Drying control method and device, electronic equipment and medium
By time-series decomposing the drying process parameters and using multi-model prediction models to accurately predict the drying time of clothes, the problem of incomplete drying or excessive drying of clothes in the prior art is solved, and the user experience is improved.
Patent Information
- Application Number
- CN202410018849.5
- Authority / Receiving Office
- CN · China
- Patent Type
- Applications(China)
- Current Assignee / Owner
- Filing Date
- 2024-01-05
- Publication Date
- 2025-07-08
AI Technical Summary
The existing drying control methods only rely on the water content threshold of the clothes for drying, resulting in the clothes not being completely dry or over-drying, affecting the user experience.
By obtaining parameters during the drying process for timing decomposition, the trained drying prediction model is used to predict the completion time point of the laundry drying, and the drying operation of the washing and care equipment is controlled based on this, including the comprehensive application of the periodic drying prediction model, the trend drying prediction model and the mixed prediction model.
Accurately predict the drying time of clothes, avoid premature ending or excessive drying of clothes, and improve user experience.
Smart Images
Figure CN120273165A_ABST
Abstract
Description
Technical Field
[0001] The present invention relates to the technical field of household appliances, and specifically provides a drying control method, device, electronic device and medium. Background Art
[0002] With the development of the economic society, more and more intelligent and modern washing and care devices have emerged in people's daily lives. While these washing and care devices bring convenience to people, they also inevitably bring further problems. For example, when a dryer or a washer-dryer is drying clothes, due to different materials, weights, and water absorption capacities of the clothes to be dried, the required drying time is also different. If the judgment of whether the clothes are dried is only based on a fixed threshold of the water content of the clothes, a large error will occur, resulting in the premature end of drying before all the clothes to be dried are completely dry or over-drying of the clothes to be dried, thereby damaging the clothes to be dried.
[0003] Correspondingly, a new drying control method is needed in this field to solve the above problems. Summary of the Invention
[0004] The present invention aims to solve the above technical problems, that is, to solve the problem that the existing drying control method only judges drying based on the water content threshold of clothes, thereby affecting the user experience.
[0005] To achieve the above object, in a first aspect, the present invention provides a drying control method, and the method includes the following steps:
[0006] Obtain the drying situation parameters during the drying process, and perform time series decomposition on the obtained drying situation parameters to obtain drying decomposition parameters;
[0007] Input the drying decomposition parameters into the trained drying prediction model to obtain the predicted drying completion time point;
[0008] Based on the predicted drying completion time point, control the washing and care device to perform corresponding drying operations.
[0009] In an alternative technical solution of the above drying control method, the step of "controlling the washing and care device to perform corresponding drying operations" includes:
[0010] S11. Obtain the current drying time point of the drying process, and judge whether the current drying time point is the predicted drying completion time point;
[0011] S12. When the current drying time point has not reached the predicted drying completion time point, control the washing and care device to continue to perform the drying process and update the drying situation parameters;
[0012] S13. Perform time series decomposition on the updated drying condition parameters to obtain the drying decomposition parameters;
[0013] S14. Input the drying decomposition parameters into the trained drying prediction model to obtain the predicted drying completion time point, and loop through steps S11 - S14 until the current drying time point reaches the predicted drying completion time point.
[0014] In an alternative technical solution of the above drying control method, the method trains the drying prediction model based on at least the following steps:
[0015] Obtain the drying training parameters during the drying training process, label the drying training parameters, and perform time series decomposition on the labeled drying training parameters to obtain the first drying decomposition parameters;
[0016] Construct a drying prediction model to be trained, where the drying prediction model to be trained includes a cycle drying prediction model to be trained, a trend drying prediction model to be trained, and a hybrid drying prediction model to be trained;
[0017] Input the first drying decomposition parameters into the drying prediction model to be trained for training to obtain the trained drying prediction model.
[0018] In an alternative technical solution of the above drying control method, the washing and care device includes an inner drum, and the step of "performing time series decomposition on the obtained drying training parameters to obtain the first drying decomposition parameters" includes:
[0019] Set the decomposition period based on the rotation condition of the inner drum;
[0020] Perform time series decomposition on the drying training parameters based on the multi - period time series decomposition algorithm and the decomposition period to obtain the cycle training change curve and the trend training change curve.
[0021] In an alternative technical solution of the above drying control method, the washing and care device includes an inner drum, and the step of "performing time series decomposition on the obtained drying training parameters to obtain the first drying decomposition parameters" includes:
[0022] Set multiple decomposition periods based on the rotation condition of the inner drum;
[0023] Perform time series decomposition on the drying training parameters based on the multi - period time series decomposition algorithm and the multiple decomposition periods to obtain the cycle training change curve and the trend training change curve for each decomposition period.
[0024] In an alternative technical solution of the above drying control method, the method further includes:
[0025] Extract the features of the periodic training change curve to obtain periodic training features;
[0026] The step of "inputting the first drying decomposition parameter into the to-be-trained drying prediction model for training to obtain the trained drying prediction model" includes:
[0027] S21. Input the periodic training features into the to-be-trained periodic drying prediction model to obtain a first trained drying prediction result;
[0028] S22. Input the trend training change curve into the to-be-trained trend drying prediction model to obtain a second trained drying prediction result;
[0029] S23. Input the first trained drying prediction result and the second trained drying prediction result into the to-be-trained hybrid prediction model to obtain a trained drying completion time point;
[0030] S24. Obtain a loss function based on the trained drying completion time point and the labeled drying training parameters, and feedback the loss function to step S21. Loop through steps S21 - S24 until the loss function converges.
