Control methods, devices and systems for anti-collision drums in washing machines

Abstract

This application provides a control method, device, and system for an anti-collision drum in a washing machine. When the washing machine is in the spin-drying phase of the washing process, the attitude data of the drum at the current moment is acquired from an attitude sensor to obtain the current attitude data. The drum rotation speed is gradually increased, and neural network model technology is used to process the current attitude data to obtain a prediction result. Based on the prediction result, the washing machine is controlled to stop working or the drum rotation speed is reduced, achieving the purpose of real-time detection of the risk of drum collision. Compared with existing solutions, this improves the accuracy of the predicted eccentricity value for anti-collision drum detection and reduces the eccentricity detection process time, thus solving the problems of low accuracy and long eccentricity detection process time in existing solutions during the spin-drying process.

Description

Technical Field

[0001] This application relates to the field of washing machine anti-collision drum technology, and more specifically, to a control method, device and system for a washing machine anti-collision drum. Background Technology

[0002] Currently, drum washing machines often experience uneven distribution of laundry during the washing, drying, and spin-drying processes due to differences in the material, size, and absorbency of the clothes. Uneven distribution of laundry not only reduces the washing effect but also causes the drum to become eccentric during rotation, impacting the outer tub and generating loud noise. In severe cases, it can even cause the washing machine to move or jump, resulting in damage.

[0003] Patent application CN110872760A discloses a method for detecting eccentricity in a washing machine. At the start of the spin-drying cycle, when the accelerometer detects that the eccentricity value of the outer tub at the current spin speed is greater than the limit eccentricity value, it further compares the eccentricity value at the current spin speed with the eccentricity value of the tub impact. If the eccentricity value is greater than the impact eccentricity value, a decision is made on whether to perform eccentricity correction; if it is less than the impact eccentricity value, a decision is made to continue spin-drying. However, during the washing process, the accelerometer outputs acceleration, while the eccentricity value of the outer tub is a displacement. There is no necessary correlation between the two, and the limit eccentricity value is difficult to determine, making it difficult to accurately detect the current eccentricity situation.

[0004] The existing solution has low accuracy in predicting and detecting the eccentricity value of the anti-collision barrel during the dehydration process, and the eccentricity detection process takes a long time. Summary of the Invention

[0005] The main objective of this application is to provide a control method, device, and system for the anti-collision drum of a washing machine, so as to at least solve the problems of low accuracy of the predicted eccentricity value of the anti-collision drum and long eccentricity detection process in the existing solution during the spin-drying process.

[0006] To achieve the above objectives, according to one aspect of this application, a control method for a washing machine anti-collision drum is provided, the method comprising:

[0007] When the washing machine is in the spin-drying process, the attitude data of the washing machine drum at the current moment is acquired by the attitude sensor to obtain the current attitude data, and the rotation speed of the drum is gradually increased. The current attitude data includes the acceleration and angular velocity of the drum in each direction of the coordinate axis.

[0008] The current posture data is processed using neural network model technology to obtain a prediction result, which indicates whether the rotating bucket will collide with another bucket or not.

[0009] Based at least on the prediction results, the washing machine can be controlled to stop working or the rotation speed of the drum can be reduced.

[0010] Optionally, based on at least the prediction result, controlling the washing machine to stop working or reduce the rotation speed of the drum includes:

[0011] If the prediction result indicates that the drum will not collide and the spin-drying process is completed, the washing machine is controlled to stop working.

[0012] If the prediction result indicates that the rotating drum is about to collide with another drum, the rotation speed of the rotating drum is reduced by a preset step size.

[0013] When the rotation speed of the drum is greater than or equal to the preset maximum rotation speed, the washing machine is controlled to stop working.

[0014] Optionally, based on at least the prediction result, controlling the washing machine to stop working includes:

[0015] If the prediction result indicates that the rotating drum will collide with another drum, the rotation speed of the rotating drum is reduced by a preset step size, and the current prediction count result is updated to n+1, where n≥0 and n is a positive integer;

[0016] If the current predicted count is the preset maximum number of times, the washing machine is controlled to stop working.

[0017] Optionally, before processing the current pose data using neural network model technology to obtain the prediction result, the method further includes:

[0018] The historical pose data of the rotating drum when the eccentric blocks with different weights are set are obtained. The historical pose data represents the acceleration and angular velocity of the rotating drum in each direction of the coordinate axis collected at a preset acquisition frequency within a preset time period.

[0019] The neural network model is trained using all the historical pose data to obtain the final neural network model.

[0020] Optionally, the neural network model is trained using all the historical pose data to obtain the final neural network model, including:

[0021] according to

[0022] Determine the accuracy of the neural network model, where MSE is the accuracy of the neural network model, N is the number of samples, C is the number of classes, and yi j Let i be the true label of the i-th sample in the j-th category. Let i be the predicted probability of the i-th sample in the j-th category;

[0023] If the accuracy of the neural network model is greater than or equal to the accuracy threshold, the current neural network model is determined to be the final neural network model.

[0024] Optionally, the current pose data is processed using neural network model technology to obtain a prediction result, including:

[0025] Perform a Fourier transform on the current attitude data to obtain the Fourier transform value;

[0026] The prediction result is obtained by processing the Fourier transform quantity using neural network model technology.

[0027] Optionally, in the process of training the neural network model using all the historical pose data to obtain the final neural network model, the method further includes:

[0028] If the barrel displacement corresponding to the Fourier transform is greater than the displacement threshold, the current attitude data corresponding to the Fourier transform is determined to be barrel collision data.

