A training method and device of a washing machine AI control model and a washing machine

Abstract

The application discloses a training method and device of a washing machine AI control model and a washing machine, and relates to the technical field of artificial intelligence application; a model training module is arranged, which is provided with a peripheral parameter fault threshold value; a plurality of washing machine complete machines are arranged, which comprise a first control module, a dewatering motor connected with the first control module, a load acquisition module and an eccentricity acquisition module; the first control module is in communication connection with the model training module; a plurality of peripheral detection modules are arranged, which are connected with the washing machine complete machines respectively, are used for acquiring peripheral parameters, and are in communication connection with the model training module; dewatering load, dewatering eccentricity and peripheral parameters in an accelerated dewatering control stage are acquired cyclically; it is judged whether the peripheral parameters reach or exceed the peripheral parameter fault threshold value; if yes, the peripheral parameters, corresponding dewatering load and dewatering eccentricity are collected as first training data, and an accelerated dewatering AI control model applied to the accelerated dewatering control stage of the washing machine is trained. The application improves the training efficiency and accuracy of the AI control model.

Description

TECHNICAL FIELD

[0001] The present application belongs to the technical field of artificial intelligence application, and particularly relates to a training method and device of a washing machine AI control model and a washing machine. BACKGROUND

[0002] At present, the training data for developing intelligent control of a washing machine system mainly comes from the running data of the washing machine itself. The data without representing the machine running state makes the model for controlling the washing machine trained with poor accuracy and insufficient intelligence. In addition, the data of the machine running state needs to be recorded and imported manually, and the training efficiency is low.

[0003] The above information disclosed in the background of the application is only used to increase the understanding of the background of the application, and therefore, it can include prior art known to those skilled in the art. SUMMARY

[0004] The present application is directed to the problem of low training efficiency, low accuracy of the intelligent control model and insufficient intelligence caused by only using the running data of the washing machine itself as the training data in the prior art. A training method and device of a washing machine AI control model and a washing machine are proposed, which increases the external device data representing the machine running state to participate in the training of the AI control model through automatic acquisition, and improves the training efficiency and the accuracy and intelligence of the AI control model.

[0005] To achieve the above-mentioned application / design purposes, the present application adopts the following technical solutions.

[0006] A training method of a washing machine AI control model, comprising:

[0007] S1, setting a model training module, and presetting a peripheral parameter fault threshold;

[0008] S2, setting a plurality of experimental washing machines, which include a first control module, a dehydration motor connected with the first control module, a load acquisition module and an eccentricity acquisition module for acquiring the load and the eccentricity; the first control module is in communication connection with the model training module;

[0009] S3, setting a plurality of peripheral detection modules, which are respectively connected with each washing machine, are used for acquiring peripheral parameters, are in communication connection with the model training module, and output the peripheral parameters to the model training module;

[0010] S4, placing clothes with an initial load in the washing machine to perform dehydration; and cyclically acquiring the dehydration load, the dehydration eccentricity and the peripheral parameters in the acceleration dehydration control stage;

[0011] S5, judging whether the peripheral parameter reaches or exceeds the peripheral parameter failure threshold value; if yes, performing S6; if no, performing S4;

[0012] S6; collecting the peripheral parameter, the corresponding dehydration load, and the dehydration eccentricity as first training data for training an accelerated dehydration AI control model.

[0013] In some specific embodiments, the peripheral detection module comprises a second control module and a vibration sensor and a force sensor connected to the second control module; the vibration sensor and the force sensor are used to detect the vibration value and the force value of the washing machine and transmit them to the second control module; the second control module is in communication connection with the model training module and transmits the obtained vibration value and force value to the model training module; the model training module identifies it as the peripheral parameter;

[0014] The peripheral parameter failure threshold value comprises a vibration threshold value and a force threshold value; judging whether the peripheral parameter reaches or exceeds the peripheral parameter failure threshold value is that the vibration value reaches or exceeds the vibration threshold value and / or the force value reaches or exceeds the force threshold value.

[0015] In some specific embodiments, the peripheral detection module comprises a second control module and a noise sensor and a force sensor connected to the second control module; the noise sensor and the force sensor are used to detect the noise value and the force value of the washing machine and transmit them to the second control module; the second control module is in communication connection with the model training module and transmits the obtained noise value and force value to the model training module; the model training module identifies it as the peripheral parameter;

[0016] The peripheral parameter failure threshold value comprises a noise threshold value and a force threshold value; judging whether the peripheral parameter reaches or exceeds the peripheral parameter failure threshold value is that the noise value reaches or exceeds the noise threshold value and / or the force value reaches or exceeds the force threshold value.

[0017] In some specific embodiments, it further comprises:

[0018] S1', presetting a plurality of different distribution parameters in the model training module, which comprises a first rotation speed and a first acceleration; presetting a plurality of different distribution load corresponding distribution eccentricity threshold values;

[0019] S2', the model training module sends the distribution parameters to the first control module to control each washing machine to run in a distribution control phase; when the distribution control phase is completed, the distribution load and the distribution eccentricity of the washing machine are obtained.

[0020] S3', determine whether the distribution eccentricity reaches or exceeds the eccentricity threshold corresponding to the distribution load; if yes, execute S4'; if no, execute S5';

[0021] S4', terminate the dehydration control;

[0022] S5', obtain the initial load, the corresponding distribution parameter, the distribution load, the distribution eccentricity and collect them as second training data for training the distribution AI control model.

[0023] A training device of a washing machine AI control model, comprising a plurality of washing machine whole machines, a plurality of external device detection modules and a model training module;

[0024] Each of the washing machine whole machines is used to load clothes of an initial load, and comprises a first control module, a dehydration motor connected thereto, a load acquisition module and an eccentricity acquisition module;

[0025] Each of the external device detection modules is connected to the corresponding washing machine whole machine, and is used to detect an external device parameter when the washing machine whole machine is running;

[0026] The model training module is a hardware main body with a processor and a memory, is communicatively connected to each of the first control modules and each of the external device detection modules, is configured with a data training tool, is preconfigured with an external device parameter fault threshold, and is configured to acquire the external device parameter, a dehydration load and a dehydration eccentricity in real time during the operation of the acceleration dehydration control phase of each of the washing machine whole machines; determine whether the external device parameter reaches the external device parameter fault threshold; if yes, collect the dehydration load and the dehydration eccentricity as first training data for training an acceleration dehydration AI control model.