[0031] In an alternative technical solution of the above drying control method, the method further includes:
[0032] Construct the to-be-trained periodic drying prediction model based on a random forest model;
[0033] Construct the to-be-trained trend drying prediction model based on a Transformer model;
[0034] Construct the to-be-trained hybrid prediction model based on a support vector machine.
[0035] In a second aspect, the present invention also provides a drying control device, and the device includes:
[0036] A time series decomposition module, configured to obtain drying condition parameters during the drying process and perform time series decomposition on the obtained drying condition parameters to obtain drying decomposition parameters;
[0037] A prediction module, configured to input the drying decomposition parameters into the trained drying prediction model to obtain a predicted drying completion time point;
[0038] A control module, configured to control the washing and drying equipment to perform corresponding drying operations based on the predicted drying completion time point.
[0039] In a third aspect, the present invention further provides an electronic device, which includes a memory, a processor, and a computer program stored in the memory and executable on the processor. When the processor executes the computer program, the drying control method described in any one of the above is implemented.
[0040] In a fourth aspect, the present invention further provides a readable storage medium, in which multiple program codes are stored. The program codes are adapted to be loaded and run by a processor to execute the drying control method described in any one of the above.
[0041] Those skilled in the art can understand that in the technical solution of the present invention, by obtaining the drying condition parameters during the drying process and performing time series decomposition on the obtained drying condition parameters to obtain the drying decomposition parameters; inputting the drying decomposition parameters into the trained drying prediction model to obtain the predicted drying completion time point; and based on the predicted drying completion time point, controlling the washing and drying device to perform corresponding drying operations. Such a setting can judge the drying of the clothes to be dried based on the drying prediction model, thereby avoiding the situation that the clothes to be dried end drying prematurely or are over-dried, and improving the user experience.
[0042] Further, inputting the first drying decomposition parameter into the drying prediction model to be trained for training to obtain the trained drying prediction model includes: S21, inputting the periodic training features into the periodic drying prediction model to be trained to obtain the first trained drying prediction result; S22, inputting the trend training change curve into the trend drying prediction model to be trained to obtain the second trained drying prediction result; S23, inputting the first trained drying prediction result and the second trained drying prediction result into the hybrid prediction model to be trained to obtain the trained drying completion time point; S24, obtaining a loss function based on the trained drying completion time point and the labeled drying training parameters, and feeding the loss function back to step S21, and repeatedly executing steps S21-S24 until the loss function converges. Such a setting can predict the optimal drying time of the clothes to be dried from two aspects of the periodic drying prediction model and the trend drying prediction model without setting the drying duration in advance; at the same time, based on the hybrid prediction model, the drying judgment results of the periodic drying prediction model and the trend drying prediction model are synthesized, further improving the accuracy of the drying judgment prediction. BRIEF DESCRIPTION OF THE DRAWINGS
[0043] Referring to the accompanying drawings, the disclosure of the present invention will become more understandable. It is easy for those skilled in the art to understand that these drawings are only for illustrative purposes and are not intended to limit the protection scope of the present invention. In addition, similar numbers in the drawings are used to represent similar components, where:
[0044] Figure 1It is a schematic diagram of the main steps of the drying control method according to an embodiment of the present invention;
[0045] Figure 2 It is a schematic diagram of the main steps of controlling a washing and drying device to perform corresponding drying operations according to an embodiment of the present invention;
[0046] Figure 3 It is a schematic diagram of the main steps of training a drying prediction model according to an embodiment of the present invention;
[0047] Figure 4 It is a schematic diagram of the original data collected by a capacitance sensor according to an embodiment of the present invention;
[0048] Figure 5 It is a schematic diagram of a trend training change curve and a periodic training change curve obtained by taking one revolution of the inner cylinder forward or backward as a decomposition period according to an embodiment of the present invention;
[0049] Figure 6 It is a schematic diagram of a trend training change curve and a periodic training change curve obtained by taking three revolutions of the inner cylinder forward or backward as a decomposition period according to an embodiment of the present invention;
[0050] Figure 7 Schematic diagram of a trend training change curve and a periodic training change curve obtained by taking three revolutions of the inner cylinder forward + a preset residence time + three revolutions backward as a decomposition period according to an embodiment of the present invention;
[0051] Figure 8 Schematic diagram of a trend training change curve and a periodic training change curve obtained by taking (three revolutions forward + a preset residence time + three revolutions backward) * 2 as a decomposition period according to an embodiment of the present invention;
[0052] Figure 9 It is a schematic diagram of the main steps of inputting the first drying decomposition parameters into the drying prediction model to be trained for training according to an embodiment of the present invention;
[0053] Figure 10 It is a schematic diagram of the detailed steps of the drying control method according to an embodiment of the present invention;
[0054] Figure 11 It is a schematic diagram of the main structural block diagram of a drying control device according to an embodiment of the present invention;
[0055] Figure 12 It is a schematic diagram of the main structural block diagram of an electronic device for executing the drying control method of the present invention.
[0056] List of Reference Numerals :
[0057] 11: Time series decomposition module; 12: Prediction module; 13: Control module. Detailed implementation manners
[0058] Some implementation manners of the present invention will be described below with reference to the accompanying drawings. Those skilled in the art should understand that these implementation manners are only used to explain the technical principle of the present invention and are not intended to limit the protection scope of the present invention.