[0029] If the displacement of the rotating barrel corresponding to the Fourier transform quantity is less than or equal to the displacement threshold, the current attitude data corresponding to the Fourier transform quantity is determined to be non-collision barrel data.

[0030] Optionally, the input layer of the neural network model in the neural network model technology contains 6 neurons, which correspond to the Fourier transform quantities of the three linear accelerations and angular accelerations in the current attitude data, respectively. The neural network model also contains two hidden layers, namely a first hidden layer and a second hidden layer. The first hidden layer contains 64 neurons, and the second hidden layer contains 32 neurons. The output layer of the neural network model is 1 neuron, which is used to output the prediction result.

[0031] According to another aspect of this application, a control device for a washing machine anti-collision drum is provided, the device comprising:

[0032] The first acquisition unit is used to acquire the attitude data of the washing machine drum at the current moment collected by the attitude sensor when the washing machine is in the spin-drying process, obtain the current attitude data, and gradually increase the rotation speed of the drum. The current attitude data includes the acceleration and angular velocity of the drum in each direction of the coordinate axis.

[0033] The first processing unit is used to process the current posture data using neural network model technology to obtain a prediction result, wherein the prediction result indicates that the rotating bucket will collide with the bucket, or that the rotating bucket will not collide with the bucket.

[0034] The second processing unit is used to control the washing machine to stop working or reduce the rotation speed of the drum, at least based on the prediction result.

[0035] According to another aspect of this application, a washing machine anti-collision drum system is provided, the system comprising: one or more processors, a memory, and one or more programs, wherein the one or more programs are stored in the memory and configured to be executed by the one or more processors, the one or more programs including methods for performing any one of the methods described.

[0036] By applying the technical solution of this application, when the washing machine is in the spin-drying process, the attitude data of the washing machine drum at the current moment is acquired by the attitude sensor to obtain the current attitude data. The rotation speed of the drum is gradually increased, and the current attitude data is processed by neural network model technology to obtain a prediction result. Based on the prediction result, the washing machine is controlled to stop working or the rotation speed of the drum is reduced. This achieves the purpose of real-time detection of the risk of drum collision. Compared with the existing solution, the accuracy of the anti-collision drum prediction and detection eccentricity value is improved, while the eccentricity detection process time is reduced. This solves the problem that the existing solution has low accuracy of anti-collision drum prediction and detection eccentricity value and long eccentricity detection process time during the spin-drying process. Attached Figure Description

[0037] The accompanying drawings, which form part of this application, are used to provide a further understanding of this application. The illustrative embodiments and descriptions of this application are used to explain this application and do not constitute an undue limitation of this application. In the drawings:

[0038] Figure 1 A flowchart illustrating a control method for a washing machine anti-collision drum according to an embodiment of this application is shown.

[0039] Figure 2 The front and top views show the mounting position of an attitude sensor provided according to an embodiment of this application in a drum washing machine;

[0040] Figure 3 The data acquisition and training process of an attitude sensor according to an embodiment of this application is illustrated;

[0041] Figure 4 A flowchart illustrating another control method for a washing machine anti-collision drum provided according to an embodiment of this application is shown.

[0042] Figure 5 A structural block diagram of a control device for a washing machine anti-collision drum according to an embodiment of this application is shown. Detailed Implementation

[0043] It should be noted that, unless otherwise specified, the embodiments and features described in this application can be combined with each other. This application will now be described in detail with reference to the accompanying drawings and embodiments.

[0044] To enable those skilled in the art to better understand the present application, the technical solutions in the embodiments of the present application will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only some embodiments of the present application, and not all embodiments. Based on the embodiments in the present application, all other embodiments obtained by those of ordinary skill in the art without creative effort should fall within the scope of protection of the present application.

[0045] It should be noted that the terms "first," "second," etc., in the specification, claims, and accompanying drawings of this application are used to distinguish similar objects and are not necessarily used to describe a specific order or sequence. It should be understood that such data can be interchanged where appropriate for the embodiments of this application described herein. Furthermore, the terms "comprising" and "having," and any variations thereof, are intended to cover non-exclusive inclusion; for example, a process, method, system, product, or apparatus that comprises a series of steps or units is not necessarily limited to those steps or units explicitly listed, but may include other steps or units not explicitly listed or inherent to such processes, methods, products, or apparatus.

[0046] For ease of description, the following explains some of the nouns or terms used in the embodiments of this application:

[0047] The Fourier Transform is a mathematical technique used to represent a function as a sum of sine and cosine functions. It has wide applications in signal processing, image processing, and communications.

[0048] The Fast Fourier Transform (FFT) is an efficient algorithm for computing the Fourier transform. The FFT algorithm can significantly reduce computational complexity, especially for sequences of length powers of 2, where its computational efficiency is extremely high. The FFT is widely used in digital signal processing, image processing, and spectrum analysis.

[0049] As described in the background section, patent application CN110872760A discloses a method for detecting eccentricity in a washing machine. At the start of the spin-drying cycle, when the accelerometer detects that the eccentricity value of the outer tub at the current spin speed is greater than the limit eccentricity value, it further compares the eccentricity value at the current spin speed with the eccentricity value of the drum impact. If the eccentricity value is greater than the drum impact eccentricity value, a decision is made on whether to perform eccentricity correction; if it is less than the drum impact eccentricity value, a decision is made to continue spin-drying. However, during the washing process, the accelerometer outputs acceleration, while the eccentricity value of the outer tub is a displacement. There is no necessary connection between the two, and the limit eccentricity value is difficult to determine, making it difficult to accurately detect the current eccentricity. To address the problems of low accuracy in predicting and detecting eccentricity values ​​during the spin-drying process and the long eccentricity detection time in existing solutions, embodiments of this application provide a control method, device, and system for preventing drum impact in a washing machine.