[0027] In some specific embodiments, the external device detection module comprises a second control module and a vibration sensor and four force sensors connected to the second control module; the vibration sensor and each of the force sensors are used to detect a vibration value of the washing machine whole machine and a force vector of each foot, and transmit them to the second control module; the second control module is communicatively connected to the model training module, and transmits the vibration value and each of the force vectors to the model training module; the model training module obtains a force value of the washing machine whole machine from each of the force vectors, and identifies the vibration value and the force value as the external device parameter;

[0028] The external device parameter fault threshold comprises a vibration threshold and a force threshold; determine whether the vibration value reaches or exceeds the vibration threshold and / or the force value reaches the force threshold; if yes, collect the dehydration load and the dehydration eccentricity as the first training data for training the acceleration dehydration AI control model.

[0029] In some specific embodiments, the peripheral device detection module comprises a second control module and a noise sensor and four force sensors connected to the second control module; the noise sensor and each force sensor are used to detect a noise value and a force vector of each foot and transmit to the second control module; the second control module is in communication connection with the model training module, and transmits the noise value and each force vector to the model training module; the model training module obtains a force value of the washing machine whole machine from each force vector, and identifies the noise value and the force value as the peripheral device parameters;

[0030] The peripheral device parameter fault threshold value comprises a noise threshold value and a force threshold value; it is judged whether the noise value reaches or exceeds the noise threshold value and / or the force value reaches the force threshold value; if yes, the dehydration load and the dehydration eccentricity are collected as the first training data for training the accelerated dehydration AI control model.

[0031] In some specific embodiments, the model training module further presets a plurality of groups of distribution parameters and eccentricity threshold values corresponding to different distribution loads, and is configured to:

[0032] send the distribution parameters to each first control module, control each washing machine whole machine to operate according to the distribution parameters in the distribution control stage, and return the distribution load and distribution eccentricity measured after the distribution control stage ends;

[0033] judge whether the distribution eccentricity reaches or exceeds the eccentricity threshold value corresponding to the distribution load; if yes, stop the dehydration control; if no, control to enter the accelerated dehydration control stage, and collect the initial load, the distribution parameters, the distribution load and the distribution eccentricity as second training data for training the distribution AI control model.

[0034] A washing machine comprises a first control module, a dehydration motor, a load acquisition module and an eccentricity acquisition module connected thereto respectively; the first control module is configured with an accelerated dehydration AI control model obtained by the training device of the washing machine AI control model, which detects or predicts whether a dehydration fault occurs according to the real-time obtained dehydration load and dehydration eccentricity; and if yes, controls the dehydration to be interrupted.

[0035] In some specific embodiments, the first control module is further configured with a distribution AI control model, which controls the operation of the distribution control stage according to the initial load and the corresponding distribution parameters, so that the distribution eccentricity after the operation of the distribution control stage meets the requirements for entering the accelerated dehydration stage.

[0036] Compared with the prior art, the advantages and positive effects of the present application are:

[0037] The training method and device of the washing machine AI control model and the washing machine of the present application combine the load and eccentricity data obtained by the load and eccentricity acquisition module of the washing machine whole machine and the external parameter data obtained by the external detection module arranged outside the washing machine whole machine to form first training data to train the accelerated dewatering AI control model, which is used for control in the accelerated dewatering control stage; the data output by the washing machine whole machine itself and the external data obtained by the external detection module of the washing machine whole machine are combined to form first training data, which expands the dimension of the data used for training the AI control model and improves the accuracy and precision of the training; the acquisition of the training data and the training of the AI control model are automatically realized through communication with the washing machine whole machine and the external detection module, which improves the degree of automation and further improves the efficiency of the training data acquisition and the efficiency of the AI control model training.

[0038] Other features and advantages of the present application will become more apparent after reading the detailed description of the application in conjunction with the accompanying drawings. BRIEF DESCRIPTION OF DRAWINGS

[0039] In order to more clearly illustrate the technical solutions in the embodiments of the present application, the drawings needed in the embodiments will be briefly introduced below. Obviously, the drawings described below are some embodiments of the present application, and other drawings can also be obtained by those skilled in the art without creative labor.

[0040] Figure 1 is a schematic diagram of the composition structure of the training device of the washing machine AI control model according to the embodiment;

[0041] Figure 2 is a schematic diagram of the dewatering process of the washing machine according to the embodiment;

[0042] Figure 3 is a schematic diagram of the training method of the washing machine AI control model according to the embodiment;

[0043] Figure 4 is a schematic diagram of the first training data acquisition process according to the embodiment;

[0044] Figure 5 is a schematic diagram of the first training data acquisition process according to the embodiment;

[0045] Figure 6 is a schematic diagram of the second training data acquisition process according to the embodiment;

[0046] Figure 7 is a schematic diagram of the composition structure of the training device of the washing machine AI control model according to the embodiment;

[0047] Figure 8 is a schematic diagram of the constituent structure of a training device of a washing machine AI control model according to an embodiment.

[0048] In the drawings,

[0049] 1, model training module; 2, washing machine whole machine; 3, peripheral device detection module; 4, first communication module; 21, first control module; 22, dehydration motor; 23, load acquisition module; 24, eccentricity acquisition module; 31, second communication module; 32, second control module; 33, force sensor; 34, vibration sensor; 35, noise sensor. DETAILED DESCRIPTION

[0050] The technical solutions in the embodiments of the present application will be described clearly and completely below with reference to the drawings in the embodiments of the present application. Obviously, the described embodiments are only some of the embodiments of the present application, not all. Based on the embodiments in the present application, all other embodiments obtained by those skilled in the art without creative labor fall within the scope of protection of the present application.

[0051] In the description of the present application, it should be understood that the terms "center", "upper", "lower", "front", "rear", "left", "right", "vertical", "horizontal", "top", "bottom", "inner", "outer" and the like indicate the orientation or positional relationship shown in the drawings, and are only for the convenience of describing the present application and simplifying the description, and do not indicate or imply that the device or element referred to must have a particular orientation, be constructed and operated in a particular orientation, therefore, it cannot be understood as a limitation on the present application.

