data every preset time,

[0019] bringing the obtained user historical dehydration feature data into the dehydration eccentricity prediction model for continuous correction to obtain a more accurate dehydration eccentricity prediction model.

[0020] In a second aspect, the present application further provides a control method of a washing device, which comprises:

[0021] acquiring real-time data in the dehydration process of the washing device;

[0022] inputting the real-time data into a dehydration eccentricity prediction model for calculation;

[0023] outputting a control instruction based on the calculation result;

[0024] adjusting the dehydration program of the washing device based on the control instruction;

[0025] The dehydration eccentricity prediction model is obtained by the training method of any one of claims 1-5.

[0026] In the specific embodiment of the control method of the washing device, the real-time data comprises:

[0027] initial distribution position of the clothes to be dehydrated,

[0028] and the corresponding relationship of dehydration time, rotation speed, and eccentricity.

[0029] In the specific embodiment of the control method of the washing device, the step of "outputting a control instruction based on the calculation result" further comprises:

[0030] when the calculation result is uneven distribution of the clothes, the output control instruction is an instruction for new speed control of the drum.

[0031] In the specific embodiment of the control method of the washing apparatus, the control method further comprises:

[0032] maintaining the instruction of the existing speed control of the drum when the calculation result is that the laundry is uniformly distributed

[0033] In a third aspect, the present application further provides a washing apparatus, comprising a controller, wherein the controller stores a dehydration eccentricity prediction model obtained by the training method of the first aspect, and the controller is configured to execute the control method of the second aspect.

[0034] With the above technical solution, the present application can train a dehydration eccentricity prediction model by using dehydration characteristic data, so that the washing apparatus provided by the present application can predict the eccentric state of the laundry in real time according to real-time data of the washing apparatus and the dehydration eccentricity prediction model, and output a control instruction for the dehydration program of the washing apparatus according to the predicted eccentric state of the laundry, thereby predicting the distribution state of the laundry in advance during the dehydration process, and solving the technical problem of eccentricity in advance, and avoiding problems such as noise, vibration, displacement, and collision of the washing drum caused by the eccentricity of the laundry. BRIEF DESCRIPTION OF DRAWINGS

[0035] The preferred embodiments of the present application will be described below with reference to the accompanying drawings, in which:

[0036] Figure 1 is a whole flowchart of the training method of the dehydration eccentricity prediction model of the present application;

[0037] Figure 2 is a specific flowchart of step S102 in the training method of the dehydration eccentricity prediction model of the present application;

[0038] Figure 3 is a flowchart of the second embodiment of the training method of the dehydration eccentricity prediction model of the present application;

[0039] Figure 4 is a whole flowchart of the control method of the washing apparatus of the present application;

[0040] Figure 5 is a specific flowchart of step S303 in the control method of the washing apparatus of the present application. DETAILED DESCRIPTION

[0041] The preferred embodiments of the present application will be described below with reference to the accompanying drawings. Those skilled in the art will appreciate that the embodiments are only used to explain the technical principles of the present application, and are not intended to limit the scope of protection of the present application. Those skilled in the art can make adjustments to them as needed in order to adapt to specific application scenarios. Such changes related to application scenarios do not deviate from the basic principles of the present application and are within the scope of protection of the present application.

[0042] In the description of the present application, "module" and "controller" can include hardware, software or a combination of both. A module can include hardware circuitry, various suitable sensors, communication ports, memory, and can also include a software part such as program code, and can be a combination of software and hardware. The controller can be a central controller, a microcontroller, an image processor, a digital signal processor or any other suitable controller. The controller has data and / or signal processing functions. The controller can be implemented in software, hardware or a combination of both. The non-transitory computer readable storage medium includes any suitable medium that can store program code, such as a magnetic disk, a hard disk, an optical disk, a flash memory, a read-only memory, a random access memory, etc. The term "A and / or B" means all possible combinations of A and B, such as only A, only B or both A and B. The term "at least one of A or B" or "at least one of A and B" has a similar meaning as "A and / or B" and can include only A, only B or both A and B. The singular form of the term "one", "this" can also include the plural form.