[0050] The technical solutions of the present invention will be clearly and completely described below with reference to the accompanying drawings in the embodiments of the present invention.

[0051] This embodiment provides a control method for the anti-collision drum of a washing machine. It should be noted that the steps shown in the flowchart in the accompanying drawings can be executed in a computer system such as a set of computer-executable instructions. Also, although the logical order is shown in the flowchart, in some cases, the steps shown or described may be executed in a different order than that shown here.

[0052] Figure 1 This is a flowchart illustrating a control method for a washing machine anti-collision drum according to an embodiment of this application.

[0053] like Figure 1 As shown, the method includes the following steps:

[0054] Step S101: When the washing machine is in the spin-drying process, the attitude data of the washing machine drum at the current moment is acquired by the attitude sensor to obtain the current attitude data, and the rotation speed of the drum is gradually increased. The current attitude data includes the acceleration and angular velocity of the drum in each direction of the coordinate axis.

[0055] The rotation speed of the drum can be gradually increased by using 50 revolutions as the step size.

[0056] Specifically, a fast Fourier transform is first performed on the acceleration (ax, ay, az) and angular velocity (rx, ry, rz) output by the attitude sensor to extract the frequency domain features of the current attitude data, and then the frequency domain features of the current attitude data are used as the input of the neural network model.

[0057] like Figure 2The diagram illustrates the installation position of the attitude sensor in a drum washing machine (shown from the front and top views). To better extract the washing machine's attitude during the washing process, the attitude sensor is mounted on the outer tub, with its position centered at the top of the inner tub. The outer tub vibrates to varying degrees during washing, the intensity of which depends on the spindle speed and the degree of eccentricity of the clothes. The attitude sensor mounted on the outer tub reflects this drum vibration during the washing process.

[0058] Step S102: The current posture data is processed using neural network model technology to obtain a prediction result. The prediction result indicates that the rotating barrel will collide with the barrel, or the rotating barrel will not collide with the barrel.

[0059] Step S102 involves processing the current pose data using a neural network model to obtain a prediction result, including:

[0060] Perform a Fourier transform on the above current attitude data to obtain the Fourier transform value;

[0061] The above prediction results were obtained by processing the Fourier transform quantities using neural network model technology.

[0062] Specifically, according to Determine the Fourier transform quantity;

[0063] Where X[k] is the k-th Fourier transform value in the acquired signal sample. x [ n ] represents the nth acquired signal sample, N is the total length of the signal, i is the imaginary unit, and k is the frequency index, ranging from 0 to N-1. Through Fourier transform, the time-domain signal can be transformed into a frequency signal.

[0064] In one embodiment of this application, before processing the current pose data using neural network model technology to obtain the prediction result, the method further includes:

[0065] The historical pose data of the rotating drum when the eccentric blocks with different weights are set are obtained. The historical pose data represents the acceleration and angular velocity of the rotating drum in each direction of the coordinate axis within a preset time period and at a preset acquisition frequency.

[0066] The neural network model was trained using all of the aforementioned historical pose data to obtain the final neural network model.

[0067] When determining whether a device will collide with a barrel, the neural network model needs to be trained. Since the attitude sensors are installed in different locations and are highly sensitive to their current position and attitude, they are calibrated before training to eliminate errors from each installation, in order to obtain more consistent data.

[0068] like Figure 3 The diagram illustrates the data acquisition and training process of the attitude sensor. The drum collision primarily occurs during the high-speed, high-weight spin-drying phase. To reduce data volume, only the drum's attitude during spin-drying is collected. To accommodate variations in spin speed and eccentricity during washing, eccentric blocks of different weights (100g, 200g, 500g, 700g, 1000g, 1500g) are placed during data acquisition. The spin-drying speed is increased in 50-revolution increments. Five data samples are collected at each speed gradient, with a sampling rate of 1kHz (1000 data samples per second). The system then determines whether a drum collision is imminent or has already occurred at the current speed gradient.

[0069] Repeat the training experiment with eccentric blocks of different weights, and observe whether the loss function on the training set decreases to a stable level or reaches an acceptable small value without further significant decline. This indicates that the network has met the training requirements. You can also input some validation data into the training set to verify the training results. If the accuracy reaches 98%, it indicates that the network has completed training.

[0070] In one embodiment of this application, the neural network model is trained using all the aforementioned historical pose data to obtain the final neural network model, including:

[0071] according to

[0072] Determine the accuracy of the above neural network model, where MSE is the accuracy of the neural network model, N is the number of samples, C is the number of classes, and yij is the true label of the i-th sample in the j-th class. Let i be the predicted probability of the i-th sample in the j-th category;

[0073] If the accuracy of the above neural network model is greater than or equal to the accuracy threshold, the current neural network model is determined to be the final neural network model.

[0074] In one embodiment of this application, during the process of training the neural network model using all the aforementioned historical pose data to obtain the final neural network model, the method further includes:

[0075] If the displacement of the rotating barrel corresponding to the above Fourier transform quantity is greater than the displacement threshold, the above current attitude data corresponding to the above Fourier transform quantity is determined to be barrel collision data.

