A method and system for controlling voltage stabilization of oral irrigator

Abstract

The present invention belongs to the technical field of water flosser control, and discloses a water flosser voltage stabilization control method and system, including obtaining water flosser parameter data, ambient temperature data and user usage data; preprocessing the water flosser parameter data, ambient temperature data and user usage data to obtain a water flosser feature data set, a temperature feature data set and a user feature data set; analyzing the influence of the temperature feature data set and the user feature data set on the water flosser feature data set, and fusing the temperature feature data set, the user feature data set and the water flosser feature data set to obtain a comprehensive feature data set; training a water pressure error prediction model for the water flosser, inputting the comprehensive feature data set into the water pressure error prediction model for the water flosser, and predicting the water pressure error of the water flosser; comparing the predicted water pressure error of the water flosser with a preset water pressure error threshold of the water flosser, and determining whether to generate water pressure error warning information; ensuring that a stable water pressure is maintained when the water flosser is in use, and improving the user experience.

Description

Technical Field

[0001] The present invention relates to the technical field of water flosser control, and more specifically, to a water flosser voltage stabilization control method and system. Background Art

[0002] As a medical device widely used in the field of oral care, the stability of the water pressure of the water irrigator has always been one of the key technical indicators; appropriate water pressure can not only ensure good cleaning effects, but also minimize the discomfort of patients.

[0003] The patent application publication number CN108969136A discloses an output control method and system for a water flosser, the method comprising: sending a pulse width modulation signal to the driving module to control the rotation speed of the motor; obtaining the target voltage value under the current gear, and obtaining the input voltage value of the power module according to a preset time interval; adjusting the duty cycle of the pulse width modulation signal according to the ratio of the target voltage value to the input voltage value; and controlling the terminal voltage value of the motor according to the adjusted pulse width modulation signal. Through the above method, the duty cycle of the pulse width modulation signal can be updated in real time according to the change of the input voltage value of the power module, so that the voltage at both ends of the motor reaches a stable range, so that the water flosser can have a stable water pressure output, improve the teeth cleaning effect of the water flosser, and improve the user experience.

[0004] Existing water flossers use simple switch control to discharge water, but they have the following main problems: the water pressure of the water flosser is not adjusted according to the ambient temperature, and it is difficult to maintain a stable cleaning effect at different temperatures; when users use the water flosser in different seasons or regions, the water pressure is unstable due to the difference in ambient temperature, which affects the user experience; if the water flosser is operated at extreme temperatures without corresponding protective measures, the internal components of the device will be damaged and the service life will be shortened; different users have different preferences for water pressure and different frequencies of use. Existing water flossers do not provide adjustable water pressure levels or intelligent adaptation functions, and cannot meet the personalized needs of users. Users need to frequently replace parts or perform repairs during use, which will increase the maintenance cost and time cost of users;

[0005] The existing technology only relies on information from a single dimension for prediction, ignoring the influence of other important factors, resulting in a significant decrease in the accuracy and reliability of the prediction; without integrating feature data sets from multiple sources, the model cannot fully capture the various factors that affect water pressure adjustment, and thus cannot make accurate predictions and adjustments; without using cross-validation to evaluate the performance of the model on different data subsets, it is easy for the model to perform well on training data but poorly on new data, resulting in a significant reduction in the effect in practical applications;

[0006] In the prior art, precise control of water pressure is not considered, and the actual water pressure will fluctuate around the target value. Users will feel sudden changes in water pressure during use, affecting the cleaning effect and comfort. Unstable water pressure may cause the water flosser to be unable to effectively remove residues in the gaps between teeth, affecting the cleaning effect. Frequent adjustment of water pumps and valves under unstable water pressure will accelerate the wear of mechanical parts and shorten the service life of the equipment.

[0007] In view of this, the present invention proposes a method and system for controlling the voltage of a water flosser to solve the above problems. Summary of the invention

[0008] In order to overcome the above-mentioned defects of the prior art and to achieve the above-mentioned purpose, the present invention provides the following technical solution: a method for controlling the voltage stabilization of a water flosser, comprising:

[0009] S1, obtaining water flosser parameter data, ambient temperature data and user usage data;

[0010] S2, preprocessing the water flosser parameter data, the ambient temperature data and the user usage data to obtain a water flosser characteristic data set, a temperature characteristic data set and a user characteristic data set;

[0011] Analyze the influence of the temperature feature data set and the user feature data set on the water flosser feature data set, and fuse the temperature feature data set, the user feature data set and the water flosser feature data set to obtain a comprehensive feature data set;

[0012] S3, training a water pressure error prediction model for the water flosser, inputting the comprehensive feature data set into the water pressure error prediction model for the water flosser, and predicting the water pressure error of the water flosser;

[0013] S4, comparing the predicted water pressure error of the water flosser with a preset water pressure error threshold of the water flosser to determine whether to generate water pressure error warning information;

[0014] S5. If a water pressure error warning message is generated, the control instruction is executed to adjust the rotation speed of the water pump and the valve opening of the water flosser to stabilize the water pressure of the water flosser.

[0015] Furthermore, the water flosser parameter data includes water pressure data, water flow data and battery status data of the water flosser; the ambient temperature data includes the external ambient temperature of the water flosser; and the user usage data includes the user's usage time and frequency.

[0016] Furthermore, the method for preprocessing the water flosser parameter data, the ambient temperature data and the user usage data includes:

[0017] Density clustering algorithm is used on the water flosser parameter data, ambient temperature data and user usage data to identify abnormal values ​​in the water flosser parameter data, ambient temperature data and user usage data, and the identified abnormal values ​​are removed to obtain the processed water flosser feature data set, temperature feature data set and user feature data set;

[0018] The acquired water flosser feature data set, temperature feature data set and user feature data set were normalized by standard deviation and converted into a standard normal distribution with a mean of 0 and a standard deviation of 1 to eliminate the dimensional influence between the data. Finally, the normalized water flosser feature data set, temperature feature data set and user feature data set were obtained.