[0052] In the description of the present application, it should be noted that unless otherwise explicitly specified and limited, the terms "mounting", "connecting", "connection" should be understood broadly, for example, it can be fixed connection, or detachable connection, or integrally connected. For those skilled in the art, the specific meaning of the above terms in the present application can be understood according to the specific circumstances. In the description of the embodiments, specific features, structures, materials or characteristics can be combined in any one or more embodiments or examples in a suitable manner.

[0053] The terms "first", "second" are only for descriptive purposes, and cannot be understood as indicating or implying relative importance or implicitly indicating the number of the indicated technical features. Therefore, the features defined with "first", "second" can explicitly or implicitly include one or more of the features.

[0054] In the description of the present application, unless otherwise specified, the meaning of "multiple" is two or more.

[0055] Referring toFigure 2 This invention discloses a training method for an AI control model of a washing machine, wherein the trained AI control model is applied to the control of the washing machine's spin-drying process.

[0056] The spin-drying process of a washing machine includes a distributed control stage, corresponding to... Figure 1 The AC segment is used to control the rotation speed of the washing machine drum, so that the clothes that need to be spun are distributed in the drum and the eccentricity is reduced. Generally, it accelerates to the first speed and maintains the first speed for a first time; then it accelerates from the first speed to the second speed with the first acceleration for a second time, so that the clothes in the drum are evenly distributed.

[0057] The spin-drying process of a washing machine also includes an accelerated spin-drying judgment stage and an accelerated spin-drying control stage; the accelerated spin-drying judgment stage involves maintaining the second spin speed continuously for a third duration, corresponding to... Figure 1 The CD segment of the process involves detecting or acquiring distributed load and eccentricity during the accelerated dehydration judgment phase, and determining whether to enter the accelerated dehydration control phase based on the acquired distributed load and eccentricity. The accelerated dehydration control phase involves running at the second acceleration for a fourth duration to further dehydrate the clothes in the inner drum. Figure 1 The DE segment in the washing machine. Washing machines that have passed this stage will not experience any spin-drying malfunctions in the subsequent spin-drying stage. Spin-drying malfunctions include vibration malfunctions, movement malfunctions, etc.

[0058] Vibration fault is when the vibration or noise value of the washing machine 2 reaches or exceeds the specified threshold; movement fault is when the force value of the washing machine 2 reaches or exceeds the specified threshold.

[0059] The distributed load is the load inside the inner tube obtained during the accelerated dehydration judgment stage; the distributed eccentricity is the eccentricity of the inner tube clothing obtained during the accelerated dehydration judgment stage.

[0060] Reference Figure 1 , Figure 2 , Figure 3 The training method for the washing machine AI control model includes acquiring parameters of the operation process during the accelerated spin-drying control stage as the first training data for training the accelerated spin-drying AI control model. The accelerated spin-drying AI control model is used for intelligent control during the accelerated spin-drying control stage of the washing machine, specifically including:

[0061] S1. Set the model training module 1 to its preset peripheral parameter fault threshold;

[0062] S2, a plurality of washing machine whole machines 2 are arranged, which are respectively connected in communication with the model training module 1; the washing machine whole machine 2 includes a first control module 21 and a dehydration motor 22 connected with the first control module 21, a load acquisition module 23 and an eccentricity acquisition module 24; the dehydration load and the dehydration eccentricity of the washing machine whole machine 2 in the accelerated dehydration control stage are acquired by the load acquisition module 23 and the eccentricity acquisition module 24 and transmitted to the model training module 1; the dehydration load is the real-time load in the accelerated dehydration control stage; the dehydration eccentricity is the real-time eccentricity in the accelerated dehydration control stage;

[0063] S3, the peripheral detection module 3 is arranged corresponding to each washing machine whole machine 2; the peripheral parameter of the washing machine whole machine 2 in the accelerated dehydration control stage is acquired by the peripheral detection module 3 and transmitted to the model training module 1;

[0064] S4, the model training module 1 cyclically acquires the dehydration load, the dehydration eccentricity and the peripheral parameter in the accelerated dehydration control stage;

[0065] S5, it is judged in real time whether the peripheral parameter reaches or exceeds the peripheral parameter fault threshold value; if yes, S6 is executed; if no, S4 is executed;

[0066] S6, the dehydration eccentricity and the dehydration load corresponding to the peripheral parameter are acquired, collected as the first training data and used for training the accelerated dehydration AI control model.

[0067] The training method of the washing machine AI control model of the present application combines the load and eccentricity data acquired by the load and eccentricity acquisition module 24 of the washing machine whole machine 2 and the peripheral parameter data acquired by the peripheral detection module 3 arranged outside the washing machine whole machine 2 to form the first training data for training the accelerated dehydration AI control model, which is used for the control in the accelerated dehydration control stage; the data output acquired by the washing machine whole machine 2 itself and the external data acquired by the peripheral detection module 3 of the washing machine whole machine 2 are combined to form the first training data, which expands the dimension of the data used for training the AI control model and improves the accuracy and precision of the training; the acquisition of the training data and the training of the AI control model are automatically realized through the communication with the washing machine whole machine 2 and the peripheral detection module 3, which improves the automation degree and further improves the efficiency of the training data acquisition and the efficiency of the AI control model training.

[0068] The specific flow and principle of the training method of the washing machine AI control model of the present application will be described in detail below through specific embodiments.

[0069] In some specific embodiments, with reference to Figure 1 、 Figure 2 、 Figure 3 、 Figure 4 , the training method of the accelerated dehydration AI control model specifically includes:

[0070] S7, the peripheral parameter fault threshold value preset by the model training module 1 includes a vibration threshold value and a force threshold value, and the model training module 1 is in communication connection with the plurality of washing machine whole machines 2;

[0071] S8, a peripheral detection module 3 is arranged for each washing machine whole machine 2 respectively; the peripheral detection module 3 includes a second control module 32 and a vibration sensor 34 and a force sensor 33 connected with the second control module 32; the second control module 32 is in communication connection with the model training module 1; the model training module 1 obtains the vibration value and the force value of the washing machine whole machine 2 in the accelerated dehydration control stage of the washing machine whole machine 2 through the peripheral detection module 3, and obtains the dehydration load and the dehydration eccentricity in the accelerated dehydration control stage through a load obtaining module 23 and an eccentricity obtaining module 24 of the washing machine whole machine 2 itself; the dehydration load is the real-time load in the accelerated dehydration control stage; and the dehydration eccentricity is the real-time eccentricity in the accelerated dehydration control stage;

[0072] S9, the model training module 1 judges whether the vibration value reaches or exceeds the vibration threshold value and / or the force value reaches or exceeds the force threshold value in real time; if yes, S10 is executed; and if no, S8 is executed;

[0073] S10, the dehydration eccentricity and the dehydration load corresponding to the vibration value and / or the force value are obtained, and the first training data are collected, which are used for training the accelerated dehydration AI control model.