[0076] If the displacement of the rotating barrel corresponding to the Fourier transform quantity is less than or equal to the displacement threshold, the current attitude data corresponding to the Fourier transform quantity is determined to be non-collision barrel data.

[0077] Specifically, the standard for determining whether the bucket collision occurs is to measure whether the left and right displacement d of the bucket during the dehydration process exceeds the maximum set displacement D. If d > D, these 5 data samples are marked as bucket collision data; otherwise, they are marked as non-bucket collision data. If a bucket collision has occurred or the current rotation speed has reached its maximum value, the data collection for the current cycle ends; otherwise, the speed is increased and data collection continues until a bucket collision occurs or the rotation speed reaches the set maximum speed.

[0078] In one embodiment of this application, the input layer of the neural network model in the above-mentioned neural network model technology contains 6 neurons, which correspond to the Fourier transform quantities of the three linear accelerations and angular accelerations in the current attitude data, respectively. The neural network model also contains 2 hidden layers, namely the first hidden layer and the second hidden layer. The first hidden layer contains 64 neurons, and the second hidden layer contains 32 neurons. The output layer of the neural network model is 1 neuron, which is used to output the prediction result.

[0079] As shown in Table 1, the acceleration data in the X-axis direction at 300 rpm under different eccentric values ​​were obtained and the frequency values ​​and collision status were obtained after FFT Fourier transform. It can be seen that whether a collision occurs is positively correlated with the vibration frequency. Generally, the greater the vibration, the easier it is to cause a collision. However, using a single frequency data to predict a collision is prone to deviation. Data under a single working condition cannot cover all situations. Therefore, data collected under different eccentric blocks and different rotation speeds were imported into the neural network model for prediction.

[0080] Table 1

[0081] Eccentric mass Maximum frequency Did it hit the barrel? 200g 11Hz no 500g 23Hz no 700g 59Hz no 1000g 179Hz yes 1500g 249Hz yes

[0082] After training with a neural network, real-time collision prediction is performed during the actual processing. Collisions typically occur during the spin-drying process. Upon entering the spin-drying acceleration phase, the attitude sensor is calibrated, and acceleration begins with a sampling frequency of 1 kHz and a sampling period of 1 second. During acceleration, the attitude sensor data is sampled in real-time and fed into the neural network for prediction. When the neural network predicts a collision, it immediately decelerates and performs anti-collision maneuvers on the garments, recording the number of anti-collision attempts, n. When the number of anti-collision attempts reaches a certain threshold (n > N), it indicates that the garments cannot be prevented from colliding with the garments using the anti-collision maneuvers, and the spin-drying process ends prematurely, awaiting user intervention. If the neural network does not predict a collision or the anti-collision maneuvers are successful, the acceleration continues until spin-drying is complete.

[0083] Step S103: Based at least on the above prediction results, control the washing machine to stop working or reduce the rotation speed of the drum.

[0084] In the above steps, when the washing machine is in the spin-drying process, the attitude data of the washing machine drum at the current moment is acquired by the attitude sensor to obtain the current attitude data. The rotation speed of the drum is gradually increased, and the current attitude data is processed by neural network model technology to obtain the prediction result. Based on the prediction result, the washing machine is controlled to stop working or the rotation speed of the drum is reduced. This achieves the purpose of real-time detection of the risk of drum collision. Compared with the existing solution, the accuracy of the anti-collision drum prediction and detection eccentricity value is improved, while the eccentricity detection process time is reduced. This solves the problem that the existing solution has low accuracy of anti-collision drum prediction and detection eccentricity value and long eccentricity detection process time during the spin-drying process.

[0085] In one embodiment of this application, at least based on the above-mentioned prediction results, controlling the washing machine to stop working or reducing the rotation speed of the drum includes:

[0086] If the above prediction results indicate that the drum will not collide and the spin-drying process is completed, the washing machine will be stopped.

[0087] If the above prediction results indicate that the rotating drum is about to collide with another drum, the rotation speed of the rotating drum is reduced by a preset step size.

[0088] When the rotation speed of the drum is greater than or equal to the preset maximum speed, the washing machine is controlled to stop working.

[0089] The maximum spin speed is set during the washing process. It is stipulated that the maximum spin speed during the spin-drying process of clothes should not exceed the set spin speed. If the maximum spin speed is reached during the spin-drying process, the spin-drying work is completed. If it is not reached, the speed will continue to increase. The best effect is to reach the maximum set spin speed.

[0090] Specifically, if the prediction results indicate that the drum will not collide with the other drum and the spin-drying process is complete, it is determined that there is no need to adjust the drum speed again, and the washing machine is directly stopped. If the drum speed is greater than or equal to the preset maximum speed, it is determined that the speed cannot be increased further, and the washing machine is stopped. If the prediction results indicate that the drum is about to collide with the other drum, it is determined that reducing the speed can prevent the collision, and the drum speed is reduced by a preset step size. The preset maximum speed value is related to the washing machine model.

[0091] In one embodiment of this application, controlling the washing machine to stop working based on at least the above-mentioned prediction results includes:

[0092] If the above prediction result indicates that the rotating drum will collide with another drum, the rotation speed of the rotating drum is reduced by a preset step size, and the current prediction count result is updated to n+1, where n≥0 and n is a positive integer.

[0093] If the current predicted count is the preset maximum number of times, control the washing machine to stop working.

