Method and system for optimizing posture strength of oral irrigator

Abstract

The present invention discloses a posture strength optimization method and system for a water flosser, comprising the following steps: step S1: collecting and collating the water flossing posture of the water flosser and the corresponding water flossing strength values; step S2: obtaining the optimal water flossing strength value corresponding to the water flossing posture, and establishing a data set; wherein the water flossing posture includes the position of the nozzle in the oral cavity and the orientation of the nozzle when in this position; and the water flossing strength value is the water flossing strength corresponding to the above-mentioned water flossing posture; step S3: establishing a neural network structure to learn the correspondence between the water flossing posture and the optimal water flossing strength value; step S4: allowing the neural network to learn and evolve; the present invention greatly improves the intelligence level of the water flosser, can provide more suitable use effects for different groups of people, and continuously improves it in the subsequent use process, continuously adapts to personal water flossing habits, and makes the water flossing process more comfortable.

Description

Technical Field

[0001] The present invention relates to the field of oral irrigators, and in particular to a posture strength optimization method and system for oral irrigators. Background Art

[0002] With the development of society and advancements in technology, people are placing increasing emphasis on personal hygiene. Within this realm of personal hygiene, one area that cannot be ignored is oral hygiene. From ancient times, people have consistently cleaned their oral cavity, from the use of tools like willow branches to the widespread use of toothbrushes today. Over these thousands of years of oral hygiene, a wide variety of devices have emerged, each with increasingly sophisticated structures, expanded functions, and improved cleaning effectiveness. Among these, oral irrigators, as the newest and most cutting-edge oral hygiene device, are increasingly gaining recognition and popularity. However, existing oral irrigators on the market still have certain shortcomings, including a low level of intelligence, limited precision control during the oral cleaning process, and limited tailoring for individual oral hygiene needs. In practice, each person's oral condition is unique, requiring oral irrigators to be tailored and adaptable to provide users with a longer-lasting and more comfortable experience. Summary of the Invention

[0003] In order to overcome the above-mentioned shortcomings, the present invention aims to provide a technical solution that can solve the above-mentioned problems.

[0004] A method for optimizing the posture strength of a water flosser comprises the following steps:

[0005] Step S1: collecting and arranging the flushing postures of the water flosser and the corresponding flushing intensity values;

[0006] Step S2: Calculating the optimal flushing intensity value corresponding to the flushing posture and establishing a data set; wherein the flushing posture includes the position of the nozzle in the oral cavity and the orientation of the nozzle in this position; and the flushing intensity value is the flushing intensity corresponding to the aforementioned flushing posture;

[0007] Step S3: establishing a neural network structure to learn the corresponding relationship between the flushing posture and the optimal flushing intensity value;

[0008] Step S4: Allow the neural network to learn and evolve.

[0009] Furthermore, step S1 includes the following sub-steps:

[0010] Step S1.1: Collect and organize detailed data information on different flushing postures and corresponding flushing intensity values ​​during the use of the water flosser.

[0011] Furthermore, step S2 includes the following sub-steps:

[0012] Step S2.1: Based on the collected and collated flushing postures and corresponding flushing intensity values, establish a targeted optimization equation A→B for the flushing intensity value;

[0013] Step S2.2: Based on the above optimization equation, set a range for the upper limit B1 and lower limit B2 of the punching strength value;

[0014] Step S2.3: Discretize the flushing intensity values ​​corresponding to the flushing postures, and then solve the flushing intensity values ​​under the corresponding flushing postures based on the optimization equation established in step S2.1 and the range limit in step S2.2;

[0015] Step S2.4: Calculate the optimal solution for the flushing intensity values ​​under all different flushing postures; establish the corresponding relationship between the flushing postures and the flushing intensity values ​​based on the obtained solution, and establish an initial data set.