[0019] Furthermore, the method for analyzing the influence of the temperature characteristic data set and the user characteristic data set on the water flosser characteristic data set includes:

[0020] The influence of temperature characteristics on water flosser characteristics: The ambient temperature will affect the working efficiency of the water flosser components. When the ambient temperature is low, the viscosity of water increases, the fluidity decreases, and a greater pressure is required to maintain the same water flow rate. When the ambient temperature is high, the viscosity of water decreases, the fluidity increases, and the pressure needs to be reduced. The flow-pressure adjustment model is used to maintain the stability of the water flow rate. The flow-pressure adjustment model is: ;in, is the water flow; is water pressure; is the viscosity of water; is the viscosity of water; is a constant;

[0021] The viscosity of the water is: ;in, is the ambient temperature; Viscosity value at standard reference temperature; It is a constant related to fluid properties and temperature, describing the degree of influence of temperature on viscosity;

[0022] The influence of user characteristics on water flosser characteristics: Different users have different preferences for water pressure and usage frequency. User characteristics affect the working parameters of water flossers. By defining a target water pressure To meet user needs, the target water pressure is:

[0023] ;

[0024] in, For user preference of water pressure; is the frequency of use; The influence coefficient of user usage frequency on target water pressure; is the ambient temperature adjustment factor;

[0025] At different ambient temperatures Under this condition, the target water pressure of the water flosser is adjusted by setting the ambient temperature adjustment factor according to the user's preference. ; The ambient temperature adjustment factor is: ;in, ;in, ;in, , the user feature dataset is recorded as , the feature dataset of the water flosser is recorded as ;

[0026] The weighted model is:

[0027] ;

[0028] in, is the weight coefficient of the temperature feature data set; is the weight coefficient of the user feature data set; is the weight coefficient of the water flosser feature data set;

[0029] For the weight design of each feature data set in the weighted model, the temperature feature data set , user feature dataset and water flosser feature dataset Combined into a feature matrix , and the corresponding target variable ;set up for Matrix; where is the number of features;

[0030] Define a set of weighted combination search grids, and set the weights , and Divide into several equally spaced points from 0 to 1;

[0031] The weighted combination search grid is:

[0032] ;

[0033] in, is the weight coefficient of the temperature feature data set The index of is the weight coefficient of the user feature dataset The index of is the weight coefficient of the water flosser feature data set The index of

[0034] Use cross-validation to evaluate the performance of each weight combination, and use each weight combination as a dataset; divide the dataset into mutually exclusive subsets; in each partition, The weighted model was trained using the training set, and the performance of the model was evaluated using the root mean square error on the validation set.

[0035] The root mean square error is:

[0036] ;

[0037] in, The number of data sets on the validation set; the predicted values ​​calculated for the weighted model, is the actual value of the target variable; for each weight combination, the average performance index is calculated on all partitions to obtain the average performance of the weight combination; the weight combination that makes the average performance index optimal is selected;

[0038] Set weight combination The average performance index is , then the optimal weight combination is:

[0039] ;

[0040] in, Optimal weight combination; An operator that returns the value of the variable that minimizes the objective function.

[0041] Furthermore, the training method of the water pressure error prediction model of the water flosser includes:

[0042] The dataset was divided into training set, validation set and test set, and the model was trained and its performance was evaluated. A water pressure error prediction model for water flossers was constructed, including input layer, hidden layer and output layer. The hidden layer used the ReLU activation function to avoid oversimplification of the model and alleviate the gradient vanishing problem.

[0043] The input layer of the water pressure error prediction model for water flossers is a historical comprehensive feature data set, and the number of neurons in the input layer should match the number of features in the historical comprehensive feature data set; the output layer of the model is the water pressure error of the water flosser, and the number of neurons in the output layer corresponds to the number of prediction targets, and the prediction value is output through one neuron; the water pressure error prediction model for water flossers is a fully connected neural network model;

[0044] The mean square error is used as the loss function to measure the error between the model's predicted value and the actual value; the mean square error loss function is:

[0045] ;

[0046] in, is the number of data sets; Data points in the dataset The actual value of Data points in the dataset The predicted value of

[0047] Use the training set data to train the model and use the Adam optimizer to minimize the loss function. Use the validation set to evaluate the performance of the water pressure error prediction model for the water flosser, calculate the accuracy index to measure the performance of the model, and tune the model's hyperparameters to improve the model performance.

[0048] The performance of the model in the prediction task is evaluated through the test set. When the performance of the model in the prediction task reaches the preset performance threshold, the test is stopped to obtain the water pressure error prediction model of the water flosser. The current comprehensive feature data set is input into the water pressure error prediction model of the water flosser to obtain the water pressure error of the water flosser.

[0049] Furthermore, the method of tuning the hyperparameters of the model to improve the model performance includes:

[0050] The hyperparameter set of the model is defined as , for each hyperparameter , The distribution of ;in, for Distribution of for The lower limit of the distribution of for The upper limit of the distribution of , in each iteration In the example, we randomly select the value of each hyperparameter from Extract from distribution , forming a complete set of hyperparameter configurations ;in, is the number of iterations; is the hyperparameter of the model; is the number of model hyperparameters;

[0051] Use the selected hyperparameter combination Train the model and use it on the validation set Evaluate model performance on ; Repeat the above process until the preset number of iterations is reached , and obtain a series of performance evaluations ; Select performance from all iterations The highest hyperparameter combination ;

[0052] Use the best hyperparameter combination The model is trained on the training set and evaluated on the test set to confirm the generalization ability of the model.