[0094] Specifically, for example, if the preset maximum number of times is set to 10, then when the current predicted count result is 10, the washing machine is controlled to stop working. If the predicted result indicates that the drum is about to collide, the drum speed is reduced by a preset step of 50 revolutions to prevent the drum from colliding.

[0095] To enable those skilled in the art to better understand the technical solution of this application, the implementation process of the control method for the anti-collision drum of the washing machine of this application will be described in detail below with reference to specific embodiments.

[0096] This embodiment relates to a specific control method for the anti-collision drum of a washing machine, such as... Figure 4 Shown, including:

[0097] Acquire historical pose data of the rotating drum when setting eccentric blocks of different weights. The historical pose data represents the acceleration and angular velocity of the rotating drum in each direction of the coordinate axis, collected at a preset acquisition frequency within a preset time period. Use all historical pose data to train the neural network model to obtain the final neural network model.

[0098] When the washing machine is in the spin-drying process, the attitude data of the washing machine drum at the current moment is acquired by the attitude sensor to obtain the current attitude data, and the rotation speed of the drum is gradually increased. The current attitude data includes the acceleration and angular velocity of the drum in each direction of the coordinate axis.

[0099] Perform a Fourier transform on the current posture data to obtain the Fourier transform value; use neural network model technology to process the Fourier transform value to obtain the prediction result, which indicates whether the rotating bucket will collide with the bucket or not.

[0100] If the prediction results indicate that there will be no drum collision during the spin cycle and it is confirmed that the spin-drying process has been completed, control the washing machine to stop working;

[0101] If the prediction result indicates that the rotating drum will collide with another drum, the rotation speed of the rotating drum is reduced by a preset step size, and the current prediction count result is updated to n+1, where n≥0 and n is a positive integer;

[0102] If the drum speed is greater than or equal to the preset maximum speed, the washing machine will stop working.

[0103] If the current predicted count is the preset maximum number of times, control the washing machine to stop working.

[0104] It should be noted that the steps shown in the flowchart in the accompanying drawings can be executed in a computer system such as a set of computer-executable instructions, and although a logical order is shown in the flowchart, in some cases the steps shown or described may be executed in a different order than that shown here.

[0105] This application also provides a control device for a washing machine anti-collision drum. It should be noted that the control device for the washing machine anti-collision drum in this application can be used to execute the control method for the washing machine anti-collision drum provided in this application. This device is used to implement the above embodiments and preferred embodiments; details already described will not be repeated. As used below, the term "module" can refer to a combination of software and / or hardware that implements a predetermined function. Although the device described in the following embodiments is preferably implemented in software, hardware implementation, or a combination of software and hardware, is also possible and contemplated.

[0106] The following describes the control device for the anti-collision drum of a washing machine provided in the embodiments of this application.

[0107] Figure 5 This is a structural block diagram of a control device for a washing machine anti-collision drum provided according to an embodiment of this application. Figure 5 As shown, the device includes:

[0108] The first acquisition unit 51 is used to acquire the attitude data of the washing machine drum at the current moment collected by the attitude sensor when the washing machine is in the spin-drying process, obtain the current attitude data, and gradually increase the rotation speed of the drum. The current attitude data includes the acceleration and angular velocity of the drum in each direction of the coordinate axis.

[0109] The first processing unit 52 is used to process the current posture data using neural network model technology to obtain a prediction result. The prediction result indicates that the rotating barrel will collide with the barrel, or that the rotating barrel will not collide with the barrel.

[0110] The second processing unit 53 is used to control the washing machine to stop working or reduce the rotation speed of the drum, at least based on the above prediction results.

[0111] In the aforementioned device, when the washing machine is in the spin-drying process, the attitude data of the washing machine drum at the current moment is acquired by the attitude sensor to obtain the current attitude data. The rotation speed of the drum is gradually increased, and the current attitude data is processed using neural network model technology to obtain a prediction result. Based on the prediction result, the washing machine is controlled to stop working or the rotation speed of the drum is reduced. This achieves the purpose of real-time detection of the risk of drum collision. Compared with the existing solution, the accuracy of the anti-collision drum prediction and detection eccentricity value is improved, while the eccentricity detection process time is reduced. This solves the problem that the existing solution has low accuracy of anti-collision drum prediction and detection eccentricity value and long eccentricity detection process time during the spin-drying process.

[0112] In one embodiment of this application, the second processing unit includes a first processing module, a second processing module, and a third processing module.

[0113] The first processing module is used to control the washing machine to stop working when the above prediction results indicate that the drum will not collide and the spin-drying work is completed.

[0114] The second processing module is used to reduce the rotation speed of the rotating drum by a preset step size when the prediction result indicates that the rotating drum will collide with the drum.

[0115] The third processing module is used to control the washing machine to stop working when the rotation speed of the drum is greater than or equal to the preset maximum rotation speed.

[0116] In one embodiment of this application, the second processing unit includes a fourth processing module and a fifth processing module.

[0117] The fourth processing module is used to reduce the rotation speed of the rotating drum by a preset step size when the above prediction result indicates that the rotating drum will collide with the drum, and update the current prediction count result to n+1, where n≥0 and n is a positive integer.

[0118] The fifth processing module is used to control the washing machine to stop working when the current predicted count result is the preset maximum number of times.

[0119] In one embodiment of this application, the above-described apparatus further includes a second acquisition unit and a second processing unit.

[0120] The second acquisition unit is used to acquire historical pose data of the rotating barrel when different weights of eccentric blocks are set before processing the current pose data using neural network model technology to obtain the prediction result. The historical pose data represents the acceleration and angular velocity of the rotating barrel in each direction of the coordinate axis collected at a preset acquisition frequency within a preset time period.