[0043] In the embodiments of the present disclosure, the terms "upper", "lower", "inner", "middle", "outer", "front", "back" and the like indicate the orientation or positional relationship shown in the drawings. These terms are mainly used to better describe the embodiments of the present disclosure and its embodiments, and are not intended to limit the indicated devices, elements or components to have a specific orientation, or to be constructed and operated in a specific orientation. In addition, in addition to indicating the orientation or positional relationship, the above-mentioned terms can also be used to indicate other meanings, for example, the term "upper" can also be used to indicate a certain attachment relationship or connection relationship in some cases. For those skilled in the art, the specific meaning of these terms in the embodiments of the present disclosure can be understood according to the specific circumstances.

[0044] It should be noted that in the description of the preferred embodiments, unless otherwise explicitly specified and limited, the terms "connected", "communicated" should be understood in a broad sense, for example, it can be directly connected, or indirectly connected through an intermediate medium, or the connection between two elements inside, which cannot be understood as a limitation of the present application. In addition, the terms "first", "second" are for the purpose of description, and those skilled in the art can understand the specific meaning of the above terms in the present application according to the specific circumstances.

[0045] The dehydration model in the conventional washing equipment is usually adjusted based on the real-time speed threshold change or weight correspondence, so it is difficult to predict and judge in advance in the dehydration process, when the judgment opportunity is too early, it will cause the dehydration time to be too long, when the judgment opportunity is too late, it will cause the critical state to shift and the drum to collide. To solve this problem, the present application provides the following scheme, a time series prediction model is established by using semi-supervised learning, whether the clothes distribution is uniform in the dehydration process is predicted in real time, the accuracy and timeliness of the prediction model are increased by combining user washing data and laboratory data, and the following will be introduced in combination with Figures 1-3 .

[0046] As Figure 1 , Figure 2 shown, to solve the problem that the dehydration model of the washing equipment in the prior art cannot predict the eccentricity of clothes in the dehydration process in advance, the present application provides a training method of a dehydration eccentricity prediction model, the training method comprises:

[0047] S101: obtaining dehydration feature data required for predicting eccentricity of the washing equipment in the dehydration process;

[0048] S102: bringing the obtained dehydration feature data into an initial model for training;

[0049] S103: obtaining a dehydration eccentricity prediction model after training.

[0050] In the preferred embodiment, the dehydration feature data required for predicting eccentricity of the washing equipment in the dehydration process in the above step S101 comprises: initial distribution position of clothes to be dehydrated, and corresponding relationship of dehydration time, rotation speed and eccentricity. Wherein, the initial distribution position of clothes to be dehydrated is the position of clothes distributed in the washing drum before dehydration of the washing equipment, such as front, back, one side, uniform, etc., and the corresponding relationship of dehydration time, rotation speed and eccentricity includes the change trend of rotation speed and eccentricity in the dehydration process. It can be understood that the corresponding relationship of dehydration time, rotation speed and eccentricity will change with the different distribution of clothes before dehydration of the washing equipment.

[0051] In the preferred embodiment, the dehydration characteristic data includes experimental acquisition dehydration characteristic data and user historical dehydration characteristic data. The experimental acquisition dehydration characteristic data is obtained through experiments, and specifically includes: placing different weights of counterweights at different positions of the washing drum to simulate the distribution state of the laundry, including front-biased, rear-biased, side-biased, uniform, etc., and obtaining the corresponding relationship among the simulated dehydration time, rotation speed, and eccentricity of the laundry in different distribution states through repeated experiments.

[0052] The step of "bringing the obtained dehydration characteristic data into the initial model for training" in the above step S102 further includes:

[0053] S1021: Bringing the obtained experimental acquisition dehydration characteristic data into the initial model for training to obtain a transition model;

[0054] S1022: Bringing the obtained user historical dehydration characteristic data into the transition model to correct the prediction model, so as to obtain a dehydration eccentricity prediction model.