[0059] In the description of the present invention, "module" and "processor" may include hardware, software, or a combination of both. A module may include a hardware circuit, various suitable sensors, communication ports, memories, and may also include a software part, such as program code, or may be a combination of software and hardware. The processor may be a central processing unit, a microprocessor, an image processor, a digital signal processor, or any other suitable processor. The processor has data and / or signal processing functions. The processor may be implemented in software, in hardware, or in a combination of both. The non-transitory computer-readable storage medium includes any suitable medium for storing program code, such as magnetic disks, hard disks, optical disks, flash memories, read-only memories, random access memories, and so on. The term "A and / or B" represents all possible combinations of A and B, such as only A, only B, or A and B. The term "at least one A or B" or "at least one of A and B" has a similar meaning to "A and / or B" and may include only A, only B, or A and B. The singular terms "a" and "this" may also include the plural form.
[0060] As described in the background art section, in view of the problem that the existing drying control method only judges drying based on the moisture content threshold of the clothes, thus affecting the user experience, the present invention provides a drying control method.
[0061] Refer to the attached Figure 1 , Figure 1 is a schematic diagram of the main steps of a drying control method according to an embodiment of the present invention, where the drying control method may be executed by a server, may also be executed by a washing and care device, or may also be jointly executed by a server and a washing and care device; the washing and care device may be a dryer, or a washing and drying integrated machine, etc., and no specific limitation is made here. As Figure 1 shown, the drying control method of the present invention includes the following steps:
[0062] Step S101: Obtain the drying condition parameters during the drying process, and perform time series decomposition on the obtained drying condition parameters to obtain the drying decomposition parameters.
[0063] Specifically, drying clothes requires a certain drying duration. Therefore, when the clothes to be dried start drying, there is no need to perform a dryness prediction on them. After the washing and drying device has executed the drying process for a preset duration, for example, after performing a 20-minute drying operation, the intelligent dryness determination program is started to predict the drying situation of the clothes to be dried. When the washing and drying device starts the intelligent dryness determination program, the water content of the clothes to be dried during the drying process is obtained through a sensor arranged in the washing and drying device to detect the water content of the clothes to be dried. Exemplarily, the sensor for detecting the water content of the clothes to be dried can be a capacitance sensor. The capacitance sensor can detect the capacitance value of the capacitance set between two electrodes in the washing and drying device. There is a certain conversion relationship between the capacitance value and the water content of the clothes. Thus, the water content of the clothes to be dried can be determined through the detected capacitance value and the conversion relationship. The capacitance sensor can directly detect the water content of the clothes to be dried within its detection range as the inner drum of the washing and drying device rotates. Therefore, it is more direct and accurate than the traditional humidity sensor. The types of sensors for detecting the water content of the clothes to be dried described above are only for exemplary illustration, and can be selected according to actual needs in practical applications.
[0064] In some embodiments, the washing and drying device includes an inner drum. Decomposing the obtained drying situation parameters in time series to obtain the drying decomposition parameters includes: setting one or more decomposition cycles based on the rotation situation of the inner drum; decomposing the obtained drying situation parameters in time series based on the multi-cycle time series decomposition algorithm and one or more decomposition cycles to obtain a periodic change curve and a trend change curve. Specifically, the multi-cycle time series decomposition algorithm (Multi Seasonal and Trend decomposition using Loess, i.e., MSTL) is an algorithm for analyzing time series data. Common multi-cycle time series decomposition algorithms include an additive model and a multiplicative model. Among them, the additive model decomposes the time series into four parts: long-term trend, seasonal variation, cyclic variation, and irregular variation; the long-term trend represents the overall change trend of the time series, the seasonal variation represents the fixed periodic change existing in the time series, the cyclic variation represents the non-fixed periodic change existing in the time series, and the irregular variation represents the random change existing in the time series. The multiplicative model decomposes the time series into three parts: long-term trend, seasonal variation, and irregular variation; the product of the long-term trend and the seasonal variation represents the overall change trend of the time series, and the irregular variation represents the random change existing in the time series.
[0065] Step S102: Input the drying decomposition parameters into the trained drying prediction model to obtain the predicted drying completion time point.
[0066] In some embodiments, before inputting the drying decomposition parameters into the trained drying prediction model, the method further includes: extracting the features of the periodic change curve to obtain periodic features; and then inputting the periodic features and the trend change curve into the trained drying prediction model to obtain the predicted drying completion time point.
[0067] Step S103: Based on the predicted drying completion time point, control the washing and care device to perform the corresponding drying operation.
[0068] Based on the above steps S101 to S103, the present invention obtains the drying condition parameters during the drying process, performs time series decomposition on the obtained drying condition parameters to obtain the drying decomposition parameters, inputs the drying decomposition parameters into the trained drying prediction model to obtain the predicted drying completion time point, and based on the predicted drying completion time point, controls the washing and care device to perform the corresponding drying operation. Such a setting can judge the drying of the clothes to be dried based on the drying prediction model, thereby avoiding the situation that the clothes to be dried end drying prematurely or are over-dried, and improving the user experience.
[0069] Refer to the appendix Figure 2 , Figure 2 is a schematic diagram of the main step flow for controlling the washing and care device to perform the corresponding drying operation according to an embodiment of the present invention. As Figure 2 shown, in some embodiments, controlling the washing and care device to perform the corresponding drying operation includes the following steps:
[0070] Step S201: Obtain the current drying time point of the drying process and determine whether the current drying time point is the predicted drying completion time point.
[0071] Step S202: When the current drying time point has not reached the predicted drying completion time point, control the washing and care device to continue the drying process and update the drying condition parameters.
