Washing machine, control method and control device thereof and storage medium

Abstract

The embodiment of the invention provides a washing machine, a control method and device of the washing machine and a storage medium, the control method of the washing machine comprises the steps that clothes information of clothes to be washed in the washing machine and temperature and humidity information in the washing machine are obtained, and the clothes information comprises materials, stain types and weight; predicting washing parameters according to the clothes information and the temperature and humidity information; and controlling the washing machine to operate based on the washing parameters. The washing parameters are predicted according to the clothes information and the temperature and humidity information, that is, the washing parameters can be dynamically adjusted, compared with existing fixed washing parameters, changes of the washing environment can be dynamically responded, the flexibility of the washing machine is improved, and then the energy utilization efficiency is improved.

Description

Technical Field

[0001] This application belongs to the technical field of washing machines, and particularly relates to a washing machine, a control method thereof, a control device and a storage medium. Background Art

[0002] In traditional washing machine designs, washing parameters such as water level, rotation speed, and washing time are usually set based on preset programs. These programs are often determined according to extensive test results and experience summaries during the product development stage, aiming to cover the usage scenarios of most users.

[0003] However, due to the lack of flexibility, the washing machine cannot dynamically respond to changes in the washing environment, resulting in low energy utilization efficiency. Summary of the Invention

[0004] This application provides a washing machine, a control method thereof, a control device and a storage medium to improve the flexibility of the washing machine, enabling it to dynamically respond to changes in the washing environment, thereby improving energy utilization efficiency.

[0005] In a first aspect, an embodiment of this application provides a control method for a washing machine, including:

[0006] Obtain the clothing information of the laundry to be washed in the washing machine and the temperature and humidity information in the washing machine. The clothing information includes material, stain type, and weight;

[0007] Predict washing parameters based on the clothing information and the temperature and humidity information;

[0008] Control the operation of the washing machine based on the washing parameters.

[0009] Optionally, the predicting washing parameters based on the clothing information and the temperature and humidity information includes:

[0010] Obtain a trained neural network model;

[0011] Predict washing parameters based on the clothing information and the temperature and humidity information according to the neural network model.

[0012] Optionally, the predicting washing parameters based on the clothing information and the temperature and humidity information according to the neural network model includes:

[0013] Input the clothing information and the temperature and humidity information into the neural network model. The neural network model includes a matching sub-network and a prediction sub-network;

[0014] Apply the matching sub-network to match the clothing information and the temperature and humidity information with the washing types to obtain a matching result;

[0015] Use the prediction sub-network to predict the washing parameters for the matching result.

[0016] Optionally, the obtaining of the trained neural network model includes:

[0017] Establish a convolutional neural network model and initialize the convolutional neural network model;

[0018] Obtain a plurality of clothing samples and a plurality of temperature and humidity samples, preprocess the plurality of clothing samples and the plurality of temperature and humidity samples to obtain a training set;

[0019] Iteratively train the convolutional neural network model according to the training set to obtain a trained neural network model.

[0020] Optionally, the iteratively training the convolutional neural network model according to the training set to obtain a trained neural network model includes:

[0021] Obtain the optimal washing parameters corresponding to each test sample obtained through washing tests;

[0022] Use each optimal washing parameter as label data, and make the training set iteratively train the convolutional neural network model with reference to the label data to obtain a trained neural network model.

[0023] Optionally, the obtaining of the clothing information of the laundry to be washed in the washing machine and the temperature and humidity information in the washing machine, where the clothing information includes material, stain type and weight, includes:

[0024] Detect the initial temperature and humidity information in the washing machine;

[0025] Compare the initial temperature and humidity information with a preset temperature and humidity threshold;

[0026] When the initial temperature and humidity information is within the temperature and humidity threshold, use the initial temperature and humidity information as the temperature and humidity information in the washing machine.

[0027] Optionally, the comparing the initial temperature and humidity information with a preset temperature and humidity threshold includes:

[0028] When the initial temperature and humidity information exceeds the temperature and humidity threshold, send an abnormal reminder, and adjust the initial temperature and humidity information until the adjusted initial temperature and humidity information is within the temperature and humidity threshold, and use the adjusted initial temperature and humidity information as the temperature and humidity information in the washing machine.

[0029] In a second aspect, an embodiment of the present application further provides a control device for a washing machine, including:

[0030] An acquisition unit, configured to acquire the clothing information of the laundry to be washed in the washing machine and the temperature and humidity information in the washing machine, where the clothing information includes material, stain type, and weight;

[0031] A prediction unit, configured to predict washing parameters according to the clothing information and the temperature and humidity information;

[0032] A control unit, configured to control the operation of the washing machine based on the washing parameters.

[0033] In a third aspect, an embodiment of the present application further provides a computer-readable storage medium, on which a computer program is stored. When the computer program is executed on a computer, the computer is caused to execute the control method of the washing machine as described in any one of the above.

[0034] In a fourth aspect, an embodiment of the present application further provides a washing machine, including:

[0035] An image sensor, configured to detect the material and stain type of the laundry to be washed;

[0036] A weight sensor, configured to detect the weight of the laundry to be washed;

[0037] A temperature and humidity sensor, configured to detect the temperature and humidity information in the washing machine;

[0038] A processor, electrically connected to the image sensor, the weight sensor, and the temperature and humidity sensor respectively. The processor is configured to execute the control method of the washing machine as described in any one of the above.