[0055] In the preferred embodiment, since the experimental acquisition dehydration characteristic data has a relatively complete label and high data authenticity, the data is used to preliminarily train the initial model to obtain a transition model, and then semi-supervised training is performed through the user historical dehydration characteristic data, so that the obtained dehydration eccentricity prediction model is more accurate. Optionally, the training method of bringing the dehydration characteristic data into the initial model for training can include a neural network training method, a support vector machine method, etc.; and the training method of semi-supervised training of the user historical dehydration characteristic data includes a pseudo-label prediction method, a direct-push support vector machine method, etc.

[0056] In the case of using the above embodiment, the dehydration eccentricity prediction model of the washing device in the dehydration process can be trained by using the dehydration characteristic data required for the predicted eccentricity of the washing device in the dehydration process.

[0057] The following refers to Figure 3 A second embodiment of the training method of the dehydration eccentricity prediction model of the present application is introduced. In the second embodiment of the training method of the dehydration eccentricity prediction model of the present application, Figure 3 The flowchart of the second embodiment of the training method of the dehydration eccentricity prediction model of the present application is shown in the second embodiment of the present application, and the training method of the dehydration eccentricity prediction model includes:

[0058] S201: obtaining the dehydration characteristic data required for the predicted eccentricity of the washing device in the dehydration process;

[0059] S202: bring the obtained dehydration feature data into the initial model for training;

[0060] S203: obtain the trained dehydration eccentricity prediction model;

[0061] S204: obtain new user historical dehydration feature data every preset time;

[0062] S205: bring the obtained user historical dehydration feature data into the dehydration eccentricity prediction model for continuous correction to obtain a more accurate dehydration eccentricity prediction model.

[0063] In the preferred embodiment, the "dehydration feature data required for predicting the eccentricity of the washing device during the dehydration process" in the above step S201 includes the initial distribution position of the clothes to be dehydrated, and the corresponding relationship of the dehydration time, the rotation speed, and the eccentricity. The initial distribution position of the clothes to be dehydrated is the position of the clothes distributed in the washing drum before dehydration, such as front, back, one side, uniform, etc. The corresponding relationship of the dehydration time, the rotation speed, and the eccentricity includes the change trend of the rotation speed and the eccentricity during the dehydration process. It can be understood that the corresponding relationship of the dehydration time, the rotation speed, and the eccentricity will change with the different distribution of the clothes before dehydration.

[0064] In the preferred embodiment, the dehydration feature data includes experimental dehydration feature data and user historical dehydration feature data. The experimental dehydration feature data is obtained through experiments, and specifically includes: placing different weights of counterweights on different positions of the washing drum to simulate the distribution state of the clothes, including front, back, one side, uniform, etc. After multiple repeated experiments, the corresponding relationship of the simulated dehydration time, the rotation speed, and the eccentricity of the clothes in different distribution states is obtained. The user historical dehydration feature data is obtained by collecting the real use data of the user when using the washing device, and the corresponding relationship of the simulated dehydration time, the rotation speed, and the eccentricity of the clothes in different distribution states is obtained according to the various washing methods of the user.

[0065] In the preferred embodiment, since the experimental dehydration feature data label is more complete and the data authenticity is higher, the data is used to preliminarily train the initial model to obtain a transition model, and then semi-supervised training is performed through the user historical dehydration feature data, so that the obtained dehydration eccentricity prediction model is more accurate. Optionally, the training method of bringing the dehydration feature data into the initial model for training can include a neural network training method, a support vector machine method, etc.; and the training method of semi-supervised training of the user historical dehydration feature data includes a pseudo-label prediction method, a direct support vector machine method, etc.

[0066] In the case of adopting the above-mentioned embodiments, the dewatering eccentricity prediction model of the washing equipment in the dewatering process can be trained by the dewatering characteristic data required for predicting the eccentricity of the washing equipment in the dewatering process. Meanwhile, the dewatering eccentricity prediction model is more accurate and has stronger applicability by continuously obtaining new user historical dewatering characteristic data and continuously correcting the dewatering eccentricity prediction model.