[0016] Furthermore, step S3 includes the following sub-steps:

[0017] Step S3.1: Divide the initial data set established in step S2.4 into an initial training set and an initial test set, establish an initial neural network model to learn the mapping relationship, train the initial neural network on the initial training set, and test the initial neural network on the initial test set to determine whether the initial neural network is effective;

[0018] Step S3.2: Establish a neural network search space; use the number of hidden layers of neurons, the number of neurons hidden in each hidden layer, and the learning rate as parameters that can be changed in the process of building a neural network; based on the initial neural network, build multiple neural network models, train them in the initial training set, and then test them in the initial test set; based on the performance on the test set, select multiple neural network models and the network model with the best performance of the initial neural network architecture, and apply them to the system to complete the selection of the optimal flushing intensity value.

[0019] Furthermore, step S4 includes the following sub-steps:

[0020] Step S4.1: Applying the oral irrigator to the user's personal oral care environment and using the currently optimized neural network model to output the optimal irrigating intensity value corresponding to each irrigating posture;

[0021] Step S4.2: During product testing or the actual flushing process of the user, multiple flushing postures during the entire flushing process are recorded to obtain flushing intensity values ​​corresponding to the flushing postures;

[0022] Step S4.3: Recording the user's adjustment data of the flushing intensity value for a certain flushing posture during actual use;

[0023] Step S4.4: Comparing the flushing intensity value output by the current neural network during the experiment or user use with the global optimal flushing intensity value adjusted by the user based on actual user experience. If the absolute value of the deviation between the two is greater than a set threshold, retraining the neural network;

[0024] Step S4.5: Add the flushing postures and corresponding flushing intensity values ​​recorded during the experiment or user use as relationship pairs to the existing dataset, and divide the dataset into an augmented training set and an augmented test set. Search in the established neural network space, construct multiple neural network models, train them in the augmented training set, and then test them in the augmented test set. The network model with the best architectural performance among the multiple neural network models is used as the current optimal neural network model in the water flosser system, and is used to complete future product experiments or users' actual water flossing operations.

[0025] Step S4.6: In future product experiments or actual tooth flushing operations, the current neural network is judged to be the optimal model by continuously determining whether the difference is greater than the threshold, thereby continuously optimizing the model.

[0026] Furthermore, the distance between two adjacent flushing postures is the width of a tooth.

[0027] The present invention also provides a posture strength optimization device for a water flosser, comprising:

[0028] The collection and sorting module is used to collect and sort the flushing posture of the water flosser and the corresponding flushing intensity value;

[0029] A flushing intensity calculation module is used to determine the optimal flushing intensity value corresponding to the flushing posture and establish a data set; the flushing posture includes the position of the nozzle in the mouth and the orientation of the nozzle when in this position; and the flushing intensity value is the flushing intensity corresponding to the aforementioned flushing posture;

[0030] A relationship learning module is used to establish a neural network structure to learn the corresponding relationship between flushing posture and optimal flushing intensity value;

[0031] The learning evolution module is used to allow neural networks to learn and evolve.

[0032] The present invention also provides a posture strength optimization system for a water flosser, comprising:

[0033] Module 1: a module for confirming the posture of the water flosser; the module for confirming the posture of the water flosser comprises the collection and arrangement module according to claim 7;

[0034] Module 2: a watering intensity value adjustment control module for the watering device; the watering intensity value adjustment control module for the watering device comprises the watering intensity calculation module and the relationship learning module according to claim 7;

[0035] Module 3: A learning module for self-evolution of a neural network module; the learning module for self-evolution of a neural network module comprises the learning evolution module as described in claim 7.

[0036] The present invention also provides a computer device, comprising: a processor and a memory, wherein the memory stores a program module, and wherein the program module runs on the processor to implement the method according to any one of claims 1 to 6.

[0037] The present invention also provides a readable storage medium storing a program module, wherein the program module is executed in a processor to implement the method according to any one of claims 1 to 6.

[0038] Compared with the prior art, the present invention has the following beneficial effects:

[0039] 1. The present invention provides a method and system for optimizing the posture strength of an oral irrigator. By collecting and organizing the corresponding relationship between the irrigating status and the irrigating strength value, a data set can be established. The established neural network model can be used to continuously learn during actual use to better fit the user's irrigating habits and provide appropriate irrigating strength values ​​at appropriate irrigating positions, thereby ensuring a longer-lasting and more comfortable user experience while ensuring the cleaning effect. For different users, as the usage time and frequency accumulate, the oral irrigator can generate a irrigating mode that is unique to the user through the learning evolution of the neural network, which has very strong adaptability and user experience.