[0121] The second processing unit is used to train the neural network model using all the aforementioned historical pose data to obtain the final neural network model.

[0122] In one embodiment of this application, the second processing unit includes a first determining module and a second determining module.

[0123] The first determining module is used to determine based on

[0124] Determine the accuracy of the above neural network model, where MSE is the accuracy of the neural network model, N is the number of samples, C is the number of classes, and yi j Let i be the true label of the i-th sample in the j-th category. Let i be the predicted probability of the i-th sample in the j-th category;

[0125] The second determining module is used to determine the current neural network model as the final neural network model when the accuracy of the above neural network model is greater than or equal to the accuracy threshold.

[0126] In one embodiment of this application, the first processing unit includes a sixth processing module and a seventh processing module.

[0127] The sixth processing module is used to perform a Fourier transform on the above current attitude data to obtain the Fourier transform value;

[0128] The seventh processing module is used to process the above Fourier transform quantities using neural network model technology to obtain the above prediction results.

[0129] In one embodiment of this application, the second processing unit includes a third determining module and a fourth determining module.

[0130] The third determining module is used to determine the current posture data corresponding to the Fourier transform quantity as the barrel collision data when the barrel displacement corresponding to the Fourier transform quantity is greater than the displacement threshold during the process of training the neural network model using all the above-mentioned historical pose data to obtain the final neural network model.

[0131] The fourth determining module is used to determine that the current attitude data corresponding to the Fourier transform quantity is non-collision data when the displacement of the rotating barrel corresponding to the Fourier transform quantity is less than or equal to the displacement threshold.

[0132] In one embodiment of this application, the input layer of the neural network model in the above-mentioned neural network model technology contains 6 neurons, which correspond to the Fourier transform quantities of the three linear accelerations and angular accelerations in the current attitude data, respectively. The neural network model also contains 2 hidden layers, namely the first hidden layer and the second hidden layer. The first hidden layer contains 64 neurons, and the second hidden layer contains 32 neurons. The output layer of the neural network model is 1 neuron, which is used to output the prediction result.

[0133] The aforementioned control device for the washing machine's anti-collision drum includes a processor and a memory. The first acquisition unit, the first processing unit, and the second processing unit are all stored as program units in the memory. The processor executes these program units stored in the memory to achieve the corresponding functions. All of the above modules reside in the same processor; alternatively, the modules may be located in different processors in any combination.

[0134] The processor contains a kernel, which retrieves the corresponding program unit from memory. One or more kernels can be configured. By adjusting the kernel parameters, the problems of low accuracy in predicting and detecting eccentricity values ​​of the anti-collision barrel during the dehydration process and long eccentricity detection time in existing solutions can be addressed.

[0135] The memory may include non-permanent memory in computer-readable media, such as random access memory (RAM) and / or non-volatile memory, such as read-only memory (ROM) or flash RAM, and the memory includes at least one memory chip.

[0136] This invention provides a computer-readable storage medium including a stored program, wherein, when the program is executed, it controls the device containing the computer-readable storage medium to perform the control method for the anti-collision drum of a washing machine.

[0137] This invention provides a processor for running a program, wherein the program executes the control method for the anti-collision drum of the washing machine.

[0138] This invention provides a device including a processor, a memory, and a program stored in the memory and executable on the processor. When the processor executes the program, it performs at least the following steps: When the washing machine is currently in the spin-drying phase of its washing process, it acquires the attitude data of the washing machine's drum at the current moment, collected by an attitude sensor, to obtain current attitude data; and gradually increases the rotational speed of the drum. The current attitude data includes the acceleration and angular velocity of the drum in various directions along the coordinate axes. It then processes the current attitude data using a neural network model to obtain a prediction result, which indicates whether the drum will collide with another drum or not. Finally, based on the prediction result, it controls the washing machine to stop operating or reduces the rotational speed of the drum. The device described herein can be a server, PC, PAD, mobile phone, etc.

[0139] This application also provides a computer program product, which, when executed on a data processing device, is suitable for executing an initialization program having at least the following method steps: when the washing machine is currently in the spin-drying process, acquiring the attitude data of the washing machine drum at the current moment collected by an attitude sensor, obtaining current attitude data, and gradually increasing the rotation speed of the drum, wherein the current attitude data includes the acceleration and angular velocity of the drum in various directions of the coordinate axis; processing the current attitude data using neural network model technology to obtain a prediction result, wherein the prediction result indicates that the drum will collide with another drum, or that the drum will not collide with another drum; and controlling the washing machine to stop working or reduce the rotation speed of the drum, at least based on the prediction result.

[0140] This application also provides a washing machine anti-collision drum system, which includes: one or more processors, a memory, and one or more programs, wherein the one or more programs are stored in the memory and configured to be executed by the one or more processors, and the one or more programs include methods for performing any of the above-described methods. When the washing machine is currently in the spin-drying process, the system acquires the attitude data of the washing machine drum at the current moment collected by an attitude sensor, obtains current attitude data, and gradually increases the rotation speed of the drum. A neural network model is used to process the current attitude data to obtain a prediction result. Based at least on the prediction result, the system controls the washing machine to stop working or reduces the rotation speed of the drum, achieving the purpose of real-time detection of the risk of drum collision. Compared with existing solutions, this improves the accuracy of the anti-collision drum prediction and detection eccentricity value, while reducing the eccentricity detection process time, thereby solving the problems of low accuracy and long eccentricity detection process time in existing solutions during the spin-drying process.