[0053] Furthermore, the method of comparing the predicted water pressure error of the water flosser with a preset water pressure error threshold of the water flosser to determine whether to generate water pressure error warning information includes:

[0054] If the predicted water pressure error of the water flosser is greater than the preset water pressure error threshold of the water flosser, a water pressure error warning message is generated;

[0055] If the predicted water pressure error of the water flosser is less than or equal to the preset water pressure error threshold of the water flosser, no water pressure error warning information will be generated.

[0056] Furthermore, the method for adjusting the rotation speed of the water pump and the valve opening of the water flosser to stabilize the water pressure of the water flosser includes:

[0057] If a water pressure error warning message is generated, the water pump speed and valve opening of the water flosser are adjusted by improving the PID algorithm to stabilize the water pressure of the water flosser;

[0058] The target water pressure of the preset water flosser is ,exist The actual water pressure of the water flosser measured by the force sensor is ; Calculate the error between the target water pressure and the actual water pressure ;

[0059] The adjustment amount of the pump speed and valve opening is controlled by improving the PID algorithm according to the error calculation. The adjustment amount is:

[0060] ;

[0061] in, It is the adjustment amount of water pump speed and valve opening; is the proportional gain, which determines the direct influence of the error on the control quantity; is the integral gain, which determines the influence of the cumulative error on the control quantity; is the differential gain, which responds to the rate of change of the error; is the error between the target water pressure and the actual water pressure; is the virtual time variable in the integration process, representing the time from the initial time 0 to the current time Continuous changes in From the initial time 0 to the current time , total accumulation of errors, over time; The time required to control the pump speed and valve opening;

[0062] The adjustment amount of the PID algorithm output It is used to adjust the pump speed and valve opening to affect the water flow, so that the actual water pressure gradually approaches the target water pressure; Feedback to the system, recalculate the error, and continuously adjust it to form a closed-loop control; , and The value is tuned to determine the PID controller , and The optimal value of makes the control system achieve fast response, minimum overshoot and stability;

[0063] Preset overshoot threshold is 0.05, steady-state error threshold ±0.01, stable time threshold is 2; and Set to 0 to only adjust Make the system respond stably without excessive oscillation; gradually increase Stop until the overshoot of the control system reaches the preset overshoot threshold; gradually increase To reduce or eliminate the steady-state error, stop when it is lower than the preset steady-state error threshold; gradually increase Reduce overshoot and shorten stabilization time, and stop when the stabilization time of the control system reaches the preset stabilization time threshold.

[0064] A water flosser voltage stabilization control system, comprising:

[0065] A data acquisition module is used to obtain water flosser parameter data, ambient temperature data and user usage data;

[0066] A data processing module, used for preprocessing the water flosser parameter data, the ambient temperature data and the user usage data to obtain a water flosser characteristic data set, a temperature characteristic data set and a user characteristic data set;

[0067] Analyze the influence of the temperature feature data set and the user feature data set on the water flosser feature data set, and fuse the temperature feature data set, the user feature data set and the water flosser feature data set to obtain a comprehensive feature data set;

[0068] A model training module is used to train a water pressure error prediction model for a water flosser, input a comprehensive feature data set into the water pressure error prediction model for a water flosser, and predict the water pressure error of the water flosser;

[0069] The water pressure error prediction module compares the predicted water pressure error of the water flosser with the preset water pressure error threshold of the water flosser to determine whether to generate water pressure error warning information;

[0070] The water pressure regulating module, if a water pressure error warning message is generated, executes the control instruction to adjust the speed of the water pump and the valve opening of the water flosser to stabilize the water pressure of the water flosser.

[0071] The technical effects and advantages of the method and system for controlling the voltage stabilization of a water flosser of the present invention are as follows:

[0072] By analyzing the impact of temperature characteristics on the characteristics of the water flosser, a flow-pressure adjustment model is introduced to accurately describe and predict the amount of water pressure adjustment required for the water flosser at different temperatures; the water pressure is adjusted in real time according to the ambient temperature to ensure that the water flosser can provide stable and effective cleaning effects in different environments; the impact of user characteristics on the characteristics of the water flosser is analyzed, and a target water pressure formula is defined, taking into account factors such as user preference for water pressure, frequency of use, and ambient temperature, which can more accurately meet the personalized needs of users. The introduction of the ambient temperature adjustment factor enables the water flosser to automatically adjust the target water pressure according to user preferences under different ambient temperatures; meet the personalized needs of different users for water pressure, improve user satisfaction, and enhance user dependence and loyalty to the product;

[0073] By weighted fusion of feature data sets from different sources, the model can comprehensively consider information from multiple dimensions, thereby improving the accuracy and reliability of predictions; it can capture the combined effects of temperature, user preferences, and the characteristics of the water flosser itself on water pressure adjustment, and the use of cross-validation ensures that the performance of the model on different data subsets can be evaluated, thereby enhancing the generalization ability of the model; by automatically searching for the optimal weight combination, this method can efficiently utilize limited computing resources and avoid the subjectivity and uncertainty caused by manual adjustment of weights;

[0074] By accurately calculating the error between the target water pressure and the actual water pressure and adjusting the water pump speed and valve opening accordingly, the water pressure of the oral irrigator can be accurately controlled; this helps ensure that users enjoy stable and appropriate water pressure during use; a closed-loop control strategy is adopted, that is, the new water pressure is fed back to the system, the error is recalculated and continuously adjusted; this ensures that the system can continuously correct deviations and maintain water pressure stability; by tuning the proportional, integral and differential gains in the PID controller, the control system can respond quickly while reducing overshoot and oscillation, thereby improving the stability of the system; this ensures safety and comfort of use, and avoids oral discomfort or equipment damage that may be caused by unstable water pressure. BRIEF DESCRIPTION OF THE DRAWINGS