[0072] Step S204: Perform time series decomposition on the updated drying condition parameters to obtain the drying decomposition parameters.
[0073] Step S205: Input the drying decomposition parameters into the trained drying prediction model to obtain the predicted drying completion time point, and loop through steps S201 - S204 until the current drying time point reaches the predicted drying completion time point.
[0074] Specifically, when the current drying time point has not reached the predicted drying completion time point, it indicates that the clothes to be dried have not been dried yet. Therefore, the washing and drying device needs to continue to execute the previous drying process and update the drying condition parameters to the current latest moment, so as to predict the drying completion time point again based on the updated drying condition parameters until the current drying time point reaches the predicted drying completion time point, and then control the washing and drying device to stop executing the drying process.
[0075] In some embodiments, the method further includes: when the washing and drying device executes the drying process, sending the condition information of the clothes to be dried to the user in real time; when the washing and drying device completes the drying process, sending a reminder message of the drying end to the user. Specifically, the condition information of the clothes to be dried may include at least one of the predicted drying completion time point of the clothes to be dried and the predicted drying degree of the clothes to be dried.
[0076] Exemplarily, the form of the condition information of the clothes to be dried may include but is not limited to information of voice type, information of text type, information of picture type, etc. For example, when the condition information of the clothes to be dried is information of voice type, the washing and drying device may push the condition information of the clothes to be dried to the user in the way of voice broadcast. The condition information of the clothes to be dried may be, for example, "The current drying degree of the clothes is 80%, and it is expected to finish drying in 15 minutes"; when the condition information of the clothes to be dried is information of text type, the washing and drying device may push the condition information of the clothes to be dried to the user by displaying it on its display screen or popping up a floating window on the user interface of the mobile terminal. The condition information of the clothes to be dried may be, for example, "Warm reminder: The current drying degree of the clothes is 80%, and it is expected to finish drying in 15 minutes"; when the condition information of the clothes to be dried is information of picture type, the washing and drying device may display picture information related to the drying degree of the clothes to be dried and the predicted drying completion time point of the clothes to be dried on its display screen. Similarly, the drying end reminder message may also include various forms such as voice type and text type, which will not be elaborated here. The form setting of the condition information of the clothes to be dried described above is only for exemplary illustration, and can be selected according to actual needs in practical applications.
[0077] Refer to the appendix Figure 3 , Figure 3 is a schematic diagram of the main step process for training the drying prediction model according to an embodiment of the present invention. As Figure 3 shown, in some embodiments, the method trains the drying prediction model based on at least the following steps:
[0078] Step S301: Obtain the drying training parameters in the drying training process, label the drying training parameters, and perform time series decomposition on the labeled drying training parameters to obtain the first drying decomposition parameters.
[0079] Step S302: Construct a to-be-trained drying prediction model, where the to-be-trained drying prediction model includes a to-be-trained periodic drying judgment prediction model, a to-be-trained trend drying judgment prediction model, and a to-be-trained hybrid prediction model.
[0080] Step S303: Input the first drying decomposition parameter into the to-be-trained drying prediction model for training to obtain a trained drying prediction model.
[0081] Specifically, the drying training parameters include the water content of single-material training clothes and the water content of multi-material training clothes, thus solving the problem of large drying judgment errors when clothes of different types, materials, and weights are mixed together, and improving the adaptability to multiple scenarios; label the drying training parameters with drying time to obtain the labeled drying training parameters. Exemplarily, the drying training parameters can be detected by a capacitance sensor provided in the washing and care device. Specifically, the capacitance sensor can detect the capacitance value of the capacitance set between two electrodes in the washing and care device. There is a certain conversion relationship between the capacitance value and the water content of the clothes. Therefore, the water content of single-material training clothes or multi-material training clothes can be determined through the detected capacitance value and the conversion relationship. The original data of single-material training clothes or multi-material training clothes detected by the capacitance sensor can be as Figure 4 shown, where Figure 4 the horizontal axis is the duration and the vertical axis is the capacitance value of the capacitance sensor.
[0082] In some embodiments, the washing and care device includes an inner drum. The steps for performing time series decomposition on the obtained drying training parameters to obtain the first drying decomposition parameter include:
[0083] Step S401: Set the decomposition period based on the rotation condition of the inner drum.
[0084] Step S402: Perform time series decomposition on the drying training parameters based on the multi-period time series decomposition algorithm and the decomposition period to obtain a periodic training change curve and a trend training change curve.
[0085] Specifically, one of the following can be used as the decomposition period: the inner drum rotates forward or backward 1 week, the inner drum rotates forward or backward 3 weeks, the inner drum rotates forward 3 weeks + a preset residence duration + rotates backward 3 weeks, or the inner drum (rotates forward 3 weeks + a preset residence duration + rotates backward 3 weeks)*2. Thus, time series decomposition is performed on the drying training parameters based on MSTL and the decomposition period. The MSTL algorithm is an extension of the seasonal-trend decomposition procedure based on loess (STL) algorithm. It can be decomposed into a time series with multiple decomposition periods, as shown in formula (1):
[0086] Y(t) = T(t) + S1(t) + S2(t) +... + S nY(t) + R(t) (1)
[0087] Among them, Y(t) is the drying training parameter, n is the number of periodic terms in the drying training parameter, T(t) is the decomposed trend term, S(t) is the periodic term, and R(t) is the residual term.