[0040] Additional aspects and advantages of the present invention will be set forth in part in the description which follows and, in part, will be obvious from the description which follows, or may be learned by practice of the present invention. BRIEF DESCRIPTION OF THE DRAWINGS

[0041] In order to more clearly illustrate the embodiments of the present invention or the technical solutions in the prior art, the following briefly introduces the drawings required for use in the embodiments or the description of the prior art. Obviously, the drawings described below are only some embodiments of the present invention. For ordinary technicians in this field, other drawings can be obtained based on these drawings without paying any creative work.

[0042] Figure 1 It is a schematic flow diagram of the present invention.

[0043] Figure 2 It is a detailed flow diagram of the present invention.

[0044] Figure 3 The present invention is a schematic diagram of the module structure.

[0045] Figure 4 It is a structural schematic diagram of the use of the present invention.

[0046] In the picture:

[0047] 1. Water flosser; 2. Nozzle;

[0048] 3. Oral flosser posture confirmation module; 31. Gravity sensing module; 32. Rotation sensing module; 33. Displacement sensing module; 34. Position sensing module;

[0049] 4. Neural network module self-evolving learning module; 41. Memory; 42. Smart chip;

[0050] 5. Water flosser intensity value adjustment control module; 51. Flow regulator; 52. Pressure detector; 53. Stepless speed regulation water pump. DETAILED DESCRIPTION

[0051] The following is a clear and complete description of the technical solutions in the embodiments of the present invention. Obviously, the embodiments described are only some embodiments of the present invention, not all embodiments. Based on the embodiments of the present invention, all other embodiments obtained by ordinary technicians in this field without making any creative efforts are within the scope of protection of the present invention.

[0052] See also Figures 1 to 4 In an embodiment of the present invention, a method for optimizing the posture strength of a water flosser includes the following steps:

[0053] Step S1: collecting and arranging the flushing postures and corresponding flushing intensity values ​​of the water flosser 1;

[0054] Step S2: Obtaining an optimal flushing intensity value corresponding to a flushing posture and establishing a data set; wherein the flushing posture includes the position of the nozzle 2 in the oral cavity and the orientation of the nozzle 2 in this position; and the flushing intensity value is the flushing intensity corresponding to the aforementioned flushing posture;

[0055] Step S3: establishing a neural network structure to learn the corresponding relationship between the flushing posture and the optimal flushing intensity value;

[0056] Step S4: Allow the neural network to learn and evolve.

[0057] Specifically, in some embodiments, data on the relative position and orientation of the nozzle 2 of the water flosser 1 in the human mouth when a large number of people use the water flosser 1 are collected and organized through experiments or recruiting volunteers, and corresponding data such as the force and flow rate of the water flosser are collected; then, an optimization equation and a data set with relative universality are established through the collected and organized data; and then, multiple neural network models are established, and the neural network models are trained, tested, and screened through the data sets, so that the neural network models can continue to evolve and better fit the user's personal oral cleaning habits and oral cleaning environment.

[0058] In particular, in the selection and collection of flushing postures, we can collect the positions and directions between teeth in a more targeted manner, select these positions and directions as the points of each flushing posture, and collect the corresponding flushing intensity values, which can better help people clean their oral cavity.

[0059] Step S1 includes the following sub-steps:

[0060] Step S1.1: Collect and organize detailed data information of different flushing postures and corresponding flushing intensity values ​​during the use of the water flosser 1.

[0061] Step S2 includes the following sub-steps:

[0062] Step S2.1: Based on the collected and collated flushing postures and corresponding flushing intensity values, establish a targeted optimization equation A→B for the flushing intensity value;

[0063] In particular, when collecting and collating detailed data information on different flushing postures and corresponding flushing intensity values ​​during use of the water flosser 1, the group of people using the water flosser 1 can be divided in a targeted manner;

[0064] In some embodiments, the users of the oral irrigator 1 can be divided by gender. Since males and females have different usage postures and oral conditions for the oral irrigator 1, different optimization equations for the irrigating intensity can be established in a more targeted manner.