[0075] Figure 1 A schematic flow chart of a method for controlling a water flosser voltage according to the present invention;

[0076] Figure 2 It is a schematic diagram of the structure of a water flosser voltage stabilizing control system of the present invention. DETAILED DESCRIPTION

[0077] The following will be combined with the drawings in the embodiments of the present invention to clearly and completely describe the technical solutions in the embodiments of the present invention. Obviously, the described embodiments are only part of the embodiments of the present invention, not all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by ordinary technicians in this field without creative work are within the scope of protection of the present invention. Example

[0078] See also Figure 1 and Figure 2 As shown, the present embodiment provides a method for controlling a water flosser voltage stabilization, including:

[0079] S1, obtaining water flosser parameter data, ambient temperature data and user usage data;

[0080] S2, preprocessing the water flosser parameter data, the ambient temperature data and the user usage data to obtain a water flosser characteristic data set, a temperature characteristic data set and a user characteristic data set;

[0081] Analyze the influence of the temperature feature data set and the user feature data set on the water flosser feature data set, and fuse the temperature feature data set, the user feature data set and the water flosser feature data set to obtain a comprehensive feature data set;

[0082] S3, training a water pressure error prediction model for the water flosser, inputting the comprehensive feature data set into the water pressure error prediction model for the water flosser, and predicting the water pressure error of the water flosser;

[0083] S4, comparing the predicted water pressure error of the water flosser with a preset water pressure error threshold of the water flosser to determine whether to generate water pressure error warning information;

[0084] S5. If a water pressure error warning message is generated, the control instruction is executed to adjust the rotation speed of the water pump and the valve opening of the water flosser to stabilize the water pressure of the water flosser.

[0085] The water flosser parameter data includes water pressure data, water flow data and battery status data of the water flosser; the ambient temperature data includes the external ambient temperature of the water flosser; the user usage data includes the user's usage time and frequency;

[0086] The water pressure data of the water flosser is obtained through a pressure sensor installed at the water outlet or near the nozzle of the water pump; the water flow data is obtained through a flow sensor installed at the water inlet or outlet of the water flosser; the battery status data is obtained through a battery management system or a built-in power detection circuit; the external ambient temperature of the water flosser is obtained through a temperature sensor installed on the outer shell or inside of the water flosser; the user's usage time is obtained through a built-in microprocessor or timer of the water flosser; the usage frequency is indirectly calculated by analyzing the usage time data. For example, the user's usage time in a week can be recorded, and the average weekly usage times can be calculated based on this.

[0087] The method for preprocessing the water flosser parameter data, the ambient temperature data and the user usage data comprises:

[0088] Density clustering algorithm is used on the water flosser parameter data, ambient temperature data and user usage data to identify abnormal values ​​in the water flosser parameter data, ambient temperature data and user usage data, and the identified abnormal values ​​are removed to obtain the processed water flosser feature data set, temperature feature data set and user feature data set;

[0089] The acquired water flosser feature data set, temperature feature data set and user feature data set were normalized by standard deviation and converted into a standard normal distribution with a mean of 0 and a standard deviation of 1 to eliminate the dimensional influence between the data. Finally, the normalized water flosser feature data set, temperature feature data set and user feature data set were obtained.

[0090] The method for analyzing the influence of the temperature characteristic data set and the user characteristic data set on the water flosser characteristic data set includes:

[0091] The influence of temperature characteristics on water flosser characteristics: The ambient temperature will affect the working efficiency of the water flosser components. When the ambient temperature is low, the viscosity of water increases, the fluidity decreases, and a greater pressure is required to maintain the same water flow rate. When the ambient temperature is high, the viscosity of water decreases, the fluidity increases, and the pressure needs to be reduced. The flow-pressure adjustment model is used to maintain the stability of the water flow rate. The flow-pressure adjustment model is: ;in, is the water flow; is water pressure; is the viscosity of water; is a constant; the viscosity of the water is: ;in, is the ambient temperature; is the viscosity value at the standard reference temperature; It is a constant related to fluid properties and temperature, describing the degree of influence of temperature on viscosity;

[0092] For example, the following data are measured through experiments:

[0093]

[0094] Using these data, fitting techniques can be used to determine and Assuming that the value of =1.002mPa·s and = 2400°C, these constants are used to predict the viscosity at different temperatures.

[0095] The influence of user characteristics on water flosser characteristics: Different users have different preferences for water pressure and usage frequency. User characteristics affect the working parameters of water flossers. By defining a target water pressure To meet user needs, the target water pressure is:

[0096] ;

[0097] in, For user preference of water pressure; is the frequency of use; The influence coefficient of user usage frequency on target water pressure; is the ambient temperature adjustment factor;

[0098] At different ambient temperatures In this case, the target water pressure of the water flosser can be adjusted according to user preference by setting the ambient temperature adjustment factor. ; The ambient temperature adjustment factor is: ;in, is the ambient temperature adjustment factor; Room temperature is the reference temperature.

[0099] For example, if the user's preference for water pressure is 7, the frequency of use is 2 times, the influence coefficient of the user's use frequency on the target water pressure is 0.8, the room temperature reference temperature is 25°C, the water viscosity is 1mPa·s, and the ambient temperature adjustment factor is 0.04; then the target water pressure It is 0.344MPa.