[0039] In the washing machine, its control method, control device, and storage medium according to the embodiments of the present application, the washing parameters are predicted according to the clothing information and the temperature and humidity information, that is, the washing parameters can be dynamically adjusted. Compared with the existing fixed washing parameters, it can dynamically respond to the changes in the washing environment, improve the flexibility of the washing machine, and further improve the energy utilization efficiency. Description of the Drawings

[0040] In order to more clearly illustrate the technical solutions in the embodiments of the present application, the following will briefly introduce the drawings required for the description of the embodiments. Obviously, the following drawings are only some embodiments of the present application. For those skilled in the art, other drawings can be obtained based on these drawings without creative efforts.

[0041] In order to more completely understand the present application and its beneficial effects, the following will be described in conjunction with the drawings. Among them, the same reference numerals represent the same parts in the following description.

[0042] Figure 1 It is a first flowchart of the control method of the washing machine provided by the embodiment of the present application.

[0043] Figure 2 It is a second process schematic diagram of the control method for the washing machine provided by the embodiment of the present application.

[0044] Figure 3 It is a circuit diagram for detecting the temperature, humidity and motor fault of the washing machine provided by the embodiment of the present application.

[0045] Figure 4 It is a structural block diagram of the control device for the washing machine provided by the embodiment of the present application.

[0046] Figure 5 It is a structural block diagram of the washing machine provided by the embodiment of the present application. Detailed implementation manners

[0047] Next, the technical solutions in the embodiments of the present application will be clearly and completely described in conjunction with the accompanying drawings in the embodiments of the present application. Obviously, the described embodiments are only a part of the embodiments of the present application, rather than all the embodiments. Based on the embodiments in the present application, all other embodiments obtained by those skilled in the art without creative efforts shall fall within the protection scope of the present application.

[0048] In the design of traditional washing machines, washing parameters such as water level, rotation speed and washing time are usually set based on preset programs, which are often determined according to extensive test results and experience summaries during the product development stage, aiming to cover the usage scenarios of most users. Manufacturers usually set a series of fixed parameter combinations for different washing modes in order to meet the basic cleaning needs of different materials of clothes. Among them, the washing modes are such as cotton fabrics, synthetic fibers, gentle wash, etc.

[0049] However, the traditional method has deficiencies. First of all, due to the lack of flexibility, traditional washing machines cannot dynamically respond to changes in the washing environment, resulting in a decrease in energy utilization efficiency; secondly, traditional washing machines do not fully consider the material and stain type of clothes, which may cause damage to some fabrics or the problem of incomplete cleaning.

[0050] With the development of smart home technology, users' demand for the intelligence and personalization of washing machines is increasing day by day. Although the existing washing machine control technology can meet the basic washing needs to a certain extent, there are still many deficiencies when dealing with complex washing scenarios and diverse user needs. For example, adjusting washing parameters only relying on single factors such as temperature and humidity or the quantity and material of clothes cannot comprehensively consider the actual situation of the clothes and the changes in the washing environment; the training process of neural networks is not optimized enough, resulting in limited accuracy and adaptability of the model; there is a lack of effective mechanisms in multi-sensor fusion and collaborative control, making it difficult to achieve efficient washing control; the fault diagnosis and repair capabilities are limited and cannot handle various abnormal situations in a timely manner; energy-saving optimization and safety protection measures need to be strengthened and cannot fully meet users' needs for energy conservation and data security.

[0051] Based on the above problems, the embodiments of the present application provide a washing machine, its control method, control device and storage medium, which will be described below with reference to the accompanying drawings.

[0052] Please refer to Figure 1 , Figure 1 , which is the first flow diagram of the control method of the washing machine provided by the embodiments of the present application. From the perspective of the structure of the washing machine, the type of the washing machine can be a drum washing machine or a pulsator washing machine; from the perspective of the functions of the washing machine, the washing machine can be a washing machine with only washing function, or a washing-drying integrated machine with washing and drying functions. The washing machine can have different washing modes, such as strong washing, gentle washing, wool washing, etc. The control method of the washing machine in the embodiments of the present application can be applied to different washing modes of the washing machine. By dynamically adjusting the washing parameters, the accuracy of clothing treatment can be improved, thereby improving the clothing cleanliness rate and reducing damage to the clothing. Of course, the control method of the washing machine in the embodiments of the present application can also be used as a separate washing program for users to select, which can improve the intelligence level of the washing machine and thus improve the user experience.

[0053] Exemplarily, the control method of the washing machine in the embodiments of the present application includes:

[0054] Step S110, obtain the clothing information of the clothes to be washed in the washing machine and the temperature and humidity information in the washing machine. The clothing information includes the material, stain type and weight.

[0055] The washing machine is mainly used to clean the clothes to be washed. With the development of technology, users require the washing machine to be able to wash the clothes clean and do not want to damage the clothes.

[0056] Based on this, it is necessary to obtain the clothing information of the laundry to be washed, and the clothing information includes material, stain type and weight. Clothing materials such as cotton, linen, silk, wool, chemical fiber, etc. Among them, cotton materials are relatively soft and easy to deform. Therefore, washing strategies such as gentle washing are set. Gentle washing can be set at a relatively low rotation speed, for example. Stain types such as oil stains, blood stains, ink stains, mud stains, etc. According to different stain types, the required cleaning degree is different. For example, general washing methods can be used for mud stains. The weight of the clothing is mainly related to the water intake in the washing parameters and also determines the amount of detergent used. Therefore, obtaining the material, stain type and weight of the clothing can make the setting of the washing parameters more in line with the actual situation.