[0065] In some other embodiments, the user population can also be divided according to their age. The oral environments of teenagers, adults and the elderly are quite different, and their requirements and adaptability for the intensity value of tooth flushing are also different. The collected data can be further divided into detailed categories to establish a more targeted optimization equation.

[0066] In some other embodiments, the collected data information can also be divided according to regionality. People in different regions have different dietary conditions and eating habits, and also have greatly different oral environments, so optimization equations can be established specifically.

[0067] Step S2.2: Based on the above optimization equation, set a range for the upper limit B1 and lower limit B2 of the punching strength value;

[0068] Specifically, according to the adaptability and tooth flushing needs of different groups of people, the strength of the tooth flusher 1 can be limited to a range. For example, for the tooth flusher 1 used by the elderly, because most of the elderly’s teeth have some problems of falling out or instability, the upper limit of the tooth flushing strength can be appropriately lowered to better protect the oral environment of the elderly.

[0069] Step S2.3: Discretize the flushing intensity values ​​corresponding to the flushing postures, and then solve the flushing intensity values ​​under the corresponding flushing postures based on the optimization equation established in step S2.1 and the range limit in step S2.2;

[0070] Specifically, in some embodiments, because individual users have different usage habits of the water flosser 1, a discrete value within a certain range is selected for the water floss intensity value under a certain water flossing posture to simulate different usage needs of different people.

[0071] Step S2.4: Calculate the optimal solution for the flushing intensity values ​​under all different flushing postures; establish the corresponding relationship between the flushing postures and the flushing intensity values ​​based on the obtained solution, and establish an initial data set.

[0072] Specifically, the optimal solution obtained by the artificially designed optimization equation for different flushing postures is a relatively universal solution, which is applicable to most people or a specific population. The artificially designed optimization equation is based on the collected and organized data, and the data set established in this way is universal and is also an initial data set.

[0073] The step S3 includes the following sub-steps:

[0074] Step S3.1: Divide the initial data set established in step S2.4 into an initial training set and an initial test set, establish an initial neural network model to learn the mapping relationship, train the initial neural network on the initial training set, and test the initial neural network on the initial test set to determine whether the initial neural network is effective;

[0075] Step S3.2: Establish a neural network search space; use the number of hidden layers of neurons, the number of neurons hidden in each hidden layer, and the learning rate as parameters that can be changed in the process of building a neural network; based on the initial neural network, build multiple neural network models, train them in the initial training set, and then test them in the initial test set; based on the performance on the test set, select multiple neural network models and the network model with the best performance of the initial neural network architecture, and apply them to the system to complete the selection of the optimal flushing intensity value.

[0076] Specifically, a neural network model is used to control the flushing intensity value under different flushing postures during the flushing process. When the irrigator 1 moves in the human mouth and reaches a flushing posture, the neural network model outputs a current optimal flushing intensity value. The smart chip 42 uses the flushing intensity value to control the actual components of the irrigator 1 such as the water pump to adjust the flushing force, giving the user the current optimal flushing experience.

[0077] The step S4 includes the following sub-steps:

[0078] Step S4.1: Applying the oral irrigator 1 to the user's personal oral care environment and using the currently optimized neural network model to output the optimal irrigating intensity value corresponding to each irrigating posture;

[0079] Step S4.2: During product testing or the actual flushing process of the user, multiple flushing postures during the entire flushing process are recorded to obtain flushing intensity values ​​corresponding to the flushing postures;

[0080] Step S4.3: Recording the user's adjustment data of the flushing intensity value for a certain flushing posture during actual use;

[0081] Step S4.4: Comparing the flushing intensity value output by the current neural network during the experiment or user use with the global optimal flushing intensity value adjusted by the user based on actual user experience. If the absolute value of the deviation between the two is greater than a set threshold, retraining the neural network;