[0100] The method of fusing the temperature feature data set, the user feature data set and the water flosser feature data set to obtain a comprehensive feature data set includes:

[0101] The temperature feature dataset, user feature dataset and water flosser feature dataset are fused through a weighted model to form a comprehensive feature dataset. The temperature feature dataset is recorded as , the user feature dataset is recorded as , the feature dataset of the water flosser is recorded as ;

[0102] The weighted model is: ;in, is the weight coefficient of the temperature feature data set; is the weight coefficient of the temperature feature data set; is the weight coefficient of the water flosser feature data set;

[0103] For the weight design of each feature data set in the weighted model, the temperature feature data set , user feature dataset and water flosser feature dataset Combined into a feature matrix , and the corresponding target variable ;set up for Matrix; where is the number of features;

[0104] Define a set of weighted combination search grids, and set the weights , and Divide into several equally spaced points from 0 to 1;

[0105] The weighted combination search grid is:

[0106] ;

[0107] in, is the weight coefficient of the temperature feature data set The index of is the weight coefficient of the user feature dataset The index of is the weight coefficient of the water flosser feature data set The index of

[0108] Use cross-validation to evaluate the performance of each weight combination, and use each weight combination as a dataset; divide the dataset into mutually exclusive subsets; in each partition, The weighted model was trained using the training set, and the performance of the model was evaluated using the root mean square error on the validation set.

[0109] The root mean square error is:

[0110] ;

[0111] in, is the number of data sets on the validation set; is the predicted value calculated by the weighted model, is the actual value of the target variable;

[0112] For each weight combination, the average performance index is calculated on all partitions to obtain the average performance of the weight combination; the weight combination that makes the average performance index optimal is selected;

[0113] Set weight combination The average performance index is , then the optimal weight combination is:

[0114] ;

[0115] in, is the best weight combination; An operator that returns the value of the variable that minimizes the objective function.

[0116] For example, suppose there are the following weight combinations and the corresponding average performance indicators are Minimum weight:

[0117]

[0118] In this table, the average performance index The smallest weight combination is (0.5, 0.3, 0.2), and the corresponding average performance index is is 0.3, so is (0.5, 0.3, 0.2), which means that (0.5, 0.3, 0.2) is chosen as the best weight combination.

[0119] The training method of the water pressure error prediction model of the water flosser includes:

[0120] The dataset was divided into training set, validation set and test set, and the model was trained and its performance was evaluated. A water pressure error prediction model for water flossers was constructed, including input layer, hidden layer and output layer. The hidden layer used the ReLU activation function to avoid oversimplification of the model and alleviate the gradient vanishing problem.

[0121] The input layer of the water pressure error prediction model for water flossers is a historical comprehensive feature data set, and the number of neurons in the input layer should match the number of features in the historical comprehensive feature data set; the output layer of the model is the water pressure error of the water flosser, and the number of neurons in the output layer corresponds to the number of prediction targets, and the prediction value is output through one neuron; the water pressure error prediction model for water flossers is a fully connected neural network model;

[0122] The mean square error is used as the loss function to measure the error between the model's predicted value and the actual value; the mean square error loss function is: ;in, Number of datasets; Data points in the dataset The actual value of Data points in the dataset The predicted value of

[0123] Use the training set data to train the model and use the Adam optimizer to minimize the loss function. Use the validation set to evaluate the performance of the water pressure error prediction model for the water flosser, calculate the accuracy index to measure the performance of the model, and tune the model's hyperparameters to improve the model performance.

[0124] The performance of the model in the prediction task is evaluated through the test set. When the performance of the model in the prediction task reaches the preset performance threshold, the test is stopped to obtain the water pressure error prediction model of the water flosser. The current comprehensive feature data set is input into the water pressure error prediction model of the water flosser to obtain the water pressure error of the water flosser.

[0125] The method of tuning the hyperparameters of the model to improve the model performance includes:

[0126] The hyperparameter set of the model is defined as , for each hyperparameter , The distribution of ;in, for Distribution of for The lower limit of the distribution of for The upper limit of the distribution of , in each iteration In the example, we randomly select the value of each hyperparameter from Extract from distribution , forming a complete set of hyperparameter configurations ;in, is the number of iterations; is the hyperparameter of the model; is the number of model hyperparameters;

[0127] Use the selected hyperparameter combination Train the model and use it on the validation set Evaluate model performance on ; Repeat the above process until the preset number of iterations is reached , and obtain a series of performance evaluations ; Select performance from all iterations The highest hyperparameter combination ;

[0128] Use the best hyperparameter combination The model is trained on the training set and evaluated on the test set to confirm the generalization ability of the model.

[0129] The method of comparing the predicted water pressure error of the water flosser with a preset water pressure error threshold of the water flosser to determine whether to generate water pressure error warning information includes:

[0130] If the predicted water pressure error of the water flosser is greater than the preset water pressure error threshold of the water flosser, a water pressure error warning message is generated;

[0131] If the predicted water pressure error of the water flosser is less than or equal to the preset water pressure error threshold of the water flosser, no water pressure error warning information will be generated.