[0074] The vibration value reaching or exceeding the vibration threshold value is vibration fault occurrence; the force value reaching or exceeding the force threshold value is movement fault occurrence; and the movement fault is that the force value of the washing machine includes a lateral force for overcoming friction to move the washing machine laterally.

[0075] In some specific embodiments, with reference to Figure 1 、 Figure 2 、 Figure 3 、 Figure 5 , the training method of the accelerated dehydration AI control model specifically includes:

[0076] S7', the model training module 1 presets a noise threshold value and a force threshold value, and is in communication connection with the plurality of washing machine whole machines 2; that is, the peripheral parameter fault threshold value includes a noise threshold value and a force threshold value;

[0077] S8', an external detection module 3 is arranged on each washing machine 2; the external detection module 3 comprises a second control module 32, a noise sensor 35 and a force sensor 33 connected with the second control module 32; the second control module 32 is in communication connection with the model training module 1; the model training module 1 obtains the noise value and the force value of the washing machine 2 in the accelerated dehydration control stage through the external detection module 3, and obtains the dehydration load and the dehydration eccentricity in the accelerated dehydration control stage through the load obtaining module 23 and the eccentricity obtaining module 24 of the washing machine 2 itself;

[0078] S9', the model training module 1 judges whether the noise value reaches or exceeds the noise threshold value and / or the force value reaches or exceeds the force threshold value; if yes, S10' is executed; if no, S8' is executed;

[0079] S10', the dehydration eccentricity and the dehydration load corresponding to the noise value and / or the force value are obtained, and the first training data are collected for training the accelerated dehydration AI control model.

[0080] The vibration value exceeding the vibration threshold value indicates that the vibration fault occurs; the force value exceeding the force threshold value indicates that the moving fault occurs; the moving fault is that the force value of the washing machine includes the lateral force for overcoming the friction force to move the washing machine laterally.

[0081] In some specific embodiments, with reference to Figure 1 、 Figure 2 、 Figure 3 、 Figure 4 、 Figure 5 、 Figure 6 , the training method of the washing machine AI control model further comprises obtaining the parameters of the distribution load and the distribution eccentricity corresponding to different initial loads, first rotation speeds and first accelerations as training data for training the distribution AI control model, and specifically comprising:

[0082] S1', a plurality of different distribution parameters are preset in the model training module 1, which include the first rotation speed and the first acceleration; a plurality of different distribution load corresponding distribution eccentricity threshold values are preset;

[0083] S2', the model training module 1 sends the distribution parameters to the first control module 21 to control the operation of each washing machine 2 in the distribution control stage;

[0084] S3', the distribution control stage is completed, that is, the accelerated dehydration judgment stage, the distribution load and the distribution eccentricity of the washing machine 2 are obtained;

[0085] S4', determine whether the distribution eccentricity reaches or exceeds the eccentricity threshold corresponding to the distribution load; if yes, execute S5'; if no, execute S6';

[0086] S5', terminate the dehydration control;

[0087] S6', collect the initial load, the corresponding distribution parameter, the distribution load, and the distribution eccentricity as the second training data for training the distribution AI control model; and enter the accelerated dehydration control phase.

[0088] The training method of the washing machine AI control model of the embodiment obtains the distribution AI control model by obtaining the initial load and the distribution parameter that meet the distribution eccentricity threshold, and is used for control in the distribution control phase of the dehydration process, avoids the situation that the distribution eccentricity exceeds the eccentricity threshold, and improves the intelligence of the dehydration process control, the accuracy of the control, and the efficiency of the dehydration.

[0089] In some specific embodiments, the eccentricity can be obtained by installing an acceleration sensor at the bearing or outer cylinder support structure of the washing machine inner cylinder to collect the acceleration signal in real time when the washing machine inner cylinder rotates; the acceleration signal is subjected to Fourier transform to analyze the amplitude change of the fundamental frequency and harmonic components to obtain the eccentricity value. That is, the eccentricity value is measured by the amplitude change value of the fundamental frequency and harmonic components of the acceleration.

[0090] In some specific embodiments, the acceleration is obtained by differentiating the rotational speed data of the dehydration motor 22 in the uniform speed stage to obtain the acceleration, and the difference between the maximum acceleration and the minimum acceleration in the continuous interval is used to measure the eccentricity in the accelerated dehydration judgment phase.

[0091] In some specific embodiments, the real-time torque current of the dehydration motor 22 is obtained, and the above real-time torque current signal is subjected to low-pass digital filtering processing to eliminate high-frequency noise interference; the maximum torque current and the minimum torque current in a plurality of continuous dehydration cycles are extracted, and the average values of each maximum torque current and each minimum torque current are calculated, and the difference value is calculated to represent the eccentricity value. That is,

[0092] e=K(Imax_avg-Imin_avg )

[0093] wherein,

[0094] e is the eccentricity;

[0095] K is a model calibration coefficient to avoid model errors;

[0096] Imax_avg is the average value of each maximum torque current;

[0097] Imin_avg is the average value of each minimum torque current.

[0098] In some specific embodiments, in the training method of the washing machine AI control model, the plurality of washing machine whole machines 2 respectively place different initial loads or the plurality of washing machine whole machines 2 group different initial loads of clothes and each washing machine whole machine 2 or each group of washing machine whole machines 2 places different initial loads in turn, while receiving the distribution parameters sent by the model training module 1, and controlling the distribution control stage of each washing machine whole machine 2 with the received distribution parameters, for obtaining the first training data and the second training data.

[0099] The training method of the washing machine AI control model of the present application obtains the first training data and the second training data by placing different initial loads of clothes in the plurality of washing machine whole machines 2 or grouping different initial loads of clothes and controlling at the same time with the same distribution parameters, improves the efficiency of obtaining the first training data and the second training data, and the breadth and dimension of the data, and improves the efficiency and accuracy of model training.