[0057] Exemplarily, the material information and stain type of the clothing can be obtained by acquiring images. For example, multiple image sensors can be set at the entrance of the inner drum of the washing machine. By collecting the images of the laundry to be washed and extracting and analyzing the feature information in the images, the material information and stain type of the clothing can be obtained. For example, when there are multiple materials in the laundry to be washed, multiple feature information can be extracted from the image, and the proportion of each material can be analyzed. The clothing material information can be comprehensively determined according to the material type and material proportion. The identification of the stain type can also refer to the above-mentioned identification of the material, which will not be elaborated here.

[0058] Exemplarily, the weight of the clothing can be obtained by a weight sensor, that is, achieved by weighing.

[0059] During the operation of the washing machine, temperature and humidity will also affect the washing parameters. For example, the temperature inside the inner drum of the washing machine will affect the water heating time, drying duration and the use effect of the detergent; for example, the humidity inside the inner drum of the washing machine will affect the operation duration of the dehydration program. It should be noted that the detergent can better exert its effect at a certain temperature. Therefore, in some implementation manners, the washing water will be heated. When the temperature in the inner drum is relatively high, the water heating duration will be shortened. The humidity inside the inner drum combined with the weight of the clothing can represent the water content of the clothing, and the water content of the clothing will affect the operation duration of the dehydration program.

[0060] It should be noted that both the clothing information and the temperature and humidity inside the inner drum will affect the setting of the washing parameters. Moreover, the clothing information and the temperature and humidity information inside the inner drum play important roles in different stages of the laundry process respectively. Therefore, before predicting the washing parameters in the embodiments of the present application, it is necessary to obtain the clothing information and the temperature and humidity information inside the inner drum.

[0061] Step S120, predict the washing parameters according to the clothing information and the temperature and humidity information.

[0062] Washing parameters such as water intake, motor speed, washing time, dehydration time, water heating time, etc. can predict the washing parameters based on clothing information and temperature and humidity information, so as to match a suitable washing method for the laundry to be washed, rather than using a fixed washing program and fixed washing parameters for washing. This can not only improve the cleaning degree of the clothing, but also reduce the damage to the clothing and reduce the waste of washing machine resources.

[0063] For example, when the clothing material is cotton, the stain degree is oil stain, the clothing weight is 2 kg, the temperature is 30 °C, and the humidity is 12% rh, predict washing parameters such as the water intake is 16 L, the motor speed such as the maximum motor speed is 1400 r / min, the washing time is 30 minutes, the dehydration time is 5 minutes, the water heating time is 3 minutes, etc.

[0064] Of course, there can be other combination methods for clothing information, temperature and humidity information, and the corresponding washing parameters. The number of washing parameters is not limited, and the deformation of other embodiments can be referred to the above description.

[0065] Step S130, control the washing machine to run based on the washing parameters.

[0066] After determining the washing parameters, the washing machine can be controlled to run according to the washing parameters. Among them, if during the operation of the washing machine, the current washing parameters can be adjusted according to the washing parameters; if before the operation of the washing machine, the washing can be directly carried out according to the washing parameters.

[0067] In the control method of the washing machine according to the embodiment of the present application, the washing parameters are predicted according to the clothing information and temperature and humidity information, that is, the washing parameters can be dynamically adjusted. Compared with the existing fixed washing parameters, it can dynamically respond to the changes in the washing environment, improve the flexibility of the washing machine, and thus improve the energy utilization efficiency.

[0068] Among them, for the prediction of the washing parameters, the preset rules can be referred to, or a trained model can be used to facilitate the timely update of the washing parameters, realize the dynamic adjustment and optimization of the washing machine washing, and improve the resource utilization efficiency. Please refer to Figure 2 as shown Figure 2 This is the second process schematic diagram of the control method of the washing machine provided by the embodiment of the present application. In one implementation, step S120, predict the washing parameters according to the clothing information and temperature and humidity information, includes:

[0069] Step S121, obtain a trained neural network model.

[0070] Among them, the neural network model can be a convolutional neural network model. The convolutional neural network model has the characteristic of transfer learning and has an accurate recognition ability for local features of images, which can shorten the learning time of the neural network model.

[0071] In this embodiment, before the washing machine executes the washing program, training samples are obtained, and the neural network model is iteratively trained using these training samples to obtain neural network parameters that satisfy the corresponding relationship between clothing information, temperature and humidity information, and washing parameters, thereby obtaining a trained neural network model. Based on the clothing information and temperature and humidity information, the washing parameters are predicted according to the neural network model.

[0072] The computing platform is deployed with a pre-trained convolutional neural network. The convolutional neural network model is built based on the TensorFlow Lite framework and written in Python. The training database of the convolutional neural network model consists of a large number of washing samples under different materials, stain types, and initial temperature and humidity conditions. Each record includes clothing types, images before and after washing, and temperature and humidity change curve information. The weight parameters are optimized through the backpropagation algorithm to finally achieve the prediction of the optimal combination of washing parameters, including the setting of water intake, the adjustment of motor speed, and the planning of washing time.

[0073] During the training process of the convolutional neural network model, the Keras library is used for model construction, and the training process is accelerated on the NVIDIA GPU to improve the training efficiency and accuracy. During the training process, a large number of actual washing cases are collected to build a comprehensive training database to ensure that the model can adapt to various different washing scenarios. The model needs to regularly update the training database, combine the latest washing technologies and user feedback, and continuously optimize the convolutional neural network model to improve the washing effect.