[0132] The method for adjusting the rotation speed of the water pump and the valve opening of the water flosser to stabilize the water pressure of the water flosser comprises:

[0133] If a water pressure error warning message is generated, the water pump speed and valve opening of the water flosser are adjusted by improving the PID algorithm to stabilize the water pressure of the water flosser;

[0134] The target water pressure of the preset water flosser is ,exist At this moment, the actual water pressure of the water flosser is measured by the pressure sensor. ; Calculate the error between the target water pressure and the actual water pressure ;

[0135] The adjustment amount of the pump speed and valve opening is controlled by improving the PID algorithm according to the error calculation. The adjustment amount is:

[0136] ;

[0137] in, It is the adjustment amount of water pump speed and valve opening; is the proportional gain, which determines the direct influence of the error on the control quantity; is the proportional gain, which determines the direct influence of the error on the control quantity; is the differential gain, which responds to the rate of change of the error; is the error between the target water pressure and the actual water pressure; is the virtual time variable in the integration process, representing the time from the initial time 0 to the current time Continuous changes in From the initial time 0 to the current time ,error Total accumulation over time; The time required to control the pump speed and valve opening;

[0138] The adjustment amount of the PID algorithm output It is used to adjust the pump speed and valve opening to affect the water flow, so that the actual water pressure gradually approaches the target water pressure; Feedback is fed back to the system, the error is recalculated, and adjustments are made continuously to form a closed-loop control;

[0139] For PID controller , and The value is tuned to determine the PID controller , and Set to 0 to only adjust Make the system respond stably without excessive oscillation; gradually increase Stop until the overshoot of the control system reaches the preset overshoot threshold; gradually increase To reduce or eliminate the steady-state error, stop when it is lower than the preset steady-state error threshold; gradually increase Reduce overshoot and shorten stabilization time, and stop when the stabilization time of the control system reaches the preset stabilization time threshold.

[0140] The preset water pressure error threshold of the water flosser is set by the staff, and different water pressure errors of the water flosser are collected through the water flosser voltage stabilization control terminal, and the average value of multiple water pressure errors of the water flosser is taken as the preset water pressure error threshold of the water flosser; the overshoot threshold, steady-state error threshold and stabilization time threshold are set similarly;

[0141] In this embodiment, by analyzing the influence of temperature characteristics on the characteristics of the water flosser, a flow-pressure adjustment model is introduced to accurately describe and predict the water pressure adjustment required for the water flosser at different temperatures; the water pressure is adjusted in real time according to the ambient temperature to ensure that the water flosser can provide a stable and effective cleaning effect in different environments; the influence of user characteristics on the characteristics of the water flosser is analyzed, and a target water pressure formula is defined, which takes into account factors such as user preference for water pressure, frequency of use, and ambient temperature, so as to more accurately meet the personalized needs of users. The introduction of the ambient temperature adjustment factor enables the water flosser to automatically adjust the target water pressure according to user preferences under different ambient temperatures; the personalized needs of different users for water pressure are met, user satisfaction is improved, and user dependence and loyalty to the product are enhanced;

[0142] By weighted fusion of feature data sets from different sources, the model can comprehensively consider information from multiple dimensions, thereby improving the accuracy and reliability of predictions; it can capture the combined effects of temperature, user preferences, and the characteristics of the water flosser itself on water pressure adjustment, and the use of cross-validation ensures that the performance of the model on different data subsets can be evaluated, thereby enhancing the generalization ability of the model; by automatically searching for the optimal weight combination, this method can efficiently utilize limited computing resources and avoid the subjectivity and uncertainty caused by manual adjustment of weights;

[0143] By accurately calculating the error between the target water pressure and the actual water pressure and adjusting the water pump speed and valve opening accordingly, the water pressure of the oral irrigator can be accurately controlled; this helps ensure that users enjoy stable and appropriate water pressure during use; a closed-loop control strategy is adopted, that is, the new water pressure is fed back to the system, the error is recalculated and continuously adjusted; this ensures that the system can continuously correct deviations and maintain water pressure stability; by tuning the proportional, integral and differential gains in the PID controller, the control system can respond quickly while reducing overshoot and oscillation, thereby improving the stability of the system; this ensures safety and comfort of use, and avoids oral discomfort or equipment damage that may be caused by unstable water pressure.

[0144] Example 2

[0145] See also Figure 1 As shown, this embodiment provides a water flosser voltage stabilization control system, including:

[0146] A data acquisition module is used to obtain water flosser parameter data, ambient temperature data and user usage data;

[0147] A data processing module, used for preprocessing the water flosser parameter data, the ambient temperature data and the user usage data to obtain a water flosser characteristic data set, a temperature characteristic data set and a user characteristic data set;

[0148] Analyze the influence of the temperature feature data set and the user feature data set on the water flosser feature data set, and fuse the temperature feature data set, the user feature data set and the water flosser feature data set to obtain a comprehensive feature data set;

[0149] A model training module is used to train a water pressure error prediction model for a water flosser, input a comprehensive feature data set into the water pressure error prediction model for a water flosser, and predict the water pressure error of the water flosser;

[0150] The water pressure error prediction module compares the predicted water pressure error of the water flosser with the preset water pressure error threshold of the water flosser to determine whether to generate water pressure error warning information;

[0151] The water pressure regulating module, if a water pressure error warning message is generated, executes the control instruction to adjust the speed of the water pump and the valve opening of the water flosser to stabilize the water pressure of the water flosser.

[0152] Since the electronic device introduced in this embodiment is an electronic device used to implement a method and system for stabilizing voltage control of a water flosser in the embodiment of this application, a person skilled in the art can understand the specific implementation of the electronic device of this embodiment and its various variations based on the method and system for stabilizing voltage control of a water flosser introduced in the embodiment of this application, so how the electronic device implements the method in the embodiment of this application will not be described in detail here. As long as a person skilled in the art implements the electronic device used by a method and system for stabilizing voltage control of a water flosser in the embodiment of this application, it belongs to the scope of protection of this application.

[0153] The above formulas are all dimensionless and numerical calculations. The formula is a formula for the most recent real situation obtained by collecting a large amount of data and performing software simulation. The preset parameters and thresholds in the formula are set by technicians in this field according to actual conditions.

[0154] The above is only a preferred embodiment of the present invention, and the protection scope of the present invention is not limited to the above embodiments. All technical solutions under the concept of the present invention belong to the protection scope of the present invention. It should be pointed out that for ordinary technical users in this technical field, some improvements and modifications without departing from the principle of the present invention should also be regarded as the protection scope of the present invention.