[0141] It is obvious to those skilled in the art that the modules or steps of the present invention described above can be implemented using general-purpose computing devices. They can be centralized on a single computing device or distributed across a network of multiple computing devices. They can be implemented using computer-executable program code, and thus can be stored in a storage device for execution by a computing device. In some cases, the steps shown or described can be performed in a different order than those described herein, or they can be fabricated as separate integrated circuit modules, or multiple modules or steps can be fabricated as a single integrated circuit module. Thus, the present invention is not limited to any particular combination of hardware and software.

[0142] Those skilled in the art will understand that embodiments of this application can be provided as methods, systems, or computer program products. Therefore, this application can take the form of a completely hardware embodiment, a completely software embodiment, or an embodiment combining software and hardware aspects. Furthermore, this application can take the form of a computer program product embodied on one or more computer-usable storage media (including but not limited to disk storage, CD-ROM, optical storage, etc.) containing computer-usable program code.

[0143] This application is described with reference to flowchart illustrations and / or block diagrams of methods, apparatus (systems), and computer program products according to embodiments of this application. It will be understood that each block of the flowchart illustrations and / or block diagrams, and combinations of blocks in the flowchart illustrations and / or block diagrams, can be implemented by computer program instructions. These computer program instructions can be provided to a processor of a general-purpose computer, special-purpose computer, embedded processor, or other programmable data processing apparatus to produce a machine, such that the instructions, which execute via the processor of the computer or other programmable data processing apparatus, generate instructions for implementing the flowchart... Figure 1 a process or multiple processes and / or boxes Figure 1 A device that provides the functions specified in one or more boxes.

[0144] These computer program instructions may also be stored in a computer readable memory that can direct a computer or other programmable data processing device to work in a specific manner, so that the instructions stored in the computer readable memory produce an article of manufacture comprising an instruction device, which implements the process Figure 1 a process or multiple processes and / or boxes Figure 1 The function specified in one or more boxes.

[0145] These computer program instructions can also be loaded onto a computer or other programmable data processing device so that a series of operational steps are executed on the computer or other programmable device to produce a computer-implemented process, thereby providing the instructions executed on the computer or other programmable device for implementing the process. Figure 1 a process or multiple processes and / or boxes Figure 1 A step that specifies a function in one or more boxes.

[0146] In a typical configuration, a computing device includes one or more processors (CPU), input / output interfaces, network interfaces, and memory.

[0147] Memory may include non-persistent memory in computer-readable media, such as random access memory (RAM) and / or non-volatile memory, such as read-only memory (ROM) or flash RAM. Memory is an example of computer-readable media.

[0148] Computer-readable media includes both permanent and non-permanent, removable and non-removable media that can store information using any method or technology. Information can be computer-readable instructions, data structures, modules of programs, or other data. Examples of computer storage media include, but are not limited to, phase-change memory (PRAM), static random access memory (SRAM), dynamic random access memory (DRAM), other types of random access memory (RAM), read-only memory (ROM), electrically erasable programmable read-only memory (EEPROM), flash memory or other memory technologies, CD-ROM, digital versatile optical disc (DVD) or other optical storage, magnetic tape, disk storage or other magnetic storage devices, or any other non-transferable medium that can be used to store information accessible by a computing device. As defined herein, computer-readable media does not include transient computer-readable media, such as modulated data signals and carrier waves.

[0149] It should also be noted that the terms "comprising," "including," or any other variations thereof are intended to cover non-exclusive inclusion, such that a process, method, article, or apparatus that comprises a list of elements includes not only those elements but also other elements not expressly listed, or elements inherent to such process, method, article, or apparatus. Unless otherwise specified, an element defined by the phrase "comprising one..." does not exclude the presence of other identical elements in the process, method, article, or apparatus that includes that element.

[0150] As can be seen from the above description, the embodiments of this application achieve the following technical effects:

[0151] 1) The washing machine anti-collision drum control method of this application, when the washing machine is in the spin-drying process, acquires the attitude data of the washing machine drum collected by the attitude sensor at the current moment, obtains the current attitude data, and gradually increases the rotation speed of the drum. The current attitude data is processed by neural network model technology to obtain a prediction result. Based on the prediction result, the washing machine is controlled to stop working or the rotation speed of the drum is reduced. This achieves the purpose of real-time detection of the risk of drum collision. Compared with the existing solution, it improves the accuracy of the anti-collision drum prediction detection eccentricity value and reduces the eccentricity detection process time, thereby solving the problem that the existing solution has low accuracy of anti-collision drum prediction detection eccentricity value and long eccentricity detection process time during the spin-drying process.

[0152] 2) The washing machine anti-collision drum control device of this application, when the washing machine is in the spin-drying process, acquires the attitude data of the washing machine drum collected by the attitude sensor at the current moment, obtains the current attitude data, and gradually increases the rotation speed of the drum. The current attitude data is processed by neural network model technology to obtain a prediction result. At least based on the prediction result, the washing machine is controlled to stop working or the rotation speed of the drum is reduced, thus achieving the purpose of real-time detection of the risk of drum collision. Compared with the existing solution, the accuracy of the anti-collision drum prediction detection eccentricity value is improved, while the eccentricity detection process time is reduced. This solves the problem that the existing solution has low accuracy of anti-collision drum prediction detection eccentricity value and long eccentricity detection process time during the spin-drying process.