[0082] Step S4.5: Add the flushing postures and corresponding flushing intensity values ​​recorded during the experiment or user use as relationship pairs to the existing data set, and divide the data set into an augmented training set and an augmented test set. Search in the established neural network space, construct multiple neural network models, train them in the augmented training set, and then test them in the augmented test set. The network model with the best architectural performance among the multiple neural network models is used as the current optimal neural network model and applied to the water flosser system 1 to complete future product experiments or users' actual water flossing operations;

[0083] Step S4.6: In future product experiments or actual tooth flushing operations, the current neural network is judged to be the optimal model by continuously determining whether the difference is greater than the threshold, thereby continuously optimizing the model.

[0084] Specifically, during the user's use process, the water flosser 1 can record the user's usage habits in real time during the actual use process, and record the water flossing status and corresponding water flossing intensity value during the user's actual water flossing process; when the user is in actual use, there may be various personal reasons for the user to manually adjust the water flossing intensity of a certain position; these individuals have their own characteristics, which may be loose or falling teeth, teeth that have just been implanted, or a certain place that needs to be cleaned carefully and vigorously; and the water flosser 1 can record all these water flossing postures and corresponding data, and add the data and relationship pairs with large adjustment differences greater than the threshold to the data set for the neural network model to learn and continuously optimize, so as to better fit and adapt to the user's personal usage habits and oral environment.

[0085] The length of the position interval between two adjacent teeth flushing postures is the width of one tooth.

[0086] Specifically, in some embodiments, the positions and directions of some flushing postures can be determined at the gaps between teeth based on the structure of the human oral cavity and the usage function of the water flosser 1, and the positions between adjacent flushing postures can be separated by one tooth, so that the water flosser 1 can adjust the flushing intensity value at the gaps between teeth more specifically and clean the gaps between teeth more carefully.

[0087] The present invention also provides a posture strength optimization device for a water flosser, comprising:

[0088] A collection and sorting module is used to collect and sort the flushing postures of the water flosser 1 and the corresponding flushing intensity values;

[0089] A flushing intensity calculation module is used to determine the optimal flushing intensity value corresponding to the flushing posture and establish a data set; the flushing posture includes the position of the nozzle in the mouth and the orientation of the nozzle when in this position; and the flushing intensity value is the flushing intensity corresponding to the aforementioned flushing posture;

[0090] A relationship learning module is used to establish a neural network structure to learn the corresponding relationship between flushing posture and optimal flushing intensity value;

[0091] The learning evolution module is used to allow neural networks to learn and evolve.

[0092] The present invention also provides a posture strength optimization system for a water flosser, comprising:

[0093] Module 1: a module for confirming the posture of the oral irrigator 1; the module for confirming the posture of the oral irrigator 1 includes the collection and arrangement module according to claim 7;

[0094] Specifically, in some embodiments, the module 1 includes: a gravity sensing module 31: used to collect and confirm the direction of the oral irrigator 1 in space and confirm the direction of the nozzle 2;

[0095] Rotation sensing module 32: used to collect and confirm the rotation angle of the oral irrigator 1 during use, and to confirm the direction of the nozzle 2 in real time;

[0096] Position sensing module 34: includes multiple staggered energy emitters and energy detectors, and is used in conjunction with an algorithm to locate and confirm the position of the nozzle 2 of the oral irrigator 1 in the human mouth, and collect and organize the position information;

[0097] Displacement sensing module 33: used to detect the displacement posture of the oral irrigator 1 and collect and organize the displacement posture data;

[0098] What is special is that in some embodiments, the energy emitter can be a near-infrared light source generator, and the energy detector is a photoelectric sensor. Multiple staggered light source generators emit light sources, which are reflected by the human face, jaw, upper jaw, and mouth, and then detected by the photoelectric sensor. Based on the received information and the algorithm, the position of the water flosser relative to the human mouth can be confirmed.