[0088] In some embodiments, the washing and care device includes an inner drum. The steps for performing time series decomposition on the obtained drying training parameter to obtain the first drying decomposition parameter are as follows:
[0089] Step S501: Set multiple decomposition periods based on the rotation condition of the inner drum.
[0090] Step S502: Perform time series decomposition on the drying training parameter based on the multi-period time series decomposition algorithm and the multiple decomposition periods to obtain the periodic training change curve and the trend training change curve for each decomposition period.
[0091] Specifically, multiple rotations such as the inner drum rotating forward or backward 1 week, the inner drum rotating forward or backward 3 weeks, the inner drum rotating forward 3 weeks + a preset residence time + rotating backward 3 weeks, or the inner drum (rotating forward 3 weeks + a preset residence time + rotating backward 3 weeks) * 2 can be used as the decomposition periods. Exemplarily, when the raw data of single-material training clothes or multi-material training clothes collected by a capacitance sensor arranged in the washing and care device can be as Figure 4 shown, based on the inner drum rotating forward or backward 1 week, the inner drum rotating forward or backward 3 weeks, the inner drum rotating forward 3 weeks + a preset residence time + rotating backward 3 weeks, and the inner drum (rotating forward 3 weeks + a preset residence time + rotating backward 3 weeks) * 2 as the decomposition periods respectively to decompose the drying training parameter, the trend training change curve (Trend) and the periodic training change curve (season) for each decomposition period obtained are respectively as Figures 5 - 8 shown.
[0092] In some embodiments, the method further includes: extracting the features of the periodic training change curve to obtain the periodic training features. Specifically, when a decomposition period is set based on the rotation condition of the inner drum, the periodic training change curve of the decomposition period obtained is subjected to feature extraction to obtain the periodic training features of the decomposition period; when multiple decomposition periods are set based on the rotation condition of the inner drum, the periodic training change curves of the multiple decomposition periods obtained are respectively subjected to feature extraction to obtain the periodic training features of the multiple decomposition periods.
[0093] Exemplarily, at least one of the slope of the periodic training variation curve within a preset stage, the maximum value and the minimum value within the preset stage, and the difference between the maximum value of the previous preset stage and the maximum value of the next preset stage can be extracted to obtain periodic training features. The above-described setting method of the periodic training variation curve features is only for exemplary illustration, and can be selected according to actual needs in practical applications.
[0094] See the appendix Figure 9 , Figure 9 is a schematic diagram of the main step flow for training by inputting the first drying decomposition parameters into a drying prediction model to be trained according to an embodiment of the present invention. As Figure 9 shown, in some embodiments, training by inputting the first drying decomposition parameters into a drying prediction model to be trained to obtain a trained drying prediction model includes the following steps:
[0095] Step S601: Input the periodic training features into the periodic drying prediction model to be trained to obtain a trained first drying prediction result.
[0096] Step S602: Input the trend training variation curve into the trend drying prediction model to be trained to obtain a trained second drying prediction result.
[0097] Step S603: Input the trained first drying prediction result and the trained second drying prediction result into the hybrid prediction model to be trained to obtain a trained drying completion time point.
[0098] Step S604: Obtain a loss function based on the trained drying completion time point and the labeled drying training parameters, and feedback the loss function to Step S601, and repeatedly execute Steps S601 - S604 until the loss function converges.
[0099] Specifically, when a decomposition period is set based on the rotation condition of the inner cylinder, the period training features of the decomposition period extracted are input into the period judgment drying prediction model to be trained, and the trend training change curve of the decomposition period obtained by decomposition is input into the trend judgment drying prediction model to be trained. When multiple decomposition periods are set based on the rotation condition of the inner cylinder, the period training features of the multiple decomposition periods extracted are input into the period judgment drying prediction model to be trained, and the optimal trend training change curve among the multiple decomposition periods obtained by decomposition is input into the trend judgment drying prediction model to be trained; alternatively, the trend training change curves of the multiple decomposition periods obtained by decomposition are all input into the trend judgment drying prediction model to be trained. The first drying prediction result is the moisture content prediction result of the single-material training clothing or multi-material training clothing, and the second drying prediction result is the predicted first drying completion time point of the single-material training clothing or multi-material training clothing; the trained first drying prediction result output by the period judgment drying prediction model and the trained second drying prediction result output by the trend judgment drying prediction model are input into the hybrid prediction model to be trained to obtain the predicted drying completion time point of the single-material training clothing or multi-material training clothing.
[0100] In some embodiments, the method further includes: constructing a period judgment drying prediction model to be trained based on a random forest model; constructing a trend judgment drying prediction model to be trained based on a Transformer model; constructing a hybrid prediction model to be trained based on a support vector machine.
[0101] Specifically, the random forest model is a relatively new machine learning model (a non-linear tree-based model) ensemble learning method. Through the bootstrap resampling technique, it repeatedly and randomly draws K samples (K is generally the same as N) from the original training sample set N with replacement to generate a new training sample set. Then, n classification trees are generated based on the bootstrap sample set to form a random forest. Its essence is an improvement of the decision tree algorithm, combining multiple decision trees together, and the establishment of each tree depends on an independently drawn sample set. The Transformer model can perform excellently in sequence-to-sequence tasks through the self-attention mechanism and the encoder-decoder structure. The self-attention mechanism can assign different importance to each element in the sequence at different positions, and the encoder-decoder structure allows the model to encode on the input sequence and then decode on the output sequence to generate the target sequence. The Support Vector Machine (SVM) is a supervised learning algorithm mainly used for classification and regression analysis. It attempts to find a hyperplane that can maximally separate data points of different classes. SVM can handle linearly separable and non-linearly separable data. For non-linearly separable data, SVM can use the kernel function to map it to a higher-dimensional space to make it linearly separable for classification.