[0074] After training is completed, the convolutional neural network model is converted into the TensorFlow Lite format and deployed to the embedded computing platform in the washing machine to achieve real-time inference.

[0075] The expert module is developed using the Prolog language. The rule library covers a variety of washing modes, including gentle wash, standard wash, and strong wash. Each mode corresponds to a specific operation instruction sequence. The rule library of this module is regularly updated, and in combination with the results output by the convolutional neural network model, the washing strategy is continuously optimized.

[0076] Step S122: Input the clothing information and temperature and humidity information into the neural network model. The neural network model includes a matching sub-network and a prediction sub-network.

[0077] Among them, the matching sub-network is used to match the clothing information and the temperature and humidity information with the washing types, and the prediction sub-network is used to predict the washing parameters according to the matching results of the matching sub-network. Exemplarily, the washing types are such as gentle wash in the first gear, gentle wash in the second gear, standard wash, strong wash in the first gear, strong wash in the second gear, strong wash in the third gear, etc. According to different clothing information and temperature and humidity information, the washing types are matched, and there are corresponding combinations of washing parameters for each washing type.

[0078] Step S123: Apply the matching sub-network to match the clothing information and the temperature and humidity information with the washing types to obtain a matching result.

[0079] A mapping table or a mapping relationship curve of the clothing information, the temperature and humidity information, and the washing types can be established. The matching sub-network can perform feature extraction and matching after feature extraction. The training process of the matching sub-network is also the process of establishing the mapping table or the process of fitting the mapping relationship curve.

[0080] Step S124: Apply the prediction sub-network to predict the washing parameters for the matching result.

[0081] Each washing type corresponds to a combination of washing parameters. For example, for the washing type of gentle wash in the first gear, the washing parameters can be a water inlet volume of 16L, a motor speed of 1400r / min, and a washing time of 30min. For the correspondence between each washing type and the washing parameters, it can be fitted through the prediction sub-network. By using the labeled data to mark the optimal washing parameters corresponding to the washing types as the reference for the fitting result, the prediction of the washing parameters is realized based on the matching result.

[0082] In the control method of the washing machine provided by the embodiment of the present application, a neural network model is used to predict the washing parameters. The neural network model can continuously perform iteration and update of the washing parameter prediction, improve the matching degree between the washing parameters and the laundry to be washed, and further improve the flexibility of washing and the utilization efficiency of resources.

[0083] Among them, for the acquisition of the trained neural network model, it can be obtained by establishing a training set and performing training. In one implementation, step S121: Acquire a trained neural network model, including:

[0084] Step S1210: Establish a convolutional neural network model and initialize the convolutional neural network model.

[0085] For example, a lightweight convolutional neural network, such as MobileNetV2, is used to adapt to the resource limitations of the embedded computing platform of the washing machine. Combined with the characteristics of the washing parameter prediction task, a global average pooling layer and a fully connected layer are added to MobileNetV2, and the network output dimension is adjusted to the number of predicted washing parameters, such as predicting three washing parameters, such as water intake, motor speed, and washing time.

[0086] According to the hardware resources and data characteristics, adjust the parameters of the convolution layer such as the convolution kernel size, step size, padding method, and number of channels; select the convolution kernel size of 3×3, the step size of 1 or 2, and the padding method of the same to balance the amount of calculation and feature extraction capabilities; select ReLU6 as the activation function to improve the nonlinear expression ability of the model.

[0087] Use Python language and Keras library to build the model, and stack each layer in order through Sequential model container; for example, first add a convolution layer with activation function; relu6, input shape; (224; 224; 3), then add a maximum pooling layer, add other convolution layers and pooling layers in sequence, and finally add a global average pooling layer and a fully connected layer. The output layer predicts 3 washing parameters.

[0088] Step S1212: obtain multiple clothing samples and multiple temperature and humidity samples, pre-process the multiple clothing samples and multiple temperature and humidity samples, and obtain a training set.

[0089] We have collaborated with home appliance manufacturers, scientific research institutions and professional washing laboratories to build a washing sample database, also known as a training set. We have collected data extensively through various methods, such as actual washing experiments, user usage data feedback and simulation of different washing scenarios. We have simulated various types of stains, such as oil stains, blood stains, ink stains, mud stains, etc., for clothes made of different materials, including common cotton, linen, silk, wool, chemical fiber, etc. We have conducted washing tests under different temperature and humidity environments, and recorded the type of clothing, degree of stains, initial temperature and humidity, images of clothing before and after washing, temperature and humidity change curves during washing, and the corresponding optimal washing parameters, i.e. water intake, motor speed, and washing time.

[0090] Use data cleaning algorithms to remove duplicate records, erroneous data, and incomplete data in the data set; for temperature and humidity data, set a reasonable range of values ​​and remove abnormal values ​​that exceed the range; for example, based on actual conditions, determine that the reasonable range of temperature is [X1, X2] degrees Celsius, and the reasonable range of humidity is [Y1, Y2]%; if the detected temperature and humidity data exceeds this range, it will be regarded as an abnormal value and removed; for image data, check the integrity and clarity of the image and remove blurred or damaged images; the clarity of the image can be judged by image clarity evaluation algorithms, such as calculating the gradient amplitude of the image.

[0091] Organize professional personnel to label the collected data, clarify the best washing parameters corresponding to each washing sample, and provide accurate label data for model training; the labeling process needs to be carried out strictly in accordance with unified standards to ensure the accuracy and consistency of labeling.