[0100] In some specific embodiments, the accelerated dehydration AI control model is a first fitting function of dehydration load and dehydration eccentricity obtained by training the first training data, and the washing machine intelligently judges whether an external parameter fault occurs according to whether the self-measured dehydration load and dehydration eccentricity meet the first fitting function, and interrupts dehydration when the external parameter fault occurs; otherwise, continue to control accelerated dehydration.

[0101] The training method of the washing machine AI control model of the present embodiment obtains the first fitting function by fitting the first training data, so that the washing machine can intelligently judge whether to interrupt dehydration according to whether the dehydration load and the dehydration eccentricity meet the first fitting function, protect the washing machine, and prolong the service life of the washing machine.

[0102] In some specific embodiments, the distribution control model is a second fitting function of initial load and distribution parameter obtained by the second training data, so that the washing machine intelligently configures different distribution parameters to control the operation of the distribution control stage of the washing machine according to the self-measured different initial loads and the second fitting function, so that the distribution eccentricity after the operation of the distribution control stage meets the requirement of entering the accelerated dehydration control stage, and improves the dehydration rate.

[0103] The training method of the washing machine AI control model of the present embodiment obtains the second fitting function by fitting the second training data, so that the washing machine can intelligently configure the distribution parameters according to the initial load, so that the distribution eccentricity meets the requirement of accelerated dehydration control, and reduces the probability of external parameter fault when entering the accelerated dehydration control from the distribution control stage, improves the dehydration efficiency, and protects the washing machine.

[0104] In some specific embodiments, when the distribution control phase of the washing machine complete machine 2 is controlled to operate with the distribution parameters, the actual speed and acceleration values of the dewatering motor 22 of the washing machine complete machine 2 are measured and output by the speed detection device of the dewatering motor 22 itself and the acceleration sensor provided in the washing machine complete machine 2 itself, which are used to correct the distribution parameters for the second training data.

[0105] The training method of the washing machine AI control model of the embodiment corrects the distribution parameters corresponding to the initial load, the distribution load, and the distribution eccentricity in the second training data by the speed and acceleration measured by the washing machine complete machine 2 itself in the distribution control phase, thereby improving the accuracy of the second training data.

[0106] In some specific embodiments, the eccentricity threshold values corresponding to different distribution loads are obtained by setting multiple standard load blocks to perform distribution control on multiple eccentricities of each initial load; and the eccentricity value that appears with a certain probability of peripheral device parameter failure is the eccentricity threshold value corresponding to the distribution load.

[0107] Referring to Figure 1 , Figure 2 , Figure 3 The application further discloses a training device of a washing machine AI control model, and the AI control model obtained by training is applied to the dewatering process of the washing machine.

[0108] The dewatering process of the washing machine includes a distribution control phase, which is used to control the speed of the inner drum of the washing machine, so that the clothes to be dewatered are distributed in the inner drum to reduce the eccentricity; generally, the speed is accelerated to a first speed and maintained at the first speed for a first time length; the speed is accelerated to a second speed from the first speed with a first acceleration for a second time length, so that the clothes in the inner drum are uniformly distributed.

[0109] The dewatering process of the washing machine further includes an acceleration dewatering judgment phase and an acceleration dewatering control phase; the acceleration dewatering judgment phase is to maintain the second speed for a third time length, and the distribution load and the distribution eccentricity are detected in the acceleration dewatering judgment phase, and whether to enter the acceleration dewatering control phase is judged according to the obtained distribution load and distribution eccentricity; the acceleration dewatering control phase is to operate for a fourth time length with the second acceleration to perform deep dewatering on the clothes in the inner drum.

[0110] The training device of the washing machine AI control model includes multiple washing machine complete machines 2, multiple peripheral detection modules 3, and a model training module 1.

[0111] The washing machine complete machine 2 comprises a first control module 21, a dehydration motor 22 connected with the first control module 21 respectively, a load acquisition module 23, and an eccentricity acquisition module 24; the load acquisition module 23 is used for acquiring initial load, distributed load, and dehydration load and transmitting to the first control module 21; the eccentricity acquisition module 24 is used for acquiring distributed eccentricity and dehydration eccentricity and transmitting to the first control module 21; the washing machine complete machine 2 is loaded with clothes for testing initial load.

[0112] Each peripheral detection module 3 is arranged corresponding to each washing machine complete machine 2 and is connected with each washing machine complete machine 2 one by one, and is used for detecting peripheral parameters when the washing machine complete machine 2 operates and outputting.

[0113] The model training module 1 is a hardware main body with a processor and a memory, is configured with a data training tool, and is in communication connection with each first control module 21 and each peripheral detection module 3 respectively. The model training module 1 is preset with a peripheral parameter fault threshold value and is configured to:

[0114] The peripheral parameters, the dehydration load, and the dehydration eccentricity of the accelerated dehydration control stage are acquired cyclically, and it is judged in real time whether the peripheral parameters reach or exceed the peripheral parameter fault threshold value; if yes, the dehydration load and the dehydration eccentricity are collected as first training data for training the accelerated dehydration AI control model; if no, the accelerated dehydration control is completed, and the corresponding data is discarded.

[0115] The training device of the washing machine AI control model of the present application trains the accelerated dehydration AI control model of the first training data composed of the peripheral parameters obtained by the peripheral detection module 3 and the dehydration load and the dehydration eccentricity obtained by the washing machine complete machine 2 itself when the peripheral parameter fault occurs, and is used for control in the accelerated dehydration control stage; the washing machine complete machine 2 itself data and the external detection data combine to train the AI control model to increase the dimension of the model training data, improve the accuracy and intelligence of the AI control model, protect the washing machine, and prolong the service life of the washing machine; the first training data is automatically acquired through the communication of the washing machine complete machine 2 itself and the peripheral detection module 3 and the model training module 1, and the efficiency of training data acquisition and the model training efficiency are improved.

[0116] The specific structure and principle of the training device of the washing machine AI control model of the present application will be described in detail below through specific embodiments.

[0117] In some specific embodiments, with reference to Figure 1 , Figure 2 , Figure 3 , Figure 4 , Figure 7 , the peripheral parameter fault threshold value comprises a vibration threshold value and a force threshold value.

[0118] The peripheral detection module 3 comprises a second control module 32, a vibration sensor 34 connected with the second control module 32, and four force sensors 33, which transmit the vibration value of the accelerated dehydration control stage and the force vector of the four feet to the second control module 32 in real time; the second control module 32 is in communication connection with the model training module 1 through the second communication module 31, and transmits the obtained vibration value and the four force vectors to the model training module 1.