[0102] It should be noted that although the description here is about constructing the to-be-trained cycle judgment prediction model based on the random forest model; constructing the to-be-trained trend judgment prediction model based on the Transformer model; constructing the to-be-trained hybrid prediction model based on the support vector machine, this is not restrictive. Those skilled in the art can also choose other suitable models to construct the corresponding prediction models. For example, those skilled in the art can also construct the to-be-trained cycle judgment prediction model based on the decision tree model, and / or construct the to-be-trained trend judgment prediction model based on the informer model, and / or construct the to-be-trained hybrid prediction model based on the linear regression model. These adjustments do not deviate from the basic principles of the present invention, so they will all fall within the protection scope of the present invention.
[0103] Refer to the appendix Figure 10 , Figure 10 is a detailed step flow diagram of the drying control method according to an embodiment of the present invention. Among them, the drying control method can be executed by the server, or by the washing and care device, or also jointly executed by the server and the washing and care device; the washing and care device can be a dryer, or a washing and drying integrated machine, etc., and no specific limitation is made here. As Figure 10 shown, the drying control method of the present invention includes the following steps:
[0104] Step S701: Obtain the drying condition parameters during the drying process based on the capacitance sensor in the washing and drying device.
[0105] Step S702: Set multiple decomposition periods based on the rotation condition of the inner drum; perform time series decomposition on the drying condition parameters based on the multiple decomposition periods and the multi-period time series decomposition algorithm to obtain the periodic change curve and the trend change curve for each decomposition period.
[0106] Step S703: Extract the features of the periodic change curve for each decomposition period to obtain the periodic features for each decomposition period.
[0107] Step S704: Input the periodic features for each decomposition period into the trained periodic drying prediction model to obtain the first drying prediction result.
[0108] Step S705: Input the trend change curve for each decomposition period into the trained trend drying prediction model to obtain the second drying prediction result.
[0109] Step S706: Input the first drying prediction result and the second drying prediction result into the trained hybrid prediction model to obtain the predicted drying completion time point.
[0110] Step S707: Obtain the current drying time point during the drying process and determine whether the current drying time point is the predicted drying completion time point.
[0111] Step S708: When the current drying time point has not reached the predicted drying completion time point, control the washing and drying device to continue the drying process and loop through steps S701 - S708 until the current drying time point reaches the predicted drying completion time point, at which point the drying process stops.
[0112] It should be noted that although the above embodiments describe the various steps in a specific order, those skilled in the art can understand that in order to achieve the effects of the present invention, it is not necessary for different steps to be executed in such an order. They can be executed simultaneously (in parallel) or in other orders, and these variations are all within the protection scope of the present invention. At the same time, all the above embodiment solutions can be combined arbitrarily to form alternative embodiments of the present invention, which will not be elaborated one by one here.
[0113] Furthermore, the present invention also provides a drying control device.
[0114] Participate in the appendix Figure 11 , Figure 11 is a schematic diagram of the main structural block diagram of the drying control device according to an embodiment of the present invention. As Figure 11As shown in the figure, the drying control device according to an embodiment of the present invention mainly includes a timing decomposition module 11, a prediction module 12, and a control module 13. In some embodiments, one or more of the timing decomposition module 11, the prediction module 12, and the control module 13 may be combined into one module. In some embodiments, the timing decomposition module 11 may be configured to obtain the drying condition parameters during the drying process, and perform timing decomposition on the obtained drying condition parameters to obtain the drying decomposition parameters; the prediction module 12 may be configured to input the drying decomposition parameters into the trained drying prediction model to obtain the predicted drying completion time point; the control module 13 may be configured to control the washing and drying device to perform corresponding drying operations based on the predicted drying completion time point.
[0115] In some embodiments, the control module 13 is configured to perform the following steps: S11, obtain the current drying time point during the drying process, and determine whether the current drying time point is the predicted drying completion time point; S12, when the current drying time point has not reached the predicted drying completion time point, control the washing and drying device to continue the drying process and update the drying condition parameters; S13, perform timing decomposition on the updated drying condition parameters to obtain the drying decomposition parameters; S14, input the drying decomposition parameters into the trained drying prediction model to obtain the predicted drying completion time point, and loop to execute steps S11 - S14 until the current drying time point reaches the predicted drying completion time point.
[0116] In some embodiments, the drying control device further includes a model training module. The model training module obtains the drying training parameters during the drying training process, labels the drying training parameters, and performs timing decomposition on the labeled drying training parameters to obtain the first drying decomposition parameters; constructs a drying prediction model to be trained, where the drying prediction model to be trained includes a drying prediction model for cycle judgment to be trained, a drying prediction model for trend judgment to be trained, and a drying prediction model for hybrid prediction to be trained; inputs the first drying decomposition parameters into the drying prediction model to be trained for training to obtain the trained drying prediction model.
[0117] In some embodiments, the drying control device further includes a model training module. The model training module sets the decomposition period based on the rotation condition of the inner cylinder; performs timing decomposition on the drying training parameters based on the multi - period time series decomposition algorithm and the decomposition period to obtain the periodic training change curve and the trend training change curve.
[0118] In some embodiments, the drying control device further includes a model training module. The model training module sets multiple decomposition periods based on the rotation condition of the inner cylinder; performs timing decomposition on the drying training parameters based on the multi - period time series decomposition algorithm and the multiple decomposition periods to obtain the periodic training change curve and the trend training change curve for each decomposition period.