[0092] Use techniques such as image rotation, flipping, cropping, and brightness adjustment to enhance the image data, increase data diversity, and improve the generalization ability of the model; perform operations such as adding random noise and data interpolation on the temperature and humidity data to expand the data volume; for example, a certain range of random noise can be added to the temperature and humidity data to simulate measurement errors and uncertainties in the actual environment.

[0093] Step S1214: Iteratively train the convolutional neural network model based on the training set to obtain the trained neural network model.

[0094] Divide the preprocessed data according to the ratio of 70% training set, 20% validation set, and 10% test set; ensure that the data under different materials, stain types, weights, and temperature and humidity conditions are evenly distributed on each subset to avoid bias in model training; a random division method can be used, but it is necessary to ensure that each subset contains various types of data.

[0095] Use the mean squared error (MSE) as the loss function to measure the difference between the model prediction value and the true value; select Adam as the optimizer, set the learning rate to 0.001, β1 = 0.9, β2 = 0.999, epsilon = 1e-07 to adaptively adjust the learning rate and improve the training efficiency and stability.

[0096] Perform training acceleration on the NVIDIA GPU, and use the CUDA toolkit and cuDNN library to achieve GPU accelerated computing; set the batch size (batchSize) of training to 32 and the number of training epochs (epochs) to 100; in each round of training, the model performs forward propagation and backward propagation calculations on the training set to update the weight parameters of the model.

[0097] After training is completed, convert the Keras model to the TensorFlow Lite format; use the TensorFlow LiteConverter tool to configure the conversion parameters, such as quantization method, target hardware, etc.; for the embedded platform of the washing machine, select dynamic range quantization (Dynamic Range Quantization) to quantize the weights and activation values in the model to 8-bit integers to reduce the model size and computational volume while maintaining a certain accuracy.

[0098] Transfer the converted TensorFlow Lite model file to the main control unit (STM32F103C8T6) of the washing machine; by writing corresponding code on the main control unit to call the TensorFlow Lite Micro library, load and infer the model; in the initialization stage, load the model into memory; during runtime, preprocess the real-time collected temperature and humidity data and clothing image data and input them into the model to obtain the predicted washing parameters.

[0099] Use the test set to evaluate the trained model, and calculate metrics such as the mean squared error (MSE), mean absolute error (MAE), and coefficient of determination (R 2 ) between the predicted results and the real washing parameters; MSE reflects the average squared error between the predicted value and the real value, MAE measures the average of the absolute errors between the predicted value and the real value, and R 2 is used to evaluate the goodness of fit of the model to the data.

[0100] Adjust the model structure: Try to increase or decrease the number of convolutional layers and fully connected layers, adjust parameters such as the convolutional kernel size and stride, retrain the model, and observe the changes in model performance; Adjust the data augmentation strategy: Further increase the ways and intensity of data augmentation, such as adding more image transformation operations, or adjusting the amplitude of temperature and humidity data augmentation, expand the dataset, and improve the generalization ability of the model; Hyperparameter tuning: Tune hyperparameters such as the learning rate, batch size, and number of training epochs of the optimizer, and find the optimal hyperparameter combination through methods such as grid search or random search.

[0101] Regularly collect new washing sample data, including the washing effect data feedback by users and the experimental data after applying new washing technologies; Clean, label, and preprocess the new data, and merge it with the original training data.

[0102] Use the merged dataset to re-divide the training set, validation set, and test set, and retrain the model according to the above model training process to update the weight parameters of the model.

[0103] Evaluate the retrained model and compare the performance metrics of the new model with the original model; If the performance of the new model has improved, convert it to the TensorFlow Lite format and deploy it to the washing machine to replace the original model; If the performance has not improved significantly or has decreased, analyze the reasons and retrain after adjusting the training strategy.

[0104] Among them, the best washing parameters can be set as the labeled data, or in other words, iterative training is carried out with the best washing parameters as a reference. In one implementation manner, step S1214, iteratively train the convolutional neural network model according to the training set to obtain the trained neural network model, including:

[0105] Step S1214a: Obtain the optimal washing parameters corresponding to each test sample obtained through washing tests.

[0106] Step S1214b: Use each optimal washing parameter as labeled data, and enable the training set to iteratively train the convolutional neural network model with reference to the labeled data to obtain a trained neural network model.

[0107] Regarding Step S1214a and Step S1214b:

[0108] Organize professionals to annotate the collected data, clarify the optimal washing parameters corresponding to each washing sample, and provide accurate labeled data for model training; the annotation process needs to be carried out strictly according to unified standards to ensure the accuracy and consistency of the annotation.

[0109] Try to increase or decrease the number of convolutional layers and fully connected layers, adjust parameters such as the convolutional kernel size and stride, retrain the model, and observe the changes in model performance; Data augmentation strategy adjustment: Further increase the methods and intensity of data augmentation, such as adding more image transformation operations, or adjusting the amplitude of temperature and humidity data augmentation, expand the dataset, and improve the generalization ability of the model; Hyperparameter adjustment: Adjust hyperparameters such as the learning rate, batch size, and number of training epochs of the optimizer, and find the optimal hyperparameter combination through methods such as grid search or random search.

[0110] It should be noted that when the temperature and humidity information is abnormal, the embodiments of the present application can also realize the prediction of washing parameters. In one implementation, Step S110: Obtain the clothing information of the clothes to be washed in the washing machine and the temperature and humidity information in the washing machine, and the clothing information includes material, stain type, and weight, including:

[0111] Step S111: Detect the initial temperature and humidity information in the washing machine.