[0067] In the dewatering process of the conventional washing equipment, the uneven distribution of clothes in the drum often causes large eccentricity during dewatering. If the clothes distribution is not adjusted in time after this situation occurs, it will cause unbalanced dewatering, and further cause serious problems such as noise, vibration, displacement, and drum collision. At the same time, it will also make the dewatering speed unable to rise to high speed, prolong the dewatering time of the clothes, cause the dewatering to be incomplete, and seriously interfere with the use of the user. In the prior art, the washing equipment judges the clothes distribution state by weight before dewatering, which is difficult to predict and judge in advance during the dewatering process. When the determination time is early, it will cause the dewatering speed to be reduced and the dewatering time to be too long. When the determination time is late, it will cause the washing drum to displace and collide in the critical state. Therefore, in order to solve the above technical problems, the present application provides the following solutions, which will be introduced below in combination with Figures 4-5 .

[0068] As Figure 4 , Figure 5 shown, in order to solve the problem that the washing equipment in the prior art is difficult to predict the clothes distribution state in advance during the dewatering process, the present application provides a control method of a washing equipment, which comprises:

[0069] S301: acquiring real-time data in the dewatering process of the washing equipment;

[0070] S302: inputting the real-time data into a dewatering eccentricity prediction model for calculation;

[0071] S303: outputting a control instruction based on the calculation result;

[0072] S304: adjusting the dewatering program of the washing equipment based on the control instruction.

[0073] In the preferred embodiment, the "acquiring real-time data in the dewatering process of the washing equipment" in the above-mentioned step S301 comprises: the initial distribution position of the clothes to be dewatered, and the real-time rotation speed and the real-time eccentricity.

[0074] In the case of employing the above-mentioned embodiments, the washing apparatus provided by the present application can predict the eccentricity state of the laundry in real time according to real-time data of the washing apparatus and the dewatering eccentricity prediction model, and output a control instruction for the dewatering program of the washing apparatus according to the predicted eccentricity state of the laundry, so as to predict the problem of the distribution state of the laundry in advance during the dewatering process, and avoid the problems of noise, vibration, displacement, and cylinder collision of the washing drum caused by the eccentricity of the laundry.

[0075] For example, when the washing apparatus enters the dewatering program, the washing apparatus acquires the initial distribution position of the laundry to be dewatered, and acquires the rotating speed and the eccentricity in real time, and inputs the real-time data into the dewatering eccentricity prediction model stored in the washing apparatus for real-time calculation, and outputs a control instruction by predicting the distribution state of the laundry of the washing apparatus, wherein the control instruction includes deceleration, shaking, swinging, and continuous speed-up, and the like, so that the washing apparatus can adjust the dewatering program according to the control instruction.

[0076] The "outputting a control instruction based on the calculation result" in the above-mentioned step S303 further includes:

[0077] S3031: When the calculation result is that the laundry is unevenly distributed, outputting a control instruction for new speed control of the drum.

[0078] S3032: When the calculation result is that the laundry is evenly distributed, maintaining the existing speed control instruction of the drum.

[0079] In the preferred embodiment, when the calculation result is that the laundry is unevenly distributed, it indicates that the drum will have an eccentricity problem, and thus a control instruction for new speed control of the drum needs to be outputted, which includes deceleration, shaking, swinging, and the like, so as to reset and change the distribution of the laundry, and then continue to speed up for dewatering after the distribution of the laundry is more uniform, so as to avoid the problems of noise, vibration, displacement, and cylinder collision of the washing drum caused by the eccentricity of the laundry. For example, when the calculation result is that the laundry is unevenly distributed, the washing apparatus outputs a shaking instruction within a preset time, and the preset time can be 10s, 15s, 20s, and the like. When the calculation result is that the laundry is evenly distributed, the existing speed control instruction of the drum is maintained to complete the dewatering task as soon as possible.

[0080] The present application also provides a washing apparatus, which comprises a controller, wherein the controller stores the dewatering eccentricity prediction model obtained by the training method of the dewatering eccentricity prediction model, and the controller is configured to execute the control method of the washing apparatus.

[0081] To sum up, the washing equipment provided by the present application can predict the eccentric state of the clothes in real time according to the real-time data of the washing equipment and the dehydration eccentric prediction model, and output control instructions to the dehydration program of the washing equipment according to the predicted eccentric state of the clothes, so that the problem of clothes distribution state can be predicted in advance during the dehydration process, and problems such as noise, vibration, displacement, and cylinder collision caused by clothes eccentricity can be avoided.