[0119] The model training module 1 obtains the dehydration load and the dehydration eccentricity of the accelerated dehydration control stage in real time, and obtains the force value of the washing machine 2 from the four force vectors; the vibration value is judged whether it reaches or exceeds the vibration threshold value and / or the force value is judged whether it reaches or exceeds the force threshold value in real time; if yes, the dehydration load and the dehydration eccentricity are collected as the first training data for training the accelerated dehydration AI control model; if no, the accelerated dehydration control is completed, and the corresponding data is discarded.

[0120] The training device of the washing machine AI control model of the application trains the accelerated dehydration AI control model by the first training data composed of the vibration value and the force value of the external detection data obtained by the peripheral detection module 3 and the dehydration load and the dehydration eccentricity obtained by itself, which is used for the control of the accelerated dehydration control stage, and the combination of the self data and the external detection data increases the dimension of the training control model and improves the accuracy and intelligence of the control model, the dehydration success rate and efficiency, the protection of the washing machine and the prolongation of the service life of the washing machine.

[0121] The vibration value exceeding the vibration threshold value is vibration failure; the force value exceeding the force threshold value is movement failure; the movement failure is that the force value of the washing machine includes the lateral force for overcoming the friction force to move laterally.

[0122] In some specific embodiments, with reference to Figure 1 , Figure 2 , Figure 3 , Figure 5 , Figure 8 , the peripheral parameter failure threshold value comprises a noise threshold value and a force threshold value.

[0123] The peripheral detection module 3 comprises a second control module 32, a noise sensor 35 connected with the second control module 32, and four force sensors 33, which transmit the noise value of the accelerated dehydration control stage and the force vector of the four feet to the second control module 32 in real time; the second control module 32 is in communication connection with the model training module 1 through the second communication module 31, and transmits the obtained noise value and the four force vectors to the model training module 1.

[0124] The model training module 1 cyclically obtains the dehydration load and the dehydration eccentricity of the accelerated dehydration control stage, obtains the force value of the whole washing machine 2 from the four force vectors, and judges in real time whether the noise value reaches or exceeds the noise threshold value and / or whether the force value reaches or exceeds the force threshold value; if yes, the dehydration load and the dehydration eccentricity are collected as the first training data for training the accelerated dehydration AI control model; if no, the accelerated dehydration control is completed, and the corresponding data is discarded.

[0125] The training device of the washing machine AI control model of the application trains the accelerated dehydration AI control model of the first training data composed of the noise value and the force value of the external detection data obtained by the external detection module 3 and the dehydration load and the dehydration eccentricity obtained by itself, which is used for the control of the accelerated dehydration control stage. The combination of the self-data and the external detection data for training the control model increases the dimension of the model training data, improves the accuracy and intelligence of the control model, improves the dehydration success rate and efficiency, protects the washing machine, and prolongs the service life of the washing machine.

[0126] The vibration value exceeding the vibration threshold value indicates that the vibration fault occurs; the force value exceeding the force threshold value indicates that the moving fault occurs; and the moving fault is that the force value of the washing machine includes the lateral force for overcoming the friction force to move the washing machine laterally.

[0127] In some specific embodiments, the eccentricity acquisition module 24 is an acceleration sensor installed at the bearing or outer cylinder support structure of the washing machine inner cylinder, which transmits the acceleration signal of the washing machine inner cylinder rotation in real time to the first control module 21; the first control module 21 performs Fourier transform on the acceleration signal to analyze the amplitude change of the fundamental frequency and harmonic components to obtain the eccentricity value. That is, the eccentricity value is measured by the amplitude change value of the fundamental frequency and harmonic components of the acceleration.

[0128] In some specific embodiments, the eccentricity acquisition module 24 is a speed sensor, which obtains the speed data of the dehydration motor 22 in the uniform speed stage and transmits the speed data to the first control module 21 for differential calculation of the speed change rate to obtain the acceleration, and calculates the difference between the maximum acceleration and the minimum acceleration in the continuous interval, which is used for obtaining the eccentricity value in the accelerated dehydration judgment stage.

[0129] In some specific embodiments, the eccentricity acquisition module 24 is a current detection module, which obtains the real-time torque current of the dehydration motor 22, transmits the above-mentioned real-time torque current signal to the first control module 21 after low-pass digital filtering processing to eliminate high-frequency noise interference; the first control module 21 extracts the maximum torque current and the minimum filtered torque current in a plurality of continuous dehydration periods, respectively calculates the average value of each maximum torque current and each minimum torque current, and calculates the difference value to represent the eccentricity value. That is,

[0130] e=K(Imax_avg-Imin_avg )

[0131] wherein,

[0132] e is eccentricity;

[0133] K is a model calibration coefficient to avoid model error;

[0134] Imax_avg is the average of each maximum torque current;

[0135] Imin_avg is the average of each minimum torque current.

[0136] In some specific embodiments, the peripheral device detection module 3 further comprises a test bench; the force sensor 33 is a multi-dimensional force sensor; the four force sensors 33 are respectively fixedly connected with the four feet of the washing machine 2 and the test bench; the force vectors of the four feet measured by the four force sensors 33 are transmitted to the model training module 1; and the model training module 1 obtains the force value of the washing machine 2 according to the force vectors.

[0137] The test bench is four joints with adjustable width and length and locking, which are fixedly connected with the four force sensors 33.

[0138] In some specific embodiments, the training device of the washing machine AI control model further comprises a first communication module 4; the first control module 21 is a driving module of the dehydration motor 22, which is in communication connection with the model training module 1 through the first communication module 4.

[0139] In some specific embodiments, referring to Figure 1 , Figure 2 , Figure 3 , Figure 6 , the model training module 1 is preconfigured with multiple sets of distribution parameters, multiple eccentricity thresholds corresponding to different distribution loads, and is configured to:

[0140] send the distribution parameters to each first control module 21, control each washing machine 2 to operate according to the distribution parameters in the distribution control stage, and when the washing machine 2 enters the accelerated dehydration judgment stage after the end of the distribution control stage, obtain the distribution load and the distribution eccentricity of the washing machine 2;

[0141] compare the distribution eccentricity with the eccentricity threshold corresponding to the distribution load; when the distribution eccentricity does not exceed the eccentricity threshold, control to enter the accelerated dehydration control stage, and collect the corresponding initial load and distribution parameters as the first training data for training the distribution AI control model; when the distribution eccentricity exceeds the eccentricity threshold, the experiment fails, and the corresponding data is discarded.