[0112] Step S112: Compare the initial temperature and humidity information with a preset temperature and humidity threshold.

[0113] Step S113: When the initial temperature and humidity information is within the temperature and humidity threshold, use the initial temperature and humidity information as the temperature and humidity information in the washing machine.

[0114] Step S114: When the initial temperature and humidity information exceeds the temperature and humidity threshold, send an abnormal reminder, and adjust the initial temperature and humidity information until the adjusted initial temperature and humidity information is within the temperature and humidity threshold, and use the adjusted initial temperature and humidity information as the temperature and humidity information in the washing machine.

[0115] Regarding Step S111 to Step S114:

[0116] A high-precision temperature and humidity sensor can be adopted and evenly distributed around the inner drum of the washing machine. Each temperature and humidity sensor is connected to a data acquisition module. The data acquisition module includes an AD conversion circuit for converting analog signals into digital signals and uploading them to the main control unit through an SPI interface. The main control unit compares and judges the temperature and humidity information. For example, it compares the initial temperature and humidity information with a preset temperature and humidity threshold. When the initial temperature and humidity information is within the threshold, the initial temperature and humidity information is used as the temperature and humidity information inside the washing machine. When the initial temperature and humidity information exceeds the threshold, it indicates that there is a problem with the temperature and humidity inside the inner drum, and an abnormal reminder can be issued to remind the user to repair it in time. At the same time, it can also try to adjust the initial temperature and humidity information, such as by increasing humidity or decreasing temperature, until the initial temperature and humidity information is adjusted within the threshold, and the adjusted initial temperature and humidity information is used as the temperature and humidity information inside the washing machine.

[0117] In the control method of the washing machine provided by the embodiment of the present application, the temperature and humidity inside the inner drum are monitored in real time, which can reduce or avoid the influence of temperature and humidity on washing parameters, thereby improving the matching degree between washing parameters and the laundry to be washed, and improving the washing effect and energy utilization efficiency.

[0118] Please refer to Figure 3 as shown in Figure 3 It is a circuit diagram for detecting the temperature, humidity and motor faults of the washing machine provided by the embodiment of the present application. In the embodiment of the present application, there is also a fault diagnosis module to monitor the working state of the washing machine in real time. Exemplarily, the fault diagnosis module integrates a zero-crossing detection circuit and an intelligent alarm system. Through the triode Q8 as a switching element, the Z_CHECK pin collects the zero-crossing moment of the AC power supply to ensure precise control; once an abnormal temperature and humidity change or abnormal motor current is detected, the module immediately triggers an alarm and notifies the user through a mobile application, and at the same time automatically records the fault code and uploads it to the cloud database for subsequent analysis and improvement.

[0119] Among them, the zero-crossing detection circuit uses a triac to control the voltage across the motor to change the speed of the corresponding motor. Therefore, it is necessary to trigger the thyristor at a determined moment so that the required voltage value can be obtained across the motor. In actual software programming, the conduction moment is determined by a timer. Therefore, the zero-crossing moment of the AC power supply is used as the reference moment, enabling the thyristor to conduct at a determined moment. Two 104 resistors are connected in series to limit the current of the incoming alternating current. This circuit is simple, easy to understand, low-cost, and highly practical. When the incoming power supply is at a high potential, the triode Q8 conducts, and the pin Z_CHECK connected to the main control chip is grounded through the 102 resistor R82 and the triode Q8. Then the chip collects a low potential. When the incoming power supply is at a low potential, the triode Q8 is cut off, and the zero-crossing acquisition terminal Z_CHECK is connected to the VCC5V power supply through R83 and R82. Then the chip collects a high potential. Therefore, the zero-crossing moment of the AC power supply can be determined by the high and low potentials collected by the main control chip.

[0120] According to the different working states and environmental conditions of the washing machine, dynamically adjust the alarm thresholds of temperature, humidity, and motor current. During the washing process, when the motor load increases, such as when there are too many clothes or they are entangled, resulting in an increase in motor resistance, the alarm threshold of the motor current is appropriately increased through an algorithm. For example, when it is detected that the motor speed drops by more than a certain proportion, the current alarm threshold is increased according to a preset formula. In a high-temperature environment, such as in a hot summer indoor environment, adjust the alarm thresholds of temperature and humidity. Appropriately increase the upper limit of the temperature alarm, and also adjust the humidity alarm range accordingly to avoid false alarms caused by environmental factors.

[0121] Once abnormal temperature and humidity changes or abnormal motor current are detected, immediately trigger an alarm. Hardware-wise, connect alarm devices such as a buzzer or an LED indicator to the GPIO pins of the main control unit. Software-wise, write the corresponding alarm handling program. When an abnormal situation is detected, control the alarm device to emit a sound and light alarm signal, and send the fault information to the mobile application and the expert module so that the user can timely understand the fault situation and take corresponding handling measures.

[0122] After the washing machine starts, the expert module generates an adaptive washing mode based on the washing parameters predicted by the convolutional neural network model, the real-time data of each sensor, and the preset rules. For example, when washing woolen clothes, the image acquisition device identifies the clothing material, combines the data of the weight sensor to judge the number of clothes, and the expert module controls the motor to rotate at a relatively low and stable speed (such as 500 revolutions per minute) to avoid felting of woolen clothes. At the same time, set a shorter washing time and an appropriate amount of detergent to be dispensed.