[0082] It should be noted that the above embodiments are only used to illustrate the principles of the present application and are not intended to limit the protection scope of the present application. Those skilled in the art can adjust the above structure without deviating from the principles of the present application, so that the present application can be applied to more specific application scenarios.

[0083] Those skilled in the art can understand that the control method of the washing equipment and the washing equipment described above also include some other well-known structures, such as processors, controllers, memories, etc. The memory includes but is not limited to random access memory, flash memory, read-only memory, programmable read-only memory, volatile memory, non-volatile memory, serial memory, parallel memory, or register, etc. The processor includes but is not limited to CPLD / FPGA, DSP, ARM processor, MIPS processor, etc. In order not to unnecessarily obscure the embodiments of the present disclosure, these well-known structures are not shown in the drawings.

[0084] Although the steps in the above embodiments are described in the above order, those skilled in the art can understand that, in order to achieve the effect of the present embodiment, the different steps do not have to be executed in such an order, they can be executed simultaneously (in parallel) or in a reversed order, and these simple changes are within the protection scope of the present application.

[0085] So far, the technical solutions of the present application have been described in combination with the preferred embodiments shown in the drawings, but those skilled in the art can easily understand that the protection scope of the present application is obviously not limited to these specific embodiments. Those skilled in the art can make equivalent changes or replacements to the related technical features without deviating from the principles of the present application, and the technical solutions after these changes or replacements will fall within the protection scope of the present application.

Claims

1. A method for training a dehydrated eccentricity prediction model, characterized in that, The training method comprises: obtaining dewatering characteristic data required for predicting eccentricity of the washing equipment in a dewatering process; feeding the obtained dewatering characteristic data into an initial model for training; obtaining a dewatering eccentricity prediction model after the training is completed; wherein the dewatering characteristic data required for predicting eccentricity of the washing equipment in a dewatering process comprises initial distribution positions of clothes to be dewatered, and corresponding relationships among dewatering time, rotation speed, and eccentricity distance, and the dewatering characteristic data comprises experimentally obtained dewatering characteristic data and user historical dewatering characteristic data; the step of "feeding the obtained dewatering characteristic data into an initial model for training" further comprises: feeding the obtained experimentally obtained dewatering characteristic data into an initial model for training to obtain a transition model; feeding the obtained user historical dewatering characteristic data into the transition model for correction of the prediction model to obtain the dewatering eccentricity prediction model. 2.The method of claim 1, wherein, The training method further comprises: obtaining new user historical dewatering characteristic data every preset time, feeding the obtained user historical dewatering characteristic data into the dewatering eccentricity prediction model for continuous correction to obtain a more accurate dewatering eccentricity prediction model.

3. A control method of a washing apparatus, characterized by, The control method comprises: obtaining real-time data in a dewatering process of the washing equipment; feeding the real-time data into a dewatering eccentricity prediction model for calculation; outputting a control instruction based on the calculation result; adjusting a dewatering program of the washing equipment based on the control instruction; wherein the dewatering eccentricity prediction model is the dewatering eccentricity prediction model obtained by the training method of any one of claims 1-2.

4. The control method of a washing apparatus according to claim 3, characterized by, The real-time data comprises: initial distribution positions of clothes to be dewatered, and corresponding relationships among dewatering time, rotation speed, and eccentricity distance.

5. The control method of a washing apparatus according to claim 3, characterized by, The step of "outputting a control instruction based on the calculation result" further comprises: when the calculation result is uneven distribution of clothes, outputting a control instruction for new speed control of the drum.

6. The control method of a washing apparatus according to claim 5, characterized by, The control method further comprises: when the calculation result is even distribution of clothes, maintaining an existing speed control instruction of the drum.

7. A washing apparatus characterized by comprising: The washing equipment comprises a controller, the controller stores the dewatering eccentricity prediction model obtained by the training method of any one of claims 1-2, and the controller is configured to execute the control method of any one of claims 3-6.

Images