[0119] In some embodiments, the drying control device further includes a model training module. The model training module extracts the features of the periodic training change curve to obtain periodic training features. The model training module is further configured to perform the following steps: S21, input the periodic training features into the to-be-trained periodic drying prediction model to obtain the first drying prediction result of the training; S22, input the trend training change curve into the to-be-trained trend drying prediction model to obtain the second drying prediction result of the training; S23, input the first drying prediction result of the training and the second drying prediction result of the training into the to-be-trained hybrid prediction model to obtain the drying completion time point of the training; S24, obtain a loss function based on the drying completion time point of the training and the labeled drying training parameters, and feedback the loss function to step S21, and loop to execute steps S21 - S24 until the loss function converges.
[0120] In some embodiments, the drying control device further includes a model training module. The model training module constructs the to-be-trained periodic drying prediction model based on the random forest model; constructs the to-be-trained trend drying prediction model based on the Transformer model; constructs the to-be-trained hybrid prediction model based on the support vector machine.
[0121] Those skilled in the art can understand that all or part of the processes in the method of the above-mentioned embodiment of the present invention can also be completed by instructing relevant hardware through a computer program. The computer program can be stored in a computer-readable storage medium. When the computer program is executed by a processor, the steps of the above-mentioned various method embodiments can be realized. Among them, the computer program includes computer program code, and the computer program code can be in the form of source code, object code, executable file or some intermediate form, etc. The computer-readable storage medium can include: any entity or device, medium, USB flash drive, mobile hard disk, magnetic disk, optical disc, computer memory, read-only memory, random access memory, electrical carrier signal, telecommunication signal, and software distribution medium that can carry the computer program code. It should be noted that the content included in the computer-readable storage medium can be appropriately increased or decreased according to the requirements of legislation and patent practice in the jurisdiction. For example, in some jurisdictions, according to legislation and patent practice, the computer-readable storage medium does not include electrical carrier signals and telecommunication signals.
[0122] Furthermore, the present invention also provides an electronic device.
[0123] Refer to the attached Figure 12 , Figure 12 is a schematic diagram of the main structural block diagram of the electronic device for executing the drying control method of the present invention. As Figure 12As shown in the figure, the present invention also provides an electronic device for implementing the drying control method of the present invention. The electronic device includes: a processor 21, a memory 22, and a computer program 23 stored in the memory 22 and executable on the processor 21. When the processor 21 executes the computer program 23, the steps in the above-mentioned method embodiments are implemented. Alternatively, when the processor 21 executes the computer program 23, the functions of the above-mentioned modules / units in the embodiments are implemented.
[0124] Exemplarily, the processor 21 may be a central processing unit (CPU), or other general-purpose processors, digital signal processors (DSPs), application-specific integrated circuits (ASICs), field-programmable gate arrays (FPGAs), or other programmable logic devices, discrete gate or transistor logic devices, discrete hardware components, etc. The general-purpose processor may be a microprocessor or the processor may also be any conventional processor, etc.
[0125] Exemplarily, the memory 22 may be an internal storage unit of the electronic device. For example, it is a hard disk or memory of the electronic device; the memory 22 may also be an external storage device of the electronic device. For example, a plug-in hard disk, a smart media card (SMC), a secure digital (SD) card, a flash card, etc. equipped on the electronic device. Further, the memory 22 may also include both the internal storage unit and the external storage device of the electronic device. The memory 22 is used to store the computer program and other programs and data required by the electronic device. The memory 22 may also be used to temporarily store the data that has been output or will be output.
[0126] In some possible implementation manners, the electronic device may include multiple processors 21 and memories 22. The program for implementing the drying control method in the above-mentioned method embodiments may be divided into multiple sub-programs, and each sub-program may be loaded and run by the processor 21 respectively to execute different steps of the drying control method in the above-mentioned method embodiments. Specifically, each sub-program may be stored in different memories 22 respectively, and each processor 21 may be configured to execute the program in one or more memories 22 to jointly implement the drying control method in the above-mentioned method embodiments, that is, each processor 21 executes different steps of the drying control method in the above-mentioned method embodiments respectively to jointly implement the drying control method in the above-mentioned method embodiments.
[0127] The above-mentioned multiple processors 21 may be processors deployed on the same device. For example, the above-mentioned electronic device may be a high-performance device composed of multiple processors, and the above-mentioned multiple processors 21 may be the processors configured on the high-performance device. In addition, the above-mentioned multiple processors 21 may also be processors deployed on different devices. For example, the above-mentioned electronic device may be a server cluster, and the above-mentioned multiple processors 21 may be the processors on different servers in the server cluster.
[0001] The electronic device may be a desktop computer, a notebook, a palm computer, a cloud server, or other electronic devices. The electronic device may include, but is not limited to, a processor 21 and a memory 22. Those skilled in the art can understand that Figure 12 These are merely examples of the electronic device and do not constitute a limitation on the electronic device. It may include more or fewer components than those shown in the figure, or combine certain components, or different components. For example, the electronic device may further include input / output devices, network access devices, a bus, etc.
[0128] Furthermore, the present invention also provides a computer-readable storage medium. In an embodiment of the computer-readable storage medium according to the present invention, the computer-readable storage medium may be configured to store a program for executing the drying control method in the above method embodiment. This program may be loaded and run by a processor to implement the above drying control method. For the sake of convenience of description, only the parts related to the embodiments of the present invention are shown. For the specific technical details not disclosed, please refer to the method part of the embodiments of the present invention. The computer-readable storage medium may be a storage device formed by various electronic devices. Optionally, the computer-readable storage medium in the embodiments of the present invention is a non-transitory computer-readable storage medium.