[0123] During the washing process, the fault diagnosis module monitors the data of each sensor and the operating status of the device in real time. If the temperature and humidity sensor detects that the temperature inside the inner drum is too high or the humidity is too low, it may indicate an abnormality in the drying process. The fault diagnosis module immediately issues an alarm and sends an instruction to the main control unit. The main control unit pauses the washing or drying program and initiates cooling or humidifying measures. For abnormal motor current, the fault diagnosis module determines the cause of the fault based on the current fluctuation. If it is caused by clothes entanglement, it controls the motor to rotate forward and backward to shake loose the clothes. If it cannot be solved, it stops the motor operation and notifies the user through a mobile application, and at the same time uploads the fault code and relevant data to the cloud database for subsequent analysis and repair by technicians. Through the collaborative work of the adaptive washing mode and fault diagnosis and repair, the stable operation of the washing machine and the safety of clothes washing are guaranteed.

[0124] In the control method of the washing machine provided by the embodiment of the present application, in this washing machine control method based on temperature and humidity perception, the high-precision temperature and humidity sensor is used to monitor the temperature and humidity changes inside the washing machine in real time and transmit the data to the data acquisition module. Subsequently, the data is uploaded to the main control unit via the SPI interface, and this unit is equipped with a Linux operating system to execute complex computing tasks. The main control unit predicts the optimal washing parameter combination based on the pre-trained convolutional neural network model and the real-time collected data, and dynamically adjusts the water intake, motor speed, and washing time. This not only improves the energy use efficiency but also ensures the best washing effect according to the environmental conditions.

[0125] In this washing machine control method based on temperature and humidity perception, through the expert module, which is developed using the Prolog language, the rule base covers various washing modes such as gentle wash, standard wash, and strong wash, etc., and each mode corresponds to a specific operation instruction sequence. The expert module receives the washing parameter suggestions output by the convolutional neural network model and dynamically adjusts the key parameters in the washing process in combination with the real-time collected data. This design enables the washing machine to intelligently select the most suitable washing program according to the specific situation of the clothes, avoiding damage to the fabric or incomplete cleaning, thereby improving the level of personalized service.

[0126] To facilitate the better implementation of the control method of the washing machine in the embodiment of the present application, the embodiment of the present application also provides a control device for the washing machine. Please refer to Figure 4 as shown in Figure 4 which is the structural block diagram of the control device for the washing machine provided by the embodiment of the present application. The control device 400 of the washing machine includes an acquisition unit 410, a prediction unit 420, and a control unit 430.

[0127] The acquisition unit 410 is used to acquire the clothing information of the clothes to be washed in the washing machine and the temperature and humidity information inside the washing machine. The clothing information includes the material, stain type, and weight.

[0128] The prediction unit 420 is used to predict the washing parameters according to the clothing information and the temperature and humidity information;

[0129] The control unit 430 is used to control the operation of the washing machine based on the washing parameters.

[0130] Any combination of the above technical solutions can form an optional embodiment of the present application, which will not be elaborated here one by one.

[0131] In the control device 400 of the washing machine provided in the embodiment of the present application, the washing parameters are predicted according to the clothing information and the temperature and humidity information, that is, the washing parameters can be dynamically adjusted. Compared with the existing fixed washing parameters, it can dynamically respond to the changes in the washing environment, improve the flexibility of the washing machine, and thus improve the energy utilization efficiency.

[0132] Correspondingly, the embodiment of the present application also provides a washing machine (not shown in the figure). The washing machine includes an image sensor, a weight sensor, and a temperature and humidity sensor. The image sensor is used to detect the material and stain type of the laundry, and the weight sensor is used to detect the weight of the laundry; the temperature and humidity sensor is used to detect the temperature and humidity information inside the washing machine. Of course, the washing machine may also include devices such as a display panel, a cabinet, and a door body, which will not be elaborated here.

[0133] Please refer to Figure 5 as shown Figure 5 which is a structural block diagram of the washing machine provided in the embodiment of the present application. The washing machine 500 may further include a processor 509 with one or more processing cores, a memory 510 with one or more computer-readable storage media, and a computer program stored in the memory 510 and executable on the processor 509. Among them, the processor 509 is electrically connected to the memory 510. Those skilled in the art can understand that the structure of the washing machine 500 shown in the figure does not constitute a limitation on the washing machine, and it may include more or fewer components than shown in the figure, or combine certain components, or have different component arrangements.

[0134] The processor 509 is the control center of the washing machine 500, connecting various parts of the entire washing machine through various interfaces and lines. For example, the processor 509 can be electrically connected to the image sensor, the weight sensor, and the temperature and humidity sensor, and by running or loading the software programs and / or modules stored in the memory 510, and calling the data stored in the memory 510, it executes various functions of the washing machine 500 and processes data, thereby monitoring the washing machine 500 as a whole.

[0135] In an embodiment of the present application, the processor 509 in the washing machine 500 will load the instructions corresponding to the processes of one or more application programs into the memory 510 according to the following steps, and the processor 509 will run the application programs stored in the memory 510 to implement various functions:

[0136] Obtain the clothing information of the laundry to be washed in the washing machine and the temperature and humidity information in the washing machine, where the clothing information includes material, stain type, and weight;

[0137] Predict the washing parameters according to the clothing information and the temperature and humidity information;

[0138] Control the operation of the washing machine based on the washing parameters.

[0139] For the specific implementation of each of the above operations, reference can be made to the previous embodiments, which will not be elaborated here.