[0142] The training device of the washing machine AI control model of the application obtains the second training data composed of the initial load and the distribution parameter through the washing machine itself, trains the distribution AI control model for the operation control of the dehydration motor 22 in the distribution control stage, improves the accuracy and intelligence of the control model, improves the dehydration success rate and efficiency, and improves the user experience.

[0143] In some specific embodiments, the accelerated dehydration control model is a first fitting function of the dehydration load and the dehydration eccentricity obtained by training the first training data, and whether the peripheral parameter fault occurs is intelligently judged by judging whether the dehydration load and the dehydration eccentricity measured by the washing machine meet the first fitting function, and the dehydration is interrupted when the peripheral parameter fault is judged; otherwise, the accelerated dehydration control is continued.

[0144] The training device of the washing machine AI control model of the embodiment obtains the first fitting function by fitting the first training data, so that the washing machine can intelligently judge whether the peripheral parameter fault occurs or is about to occur according to the dehydration load and the dehydration eccentricity, improve the intelligence and accuracy of the control, protect the washing machine, and prolong the service life of the washing machine.

[0145] In some specific embodiments, the distribution control model is a second fitting function of the initial load and the distribution parameter obtained by the second training data, so that the washing machine intelligently configures different distribution parameters according to different initial loads and the second fitting function measured by the washing machine to control the operation of the washing machine in the distribution control stage, so that the distribution eccentricity after the operation in the distribution control stage meets the requirement of entering the accelerated dehydration control stage, and the dehydration rate is improved.

[0146] The training device of the washing machine AI control model of the embodiment obtains the second fitting function by fitting the second training data, so that the washing machine can intelligently configure the distribution parameter according to the initial load, and intelligently configure the distribution parameter according to the initial load not only to make the distribution eccentricity meet the dehydration requirement, but also to reduce the probability of vibration failure and movement failure when the distribution load and the distribution eccentricity obtained in the distribution control stage are used for accelerated dehydration control, improve the dehydration rate, and improve the user experience.

[0147] The application also discloses a washing machine, which comprises a first control module 21, a dehydration motor 22, a load acquisition module 23 and an eccentricity acquisition module 24 connected with the first control module 21 respectively; the first control module 21 is configured with the accelerated dehydration AI control model obtained in the above-mentioned embodiments, is used for judging whether the peripheral parameter fault of the washing machine occurs or is about to occur according to the dehydration load cyclically acquired by the load acquisition module 23 and the dehydration eccentricity cyclically acquired by the eccentricity acquisition module 24, and if yes, controls the stop operation of the dehydration motor 22 and interrupts the dehydration; if no, controls the continuous operation of the dehydration motor 22.

[0148] The laundry washing machine of the present application improves the intelligence of the operation of the laundry washing machine, protects the laundry washing machine and prolongs the service life of the laundry washing machine by configuring the accelerated dewatering AI control model obtained by the above AI training method and AI training device to control the operation of the accelerated dewatering control stage.

[0149] The specific structure and control principle of the laundry washing machine of the present application are described in detail below through specific embodiments.

[0150] In some specific embodiments, the first control module 21 is configured with the distribution AI control model obtained in the above training method or training device embodiments, is configured with the distribution parameters according to the initial load obtained by the load obtaining module 23, controls the operation of the distribution control stage, and judges whether to enter the accelerated dewatering control stage according to the distribution load obtained by the load obtaining module 23 and the distribution eccentricity obtained by the eccentricity obtaining module 24 in the accelerated dewatering judgment stage.

[0151] This embodiment improves the success rate of dewatering by intelligently configuring the distribution parameters according to the initial load.

[0152] In some specific embodiments, the load obtaining module 23 is a pressure sensor installed at the bottom of the laundry washing machine or the support point of the suspension system, which calculates the load by measuring the pressure value.

[0153] In some specific embodiments, the load obtaining module 23 is a load sensor placed at the connection between the inner drum and the rack, which measures the load value in real time.

[0154] In some specific embodiments, the eccentricity obtaining module 24 is an acceleration sensor arranged at the bearing of the inner drum of the laundry washing machine or the support structure of the outer drum; the acceleration signal of the rotation of the inner drum of the laundry washing machine is collected in real time and transmitted to the first control module 21; the first control module 21 performs Fourier transform on the acceleration signal, analyzes the amplitude change of the fundamental frequency and harmonic components, and obtains the eccentricity value.

[0155] In some specific embodiments, the eccentricity obtaining module 24 is a speed sensor, which obtains the speed data of the dewatering motor 22 in the uniform speed stage, transmits the speed data to the first control module 21 for differential calculation of the speed change rate to obtain the acceleration, calculates the difference between the maximum acceleration and the minimum acceleration in the continuous interval, and is used for obtaining the eccentricity value in the accelerated dewatering judgment stage.

[0156] In some specific embodiments,

[0157] The above real-time torque current signal is transmitted to the first control module 21 after low-pass digital filtering processing to eliminate high-frequency noise interference; the first control module 21 extracts the maximum torque current and the minimum filtered torque current in a plurality of continuous dewatering periods, respectively calculates the average value of each maximum torque current and each minimum torque current, and calculates the difference value to represent the eccentricity value. That is,

[0158] e = K (Imax_avg - Imin_avg )

[0159] wherein,

[0160] e is eccentricity;

[0161] K is a model calibration coefficient, avoiding model error;

[0162] Imax_avg is the average of each maximum torque current;

[0163] Imin_avg is the average of each minimum torque current.

[0164] The above examples are only used to illustrate the technical solutions of the present application, but not to limit it; although the present application has been described in detail with reference to the foregoing examples, the technical solutions recorded in the foregoing examples can still be modified, or some technical features can be replaced equivalently, by those skilled in the art; and these modifications or replacements do not make the essence of the corresponding technical solutions deviate from the spirit and scope of the technical solutions claimed by the present application.