[0099] The gravity sensing module 31 and the position sensing module 34 can be used to confirm the initial posture of the water flosser 1 when it is used; and collect and organize the information of the initial posture

[0100] The displacement sensing module 33 and the position sensing module 34 can be used to detect the position change of the nozzle 2 of the water flosser 1 in the human mouth in real time, and can be used to collect and confirm the water flossing posture in real time in combination with the rotation sensing module 32.

[0101] Module 2: a watering intensity value adjustment control module 5 of the watering device; the watering intensity value adjustment control module of the watering device 1 includes the watering intensity calculation module and the relationship learning module as claimed in claim 7;

[0102] The tooth flushing intensity value calculation module calculates the optimal tooth flushing intensity value and establishes a data set.

[0103] Specifically, in some embodiments, module 2 also includes a flow regulator 51, a pressure detector 52 and a stepless speed-regulating water pump 53; the above-mentioned flow regulator 51, pressure detector 52 and stepless speed-regulating water pump 53 are used to monitor the actual output of the flushing intensity value in actual use, and facilitate the user to adjust; and these adjusted data will be added to the data set for the relationship learning module to learn, train and test at all times, and then in turn affect the flushing intensity value output by the neural network model in the future, thereby realizing automatic adjustment and control of the flushing intensity value.

[0104] Module 3: A learning module for self-evolution of a neural network module; the learning module for self-evolution of a neural network module comprises the learning evolution module as described in claim 7.

[0105] Module 3 includes a memory 41 and an intelligent chip 42 for establishing a neural network model and allowing the neural network model to learn and evolve.

[0106] The present invention further provides a computer device, comprising: a processor and a memory 41, wherein the memory 41 stores a program module, and wherein the program module runs on the processor to implement the method according to any one of claims 1 to 6.

[0107] Specifically, the computer device includes the above-mentioned module 3, the processor is the smart chip 42, the memory 41 stores the program module and the data set, the processor is used to run the program module, and establish a neural network model through the program module, so that the neural network model can learn and evolve through the continuously updated data set.

[0108] The present invention further provides a readable storage medium storing a program module, wherein the program module can implement the method according to any one of claims 1 to 6 when executed in a processor.

[0109] It will be apparent to those skilled in the art that the present invention is not limited to the details of the exemplary embodiments described above and that the invention can be embodied in other specific forms without departing from the spirit or essential characteristics of the invention. Therefore, the embodiments should be considered in all respects as illustrative and non-restrictive, and the scope of the invention is defined by the appended claims rather than the foregoing description, and it is intended that all variations that come within the meaning and range of equivalents of the claims be embraced therein.

Claims

1. A method for optimizing the posture strength of a water flosser, characterized in that: The following steps are involved: Step S1: collecting and arranging the flushing postures of the water flosser and the corresponding flushing intensity values; Step S2: Calculating the optimal flushing intensity value corresponding to the flushing posture and establishing a data set; wherein the flushing posture includes the position of the nozzle in the oral cavity and the orientation of the nozzle in this position; and the flushing intensity value is the flushing intensity corresponding to the aforementioned flushing posture; Step S3: establishing a neural network structure to learn the corresponding relationship between the flushing posture and the optimal flushing intensity value; Step S4: Allow the neural network to learn and evolve.

2. The method for optimizing the posture strength of the oral irrigator according to claim 1, characterized in that: The step S1 includes the following sub-steps: Step S1.1: Collect and organize detailed data information on different flushing postures and corresponding flushing intensity values ​​during the use of the water flosser.

3. The method for optimizing the posture strength of the oral irrigator according to claim 2, characterized in that: The step S2 includes the following sub-steps: Step S2.1: Based on the collected and collated flushing postures and corresponding flushing intensity values, establish a targeted optimization equation A→B for the flushing intensity value; Step S2.2: Based on the above optimization equation, set a range for the upper limit B1 and lower limit B2 of the punching strength value; Step S2.3: Discretize the flushing intensity values ​​corresponding to the flushing postures, and then solve the flushing intensity values ​​under the corresponding flushing postures based on the optimization equation established in step S2.1 and the range limit in step S2.2; Step S2.4: Calculate the optimal solution for the flushing intensity values ​​under all different flushing postures; based on the obtained solution, establish the corresponding relationship between the flushing posture and the flushing intensity value, and establish an initial data set.