[0140] Those of ordinary skill in the art can understand that all or part of the steps in the various methods of the above embodiments can be completed by instructions or by controlling related hardware through instructions. These instructions can be stored in a computer-readable storage medium and loaded and executed by a processor.

[0141] Therefore, an embodiment of the present application provides a computer-readable storage medium, in which multiple computer programs are stored. These computer programs can be loaded by a processor to execute the steps in the control method of the washing machine provided by the embodiment of the present application.

[0142] Among them, the storage medium may include: various media that can store program codes such as a read-only memory (ROM, Read Only Memory), a random access memory (RAM, Random Access Memory), a magnetic disk, or an optical disc.

[0143] Since the computer programs stored in the storage medium can execute the steps in the control method of the washing machine provided by the embodiment of the present application, the beneficial effects achievable by any of the control methods of the washing machine provided by the embodiment of the present application can be realized. For details, refer to the previous embodiments, which will not be elaborated here.

[0144] In the above embodiments, the descriptions of the respective embodiments have their own emphases. For parts not elaborated in a certain embodiment, reference can be made to the relevant descriptions of other embodiments.

[0145] In the description of the present application, the terms "first" and "second" are only used for descriptive purposes and cannot be construed as indicating or implying relative importance or implicitly specifying the quantity of the indicated technical features. Thus, the features defined with "first" and "second" may explicitly or implicitly include one or more features.

[0146] The above has introduced in detail the washing machine, its control method, control device and storage medium provided by the embodiments of the present application. Specific examples are used herein to elaborate on the principle and implementation manner of the present application. The description of the above embodiments is only used to help understand the method and its core idea of the present application; at the same time, for those skilled in the art, according to the idea of the present application, there will be changes in the specific implementation manner and application scope. In summary, the content of this specification should not be construed as a limitation to the present application.

Claims

1. A method for controlling a washing machine, characterized in that: include: Acquire clothing information of the laundry in the washing machine and temperature and humidity information in the washing machine, wherein the clothing information includes material, stain type and weight; predicting washing parameters according to the clothing information and the temperature and humidity information; The operation of the washing machine is controlled based on the washing parameters.

2. The control method according to claim 1, characterized in that: The predicting of washing parameters according to the clothing information and the temperature and humidity information includes: Get the trained neural network model; Based on the clothing information and the temperature and humidity information, washing parameters are predicted according to the neural network model.

3. The control method according to claim 2, characterized in that: The predicting of washing parameters based on the clothing information and the temperature and humidity information and according to the neural network model includes: Inputting the clothing information and the temperature and humidity information into the neural network model, wherein the neural network model includes a matching subnetwork and a prediction subnetwork; Applying the matching sub-network to match the clothing information and the temperature and humidity information with the washing type to obtain a matching result; The prediction subnetwork is used to predict washing parameters of the matching results.

4. The control method according to claim 2, characterized in that: The step of obtaining a trained neural network model comprises: Establish a convolutional neural network model and initialize the convolutional neural network model; Acquire a plurality of clothing samples and a plurality of temperature and humidity samples, pre-process the plurality of clothing samples and the plurality of temperature and humidity samples, and acquire a training set; The convolutional neural network model is iteratively trained according to the training set to obtain a trained neural network model.

5. The control method according to claim 4, characterized in that: The iteratively training the convolutional neural network model according to the training set to obtain the trained neural network model includes: Obtaining the optimal washing parameters corresponding to each test sample obtained through the washing test; Each optimal washing parameter is used as label data, and the training set is used to iteratively train the convolutional neural network model with reference to the label data to obtain a trained neural network model.

6. The control method according to claim 1, characterized in that: The obtaining of clothing information of the laundry in the washing machine and temperature and humidity information in the washing machine, wherein the clothing information includes material, stain type and weight, comprises: Detecting initial temperature and humidity information in the washing machine; Comparing the initial temperature and humidity information with preset temperature and humidity thresholds; When the initial temperature and humidity information is within the temperature and humidity threshold, the initial temperature and humidity information is used as the temperature and humidity information in the washing machine.

7. The control method according to claim 6, characterized in that: The comparing the initial temperature and humidity information with a preset temperature and humidity threshold value includes: When the initial temperature and humidity information exceeds the temperature and humidity threshold, an abnormal reminder is issued, and the initial temperature and humidity information is adjusted until the adjusted initial temperature and humidity information is within the temperature and humidity threshold, and the adjusted initial temperature and humidity information is used as the temperature and humidity information in the washing machine.

8. A control device for a washing machine, characterized in that: include: an acquisition unit, configured to acquire clothing information of the laundry in the washing machine and temperature and humidity information in the washing machine, wherein the clothing information includes material, stain type and weight; a prediction unit, configured to predict washing parameters according to the clothing information and the temperature and humidity information; A control unit is used to control the operation of the washing machine based on the washing parameters.

9. A computer-readable storage medium having a computer program stored thereon, characterized in that: When the computer program is executed on a computer, the computer is caused to execute the control method for the washing machine according to any one of claims 1 to 7.

10. A washing machine, characterized in that: include: Image sensors to detect the material and stain type of laundry; A weight sensor for detecting the weight of laundry; A temperature and humidity sensor, used to detect temperature and humidity information in the washing machine; A processor is electrically connected to the image sensor, the weight sensor and the temperature and humidity sensor respectively, and the processor is used to execute the control method of the washing machine according to any one of claims 1 to 7.

Images