Claims

1. A training method for an AI control model of a washing machine, characterized in that, Comprise: S1, set the model training module, preset peripheral parameter fault threshold; S2, set a plurality of experimental washing machine whole machine, it includes first control module and with the first control module connection dehydration motor, load acquisition module, eccentricity acquisition module, for obtaining load and eccentricity; The first control module and the model training module are connected in communication; S3, set a plurality of peripheral detection modules, respectively with each washing machine whole machine connection, for obtaining peripheral parameters, with the model training module communication connection, to the model training module output the peripheral parameters; S4, in the washing machine whole machine, place the initial load of clothes to carry out dehydration; Cycle to obtain the dehydration load of acceleration dehydration control stage, dehydration eccentricity, the peripheral parameter; S5, judge whether the peripheral parameter reaches or exceeds the peripheral parameter fault threshold; If yes, execute S6; if no, execute S4; S6; the peripheral parameter, corresponding dehydration load, dehydration eccentricity is collected as first training data, for training acceleration dehydration AI control model; The peripheral detection module includes second control module and vibration sensor, force sensor connected with the second control module; The vibration sensor, the force sensor are used for detecting the vibration value, the force value of the washing machine whole machine transmission to the second control module; The second control module and the model training module are connected in communication, and the vibration value, the force value obtained are transmitted to the model training module; The model training module identifies it as the peripheral parameter; The peripheral parameter fault threshold includes vibration threshold, force threshold; Whether the peripheral parameter reaches or exceeds the peripheral parameter fault threshold is that the vibration value reaches or exceeds the vibration threshold and / or the force value reaches or exceeds the force threshold; Or, the peripheral detection module includes second control module and noise sensor, force sensor connected with the second control module; The noise sensor, the force sensor are used for detecting the noise value, the force value of the washing machine whole machine transmission to the second control module; The second control module and the model training module are connected in communication, and the noise value, the force value obtained are transmitted to the model training module; The model training module identifies it as the peripheral parameter; The peripheral parameter fault threshold includes noise threshold, force threshold; Whether the peripheral parameter reaches or exceeds the peripheral parameter fault threshold is that the noise value reaches or exceeds the noise threshold and / or the force value reaches or exceeds the force threshold. 2.The training method of a washing machine AI control model according to claim 1, characterized in that, Further comprise: S1', in the model training module, preset a plurality of different distribution parameters, it includes first rotating speed, first acceleration; Prewritten a plurality of different distribution load corresponding distribution eccentricity threshold; S2', the model training module sends the distribution parameter to the first control module, controls each washing machine whole machine operation distribution control stage; When the distribution control stage is completed, the distribution load of the washing machine whole machine, the distribution eccentricity is obtained; S3', determining whether the distribution eccentricity reaches or exceeds the eccentricity threshold corresponding to the distribution load; if yes, performing S4'; if no, performing S5'; S4', terminating the dehydration control; S5', obtaining the initial load, the corresponding distribution parameter, the distribution load, the distribution eccentricity and collecting them as second training data for training the distribution AI control model. 3.A training device of a washing machine AI control model, characterized by, Comprise: A plurality of washing machine complete machines for loading initial load of clothes, comprising a first control module, a dehydration motor connected thereto respectively, a load acquisition module, an eccentricity acquisition module; A plurality of external detection modules respectively connected with each of the washing machine complete machines for detecting external parameters of the washing machine complete machines during operation; A model training module which is a hardware main body with a processor and a memory, is in communication connection with each of the first control modules and each of the external detection modules, is configured with a data training tool, is preconfigured with an external parameter fault threshold, and is configured to obtain the external parameter, the dehydration load and the dehydration eccentricity in real time during the operation of the acceleration dehydration control stage of each of the washing machine complete machines; determining whether the external parameter reaches the external parameter fault threshold; If yes, collecting the dehydration load and the dehydration eccentricity as first training data for training the acceleration dehydration AI control model; The external detection module comprises a second control module and a vibration sensor and four force sensors connected with the second control module; the vibration sensor and each of the force sensors are respectively used for detecting the vibration value of the washing machine complete machine and the force vector of each foot and transmitting to the second control module; the second control module is in communication connection with the model training module and transmits the vibration value and each of the force vectors to the model training module; the model training module obtains the force value of the washing machine complete machine from each of the force vectors and identifies the vibration value and the force value as the external parameter; the external parameter fault threshold comprises a vibration threshold and a force threshold; determining whether the vibration value reaches or exceeds the vibration threshold and / or the force value reaches the force threshold; if yes, collecting the dehydration load and the dehydration eccentricity as the first training data for training the acceleration dehydration AI control model; Or, the external detection module comprises a second control module and a noise sensor and four force sensors connected with the second control module; the noise sensor and each of the force sensors are used for detecting the noise value and the force vector of each foot and transmitting to the second control module; The second control module is in communication connection with the model training module and transmits the noise value and each of the force vectors to the model training module; the model training module obtains the force value of the washing machine complete machine from each of the force vectors and identifies the noise value and the force value as the external parameter; the external parameter fault threshold comprises a noise threshold and a force threshold; Determining whether the noise value reaches or exceeds the noise threshold and / or the force value reaches the force threshold; If yes, collect the dehydration load, the dehydration eccentricity as the first training data for training the accelerated dehydration AI control model. 4.The training device of the washing machine AI control model of claim 3, wherein, The model training module is further configured to preset a plurality of sets of distribution parameters and eccentricity thresholds corresponding to different distribution loads, and to: send the distribution parameters to each first control module to control each washing machine to operate according to the distribution parameters in a distribution control stage, and return the distribution load and distribution eccentricity measured after the distribution control stage ends; determine whether the distribution eccentricity reaches or exceeds the eccentricity threshold corresponding to the distribution load; if yes, stop the dehydration control; and if no, control to enter the accelerated dehydration control stage, and collect the initial load, the distribution parameters, the distribution load, and the distribution eccentricity as second training data for training a distribution AI control model.

5. A laundry machine characterized by The first control module is configured with an accelerated dehydration AI control model obtained by the training device of the washing machine AI control model according to claim 3 or 4, which is used to detect or predict whether a dehydration fault occurs according to the real-time obtained dehydration load and dehydration eccentricity; and if yes, control the dehydration to be interrupted.

6. A laundry washing machine according to Claim 5, characterized in that The first control module is further configured with a distribution AI control model, which is used to control the operation of a distribution control stage according to an initial load using corresponding distribution parameters, so that the distribution eccentricity after the operation of the distribution control stage meets the requirement of entering the accelerated dehydration stage for operation.

Images