[0153] The above description is merely a preferred embodiment of this application and is not intended to limit this application. Various modifications and variations can be made to this application by those skilled in the art. Any modifications, equivalent substitutions, improvements, etc., made within the spirit and principles of this application should be included within the protection scope of this application.

Claims

1. A control method for an anti-collision drum in a washing machine, characterized in that, include: When the washing machine is in the spin-drying process, the attitude data of the washing machine drum at the current moment is acquired by the attitude sensor to obtain the current attitude data, and the rotation speed of the drum is gradually increased. The current attitude data includes the acceleration and angular velocity of the drum in each direction of the coordinate axis. The current posture data is processed using neural network model technology to obtain a prediction result, which indicates whether the rotating bucket will collide with another bucket or not. Based at least on the prediction results, control the washing machine to stop working or reduce the rotation speed of the drum; Before processing the current posture data using neural network model technology to obtain the prediction result, the method further includes: acquiring historical posture data of the rotating barrel when setting eccentric blocks of different weights, wherein the historical posture data represents the acceleration and angular velocity of the rotating barrel in each direction of the coordinate axis collected at a preset acquisition frequency within a preset time period; and training the neural network model using all the historical posture data to obtain the final neural network model. The neural network model is trained using all the aforementioned historical pose data to obtain the final neural network model, including: according to , Determine the accuracy of the neural network model, where MSE is the accuracy of the neural network model, N is the number of samples, and C is the number of classes. Let i be the true label of the i-th sample in the j-th category. Let i be the predicted probability of the i-th sample in the j-th category; If the accuracy of the neural network model is greater than or equal to the accuracy threshold, the current neural network model is determined to be the final neural network model.

2. The method according to claim 1, characterized in that, Based on at least the predicted results, controlling the washing machine to stop working or reduce the rotation speed of the drum includes: If the prediction result indicates that the drum will not collide and the spin-drying process is completed, the washing machine is controlled to stop working. If the prediction result indicates that the rotating drum is about to collide with another drum, the rotation speed of the rotating drum is reduced by a preset step size. When the rotation speed of the drum is greater than or equal to the preset maximum rotation speed, the washing machine is controlled to stop working.

3. The method according to claim 1, characterized in that, At least based on the predicted results, controlling the washing machine to stop working includes: If the prediction result indicates that the rotating drum will collide with another drum, the rotation speed of the rotating drum is reduced by a preset step size, and the current prediction count result is updated to n+1, where n≥0 and n is a positive integer; If the current predicted count is the preset maximum number of times, the washing machine is controlled to stop working.

4. The method according to claim 1, characterized in that, The current pose data is processed using neural network model technology to obtain prediction results, including: Perform a Fourier transform on the current attitude data to obtain the Fourier transform value; The prediction result is obtained by processing the Fourier transform quantity using neural network model technology.

5. The method according to claim 4, characterized in that, In the process of training the neural network model using all the historical pose data to obtain the final neural network model, the method further includes: If the barrel displacement corresponding to the Fourier transform is greater than the displacement threshold, the current attitude data corresponding to the Fourier transform is determined to be barrel collision data. If the displacement of the rotating barrel corresponding to the Fourier transform quantity is less than or equal to the displacement threshold, the current attitude data corresponding to the Fourier transform quantity is determined to be non-collision barrel data.

6. The method according to any one of claims 1 to 5, characterized in that, The neural network model in the described neural network model technology has an input layer containing 6 neurons, which correspond to the Fourier transform quantities of the three linear accelerations and angular accelerations in the current attitude data. The neural network model also contains two hidden layers, namely a first hidden layer and a second hidden layer. The first hidden layer contains 64 neurons, and the second hidden layer contains 32 neurons. The output layer of the neural network model has 1 neuron, which is used to output the prediction result.

7. A control device for a washing machine anti-collision drum, characterized in that, include: The first acquisition unit is used to acquire the attitude data of the washing machine drum at the current moment collected by the attitude sensor when the washing machine is in the spin-drying process, obtain the current attitude data, and gradually increase the rotation speed of the drum. The current attitude data includes the acceleration and angular velocity of the drum in each direction of the coordinate axis. The first processing unit is used to process the current posture data using neural network model technology to obtain a prediction result, wherein the prediction result indicates that the rotating bucket will collide with the bucket, or that the rotating bucket will not collide with the bucket. The second processing unit is used to control the washing machine to stop working or reduce the rotation speed of the drum, at least based on the prediction result. The control device for the washing machine's anti-collision drum is also used to perform the following steps: Before processing the current posture data using neural network model technology to obtain the prediction result, historical posture data of the rotating barrel when setting different weights of eccentric blocks are obtained. The historical posture data represents the acceleration and angular velocity of the rotating barrel in each direction of the coordinate axis collected at a preset acquisition frequency within a preset time period. The neural network model is trained using all the historical pose data to obtain the final neural network model; according to , Determine the accuracy of the neural network model, where MSE is the accuracy of the neural network model, N is the number of samples, and C is the number of classes. Let i be the true label of the i-th sample in the j-th category. The predicted probability of the i-th sample in the j-th category; if the accuracy of the neural network model is greater than or equal to the accuracy threshold, the current neural network model is determined to be the final neural network model.

8. A washing machine anti-collision drum system, characterized in that, include: One or more processors, a memory, and one or more programs, wherein the one or more programs are stored in the memory and configured to be executed by the one or more processors, the one or more programs comprising methods for performing any one of claims 1 to 6.

Images