Claims

1. A method for controlling a water flosser voltage, characterized in that: include: S1, obtaining water flosser parameter data, ambient temperature data and user usage data; S2, preprocessing the water flosser parameter data, the ambient temperature data and the user usage data to obtain a water flosser characteristic data set, a temperature characteristic data set and a user characteristic data set; Analyze the influence of the temperature feature data set and the user feature data set on the water flosser feature data set, and fuse the temperature feature data set, the user feature data set and the water flosser feature data set to obtain a comprehensive feature data set; S3, training a water pressure error prediction model for the water flosser, inputting the comprehensive feature data set into the water pressure error prediction model for the water flosser, and predicting the water pressure error of the water flosser; S4, comparing the predicted water pressure error of the water flosser with a preset water pressure error threshold of the water flosser to determine whether to generate water pressure error warning information; S5. If a water pressure error warning message is generated, the control instruction is executed to adjust the rotation speed of the water pump and the valve opening of the water flosser to stabilize the water pressure of the water flosser; The water flosser parameter data includes water pressure data, water flow data and battery status data of the water flosser; the ambient temperature data includes the external ambient temperature of the water flosser; the user usage data includes the user's usage time and frequency; The method for analyzing the influence of the temperature characteristic data set and the user characteristic data set on the water flosser characteristic data set includes: The water flow rate is kept stable by a flow-pressure adjustment model; the flow-pressure adjustment model is: Among them, Q is the water flow rate; P is the water pressure; μ(T) is the viscosity of water; k is a constant; The viscosity of the water is: Where, T is the ambient temperature; μ0 is the viscosity value at the standard reference temperature; B is a constant related to fluid properties and temperature, describing the degree of influence of temperature on viscosity; Define a target water pressure P GA , the target water pressure is: P GA =(P user +α·f user )·β(T); where P user is the user's preference for water pressure; user is the frequency of use; α is the influence coefficient of user frequency of use on the target water pressure; β(T) is the ambient temperature adjustment factor; Under different ambient temperatures T, the target water pressure P of the water flosser can be adjusted according to user preferences by setting the ambient temperature adjustment factor. GA ; The ambient temperature adjustment factor is: Where β(T) is the ambient temperature adjustment factor; μ(T df ) is the room temperature reference temperature.

2. The method for controlling the voltage stabilization of a water flosser according to claim 1, characterized in that: The method for preprocessing the water flosser parameter data, the ambient temperature data and the user usage data comprises: Density clustering algorithm is used on the water flosser parameter data, ambient temperature data and user usage data to identify abnormal values ​​in the water flosser parameter data, ambient temperature data and user usage data, and the identified abnormal values ​​are removed to obtain the processed water flosser feature data set, temperature feature data set and user feature data set; The acquired water flosser feature data set, temperature feature data set and user feature data set are normalized by standard deviation and converted into a standard normal distribution with a mean of 0 and a standard deviation of 1, and finally the normalized water flosser feature data set, temperature feature data set and user feature data set are obtained.

3. The method for controlling the voltage stabilization of a water flosser according to claim 2, characterized in that: The method of fusing the temperature feature data set, the user feature data set and the water flosser feature data set to obtain a comprehensive feature data set includes: The temperature feature data set, the user feature data set and the water flosser feature data set are fused through a weighted model to form a comprehensive feature data set, and the temperature feature data set is recorded as F1, the user feature data set is recorded as F2, and the water flosser feature data set is recorded as F3; The weighted model is: Among them, ω is the weight coefficient of the temperature feature data set; γ is the weight coefficient of the user feature data set; δ is the weight coefficient of the water flosser feature data set; For the weight design of each feature data set in the weighted model, the temperature feature data set F1, the user feature data set F2 and the water flosser feature data set F3 are combined into a feature matrix X, and the corresponding target variable y; let X = [F1, F2, F3] be an n × 3 matrix; where n is the number of features; Define a set of weight combination search grids, dividing the weights ω, γ, and δ from 0 to 1 into several equally spaced points; The weighted combination search grid is: WR={(ω i ,γ j ,δ k )|ω i +γ j +δ k =1,ω i ,γ j ,δ k ∈{0,0.1,0.2,...1}}; where i is the index of the weight coefficient ω of the temperature feature dataset; j is the index of the weight coefficient γ of the user feature dataset; k is the index of the weight coefficient δ of the water flosser feature dataset; Use cross-validation to evaluate the performance of each weight combination, and use each weight combination as a data set; divide the data set into v mutually exclusive subsets; in each partition, use v-1 subsets as training sets and the remaining subset as validation set; use the training set to train the weighted model, and evaluate the performance indicators of the model by the root mean square error on the validation set; The root mean square error is: Among them, m is the number of data sets on the validation set; is the predicted value calculated by the weighted model, y i′ is the actual value of the target variable; For each weight combination, the average performance index is calculated on all partitions to obtain the average performance of the weight combination; the weight combination that makes the average performance index optimal is selected; Assume that the average performance index of the weight combination (ω,γ,δ) is E(ω,γ,δ), then the optimal weight combination is: (ω * ,γ * ,δ * ) = argmin (ω,γ,δ) E(ω,γ,δ); where (ω * ,γ * ,δ * ) is the optimal weight combination; argmin is an operator that returns the value of the variable that minimizes the objective function.