4. The method for optimizing the posture strength of the oral irrigator according to claim 3, characterized in that: The step S3 includes the following sub-steps: Step S3.1: Divide the initial data set established in step S2.4 into an initial training set and an initial test set, establish an initial neural network model to learn the mapping relationship, train the initial neural network on the initial training set, and test the initial neural network on the initial test set to determine whether the initial neural network is effective; Step S3.2: Establish a neural network search space; use the number of hidden layers of neurons, the number of neurons hidden in each hidden layer, and the learning rate as parameters that can be changed in the process of building a neural network; based on the initial neural network model, construct multiple neural network models, train them in the initial training set, and then test them in the initial test set; based on the performance on the test set, select multiple neural network models and the network model with the best initial neural network architecture performance, and apply them to the system to complete the selection of the optimal flushing intensity value.

5. The method for optimizing the posture strength of the oral irrigator according to claim 4, characterized in that: The step S4 includes the following sub-steps: Step S4.1: Applying the oral irrigator to the user's personal oral care environment and using the currently optimized neural network model to output the optimal irrigating intensity value corresponding to each irrigating posture; Step S4.2: During product testing or the actual flushing process of the user, multiple flushing postures during the entire flushing process are recorded to obtain flushing intensity values ​​corresponding to the flushing postures; Step S4.3: Recording the user's adjustment data of the flushing intensity value for a certain flushing posture during actual use; Step S4.4: Comparing the flushing intensity value output by the current neural network during the experiment or user use with the global optimal flushing intensity value adjusted by the user based on actual user experience. If the absolute value of the deviation between the two is greater than a set threshold, retraining the neural network; Step S4.5: Add the flushing postures and corresponding flushing intensity values ​​recorded during the experiment or user use as relationship pairs to the existing dataset, and divide the dataset into an augmented training set and an augmented test set. Search in the established neural network space to construct multiple neural network models, train them in the augmented training set, and then test them in the augmented test set. The neural network model with the best architectural performance among the multiple neural network models is used as the current optimal neural network model in the water flosser system, and is used to complete future product experiments or users' actual water flossing operations. Step S4.6: In future product experiments or actual tooth flushing operations, the current neural network model is continuously judged to be the optimal model by determining whether the difference is greater than the threshold, thereby continuously optimizing the model.

6. The method for optimizing the posture strength of the oral irrigator according to claim 5, characterized in that: The length of the position interval between two adjacent teeth flushing postures is the width of one tooth.

7. A posture strength optimization device for a water flosser, characterized in that: include: The collection and sorting module is used to collect and sort the flushing posture of the water flosser and the corresponding flushing intensity value; A flushing intensity calculation module is used to determine the optimal flushing intensity value corresponding to the flushing posture and establish a data set; the flushing posture includes the position of the nozzle in the mouth and the orientation of the nozzle when in this position; and the flushing intensity value is the flushing intensity corresponding to the aforementioned flushing posture; A relationship learning module is used to establish a neural network structure to learn the corresponding relationship between flushing posture and optimal flushing intensity value; The learning evolution module is used to allow neural networks to learn and evolve.

8. A posture strength optimization system for a water flosser, characterized in that: include: Module 1: a module for confirming the posture of the water flosser; the module for confirming the posture of the water flosser comprises the collection and arrangement module according to claim 7; Module 2: a watering intensity value adjustment control module for the watering device; the watering intensity value adjustment control module for the watering device comprises the watering intensity calculation module and the relationship learning module according to claim 7; Module 3: A learning module for self-evolution of a neural network module; the learning module for self-evolution of a neural network module comprises the learning evolution module as described in claim 7.

9. A computer device comprising: A processor and a memory, wherein the memory stores a program module, wherein the program module runs on the processor to implement the method according to any one of claims 1 to 6.

10. A readable storage medium storing a program module, characterized in that: The program module is run in a processor to implement the method according to any one of claims 1 to 6.

Images