[0129] Furthermore, it should be understood that since the setting of each module is only for explaining the functional units of the device of the present invention, the physical devices corresponding to these modules may be the processor itself, or a part of the software in the processor, a part of the hardware, or a part of the combination of software and hardware. Therefore, the number of each module in the figure is only illustrative.
[0130] Those skilled in the art can understand that the various modules in the device can be adaptively split or combined. Such splitting or combination of specific modules will not cause the technical solution to deviate from the principle of the present invention. Therefore, the technical solutions after splitting or combination will all fall within the protection scope of the present invention.
[0131] So far, the technical solution of the present invention has been described in conjunction with the preferred embodiments shown in the accompanying drawings. However, it is easily understood by those skilled in the art that the protection scope of the present invention is obviously not limited to these specific embodiments. Without departing from the principle of the present invention, those skilled in the art can make equivalent changes or substitutions to the relevant technical features, and the technical solutions after these changes or substitutions will all fall within the protection scope of the present invention.
Claims
1. A drying control method, characterized in that, The method includes the following steps: Obtain the drying condition parameters during the drying process, and perform time series decomposition on the obtained drying condition parameters to obtain drying decomposition parameters; Input the drying decomposition parameters into the trained drying prediction model to obtain the predicted drying completion time point; Based on the predicted drying completion time point, control the washing and care device to perform corresponding drying operations.
2. The drying control method according to claim 1, wherein The step of "controlling the washing and care device to perform corresponding drying operations" includes: S11. Obtain the current drying time point of the drying process, and determine whether the current drying time point is the predicted drying completion time point; S12. When the current drying time point has not reached the predicted drying completion time point, control the washing and care device to continue the drying process and update the drying condition parameters; S13. Perform time series decomposition on the updated drying condition parameters to obtain the drying decomposition parameters; S14. Input the drying decomposition parameters into the trained drying prediction model to obtain the predicted drying completion time point, and loop through steps S11 - S14 until the current drying time point reaches the predicted drying completion time point.
3. The drying control method according to claim 1, wherein, The method trains the drying prediction model at least based on the following steps: Obtain the drying training parameters during the drying training process, label the drying training parameters, and perform time series decomposition on the labeled drying training parameters to obtain the first drying decomposition parameters; Construct a drying prediction model to be trained, where the drying prediction model to be trained includes a cycle drying prediction model to be trained, a trend drying prediction model to be trained, and a hybrid drying prediction model to be trained; Input the first drying decomposition parameters into the drying prediction model to be trained for training to obtain the trained drying prediction model.
4. The drying control method according to claim 3, wherein The washing and care device includes an inner drum. The step of "performing time series decomposition on the obtained drying training parameters to obtain the first drying decomposition parameters" includes: Set the decomposition period based on the rotation condition of the inner drum; Perform time series decomposition on the drying training parameters based on the multi - period time series decomposition algorithm and the decomposition period to obtain the periodic training change curve and the trend training change curve.
5. The drying control method according to claim 3, characterized in that The washing and care device includes an inner drum. The step of "performing time series decomposition on the obtained drying training parameters to obtain the first drying decomposition parameters" includes: Set multiple decomposition periods based on the rotation condition of the inner drum; Perform time series decomposition on the drying training parameters based on the multi - period time series decomposition algorithm and the multiple decomposition periods to obtain the periodic training change curve and the trend training change curve for each decomposition period.
6. The drying control method according to any one of claims 4 or 5, characterized in that, The method further includes: Extract the features of the periodic training change curve to obtain the periodic training features; The step of "inputting the first drying decomposition parameters into the drying prediction model to be trained for training to obtain the trained drying prediction model" includes: S21. Input the periodic training features into the cycle drying prediction model to be trained to obtain the first trained drying prediction result; S22. Input the trend training change curve into the to-be-trained trend judgment and prediction model to obtain the trained second drying prediction result; S23. Input the trained first drying prediction result and the trained second drying prediction result into the to-be-trained hybrid prediction model to obtain the trained drying completion time point; S24. Obtain a loss function based on the trained drying completion time point and the labeled drying training parameters, and feedback the loss function to step S21. Loop through steps S21 - S24 until the loss function converges.
7. The drying control method according to claim 3, wherein The method further includes: Construct the to-be-trained cycle judgment and prediction model based on a random forest model; Construct the to-be-trained trend judgment and prediction model based on a Transformer model; Construct the to-be-trained hybrid prediction model based on a support vector machine.
8. A drying control device, characterized in that, The device includes: A time series decomposition module, configured to obtain the drying condition parameters during the drying process and perform time series decomposition on the obtained drying condition parameters to obtain the dried decomposition parameters; A prediction module, configured to input the dried decomposition parameters into the trained drying prediction model to obtain the predicted drying completion time point; A control module, configured to control the washing and care device to perform corresponding drying operations based on the predicted drying completion time point.
9. An electronic device, comprising a memory, a processor, and a computer program stored in the memory and executable on the processor, characterized in that, When the processor executes the computer program, it implements the drying control method according to any one of claims 1 to 7.
10. A readable storage medium, in which multiple program codes are stored, characterized in that, The program code is suitable for being loaded and run by a processor to execute the drying control method according to any one of claims 1 to 7.