4. The method for controlling the voltage stabilization of a water flosser according to claim 3, characterized in that: The training method of the water pressure error prediction model of the water flosser includes: Divide the data set into training set, validation set and test set, train the model and evaluate the model performance; build a water pressure error prediction model for the water flosser, including input layer, hidden layer and output layer; use ReLU activation function in the hidden layer; The input layer of the water pressure error prediction model for water flossers is a historical comprehensive feature data set, and the number of neurons in the input layer should match the number of features in the historical comprehensive feature data set; the output layer of the model is the water pressure error of the water flosser, and the number of neurons in the output layer corresponds to the number of prediction targets, and the prediction value is output through one neuron; the water pressure error prediction model for water flossers is a fully connected neural network model; The mean square error is used as the loss function to measure the error between the model's predicted value and the actual value; the mean square error loss function is: Where n is the number of data sets; Y i is the actual value of data point i in the data set; is the predicted value of data point i in the data set; Use the training set data to train the model, minimize the loss function through the Adam optimizer, and tune the model's hyperparameters; The model is evaluated through the test set. When the performance of the model in the prediction task reaches the preset performance threshold, the test is stopped to obtain the water pressure error prediction model of the water flosser. The current comprehensive feature data set is input into the water pressure error prediction model of the water flosser to obtain the water pressure error of the water flosser.

5. The method for controlling the voltage stabilization of a water flosser according to claim 4, characterized in that: The method for tuning the hyperparameters of the model includes: Define the model's hyperparameter set as θ. For each hyperparameter θ u ,θ u The distribution of D u =(a,b); where D u is θ u The distribution of u The lower limit of the distribution; b is θ u The upper limit of the distribution of the preset number of iterations is t. In each iteration t, a random u Draw θ from the distribution u , forming a complete set of hyperparameter configurations Where t is the number of iterations; θ n is the hyperparameter of the model; n is the number of model hyperparameters; Use the selected hyperparameter combination θ (t) Train the model and evaluate the model performance P(V / θ) on the validation set V (t) ); Repeat the above process until the preset number of iterations t is reached, and a series of performance evaluations {P(V / θ (1) ),P(V / θ (2) ),...,P(V / θ (t) )}; Select the performance P(V / θ) from all iterations (t) )The highest hyperparameter combination θ (best) ; Use the best hyperparameter combination θ (best) The model is trained on the training set and evaluated on the test set to confirm the generalization ability of the model.

6. The method for controlling the voltage stabilization of a water flosser according to claim 5, characterized in that: The method of comparing the predicted water pressure error of the water flosser with a preset water pressure error threshold of the water flosser to determine whether to generate water pressure error warning information includes: If the predicted water pressure error of the water flosser is greater than the preset water pressure error threshold of the water flosser, a water pressure error warning message is generated; If the predicted water pressure error of the water flosser is less than or equal to the preset water pressure error threshold of the water flosser, no water pressure error warning information will be generated.

7. The method for controlling the voltage stabilization of a water flosser according to claim 6, characterized in that: The method for adjusting the rotation speed of the water pump and the valve opening of the water flosser to stabilize the water pressure of the water flosser comprises: If a water pressure error warning message is generated, the water pump speed and valve opening of the water flosser are adjusted by improving the PID algorithm to stabilize the water pressure of the water flosser; The target water pressure of the water flosser is preset to Q jc At time t′, the actual water pressure of the water flosser is measured by the pressure sensor as Q jc (t′); Calculate the error Q between the target water pressure and the actual water pressure e (t′) = Q jc -Q jc (t′); The adjustment amount of the pump speed and valve opening is controlled by improving the PID algorithm according to the error calculation. The adjustment amount is: Among them, U(t′) is the adjustment amount of the pump speed and valve opening; K p is the proportional gain; K i is the integral gain; K d is the differential gain; Q e (t′) is the error between the target water pressure and the actual water pressure; τ is the virtual time variable in the integration process; From the initial time 0 to the current time t′, the error Q e (t′) is the total accumulation over time; t′ is the time required to control the pump speed and valve opening adjustment; The adjustment value U(t′) output by the PID algorithm is used to adjust the pump speed and valve opening so that the actual water pressure gradually approaches the target water pressure. jc (t′) is fed back to the system, the error is recalculated, and continuously adjusted to form a closed-loop control; For K in PID controller p , K i and K d The value is tuned to determine the K value in the PID controller. p , K i and K d The optimal value of The preset overshoot threshold is YU, the steady-state error threshold is XU, and the stable time threshold is YU; i and K d Set to 0; gradually increase K p Stop until the overshoot of the control system reaches the preset overshoot threshold; gradually increase K i To reduce or eliminate the steady-state error, stop when it is lower than the preset steady-state error threshold; gradually increase K d Reduce overshoot and shorten stabilization time, and stop when the stabilization time of the control system reaches the preset stabilization time threshold.

8. A water flosser voltage stabilization control system, used to implement a water flosser voltage stabilization control method according to any one of claims 1 to 7, characterized in that: include: A data acquisition module is used to obtain water flosser parameter data, ambient temperature data and user usage data; A data processing module, used for preprocessing the water flosser parameter data, the ambient temperature data and the user usage data to obtain a water flosser characteristic data set, a temperature characteristic data set and a user characteristic data set; Analyze the influence of the temperature feature data set and the user feature data set on the water flosser feature data set, and fuse the temperature feature data set, the user feature data set and the water flosser feature data set to obtain a comprehensive feature data set; A model training module is used to train a water pressure error prediction model for a water flosser, input a comprehensive feature data set into the water pressure error prediction model for a water flosser, and predict the water pressure error of the water flosser; The water pressure error prediction module compares the predicted water pressure error of the water flosser with the preset water pressure error threshold of the water flosser to determine whether to generate water pressure error warning information; The water pressure regulating module, if a water pressure error warning message is generated, executes the control instruction to adjust the speed of the water pump and the valve opening of the water flosser to stabilize the water pressure of the water flosser.

Images