CN117691719A - Charging control method and system for electric toothbrush - Google Patents
Charging control method and system for electric toothbrush Download PDFInfo
- Publication number
- CN117691719A CN117691719A CN202410126358.2A CN202410126358A CN117691719A CN 117691719 A CN117691719 A CN 117691719A CN 202410126358 A CN202410126358 A CN 202410126358A CN 117691719 A CN117691719 A CN 117691719A
- Authority
- CN
- China
- Prior art keywords
- brushing
- time
- charging
- time period
- probability
- Prior art date
- Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
- Granted
Links
- 238000007600 charging Methods 0.000 title claims abstract description 262
- 238000000034 method Methods 0.000 title claims abstract description 58
- 230000001680 brushing effect Effects 0.000 claims abstract description 299
- 238000007781 pre-processing Methods 0.000 claims abstract description 10
- 230000002159 abnormal effect Effects 0.000 claims description 27
- 230000008569 process Effects 0.000 claims description 11
- 238000007621 cluster analysis Methods 0.000 claims description 7
- 230000001965 increasing effect Effects 0.000 claims description 7
- 230000007774 longterm Effects 0.000 claims description 7
- 238000012545 processing Methods 0.000 claims description 6
- 230000002708 enhancing effect Effects 0.000 claims description 5
- 230000005540 biological transmission Effects 0.000 claims description 3
- 238000010276 construction Methods 0.000 claims description 3
- 230000003247 decreasing effect Effects 0.000 claims description 3
- 238000012216 screening Methods 0.000 claims description 3
- 238000012549 training Methods 0.000 description 7
- 238000004364 calculation method Methods 0.000 description 5
- 238000012935 Averaging Methods 0.000 description 3
- 238000004891 communication Methods 0.000 description 3
- 238000012544 monitoring process Methods 0.000 description 3
- 238000004422 calculation algorithm Methods 0.000 description 2
- 230000008859 change Effects 0.000 description 2
- 238000010277 constant-current charging Methods 0.000 description 2
- 238000010586 diagram Methods 0.000 description 2
- 230000000694 effects Effects 0.000 description 2
- 238000003064 k means clustering Methods 0.000 description 2
- 238000005457 optimization Methods 0.000 description 2
- 238000012360 testing method Methods 0.000 description 2
- 230000003313 weakening effect Effects 0.000 description 2
- 230000006978 adaptation Effects 0.000 description 1
- 230000003044 adaptive effect Effects 0.000 description 1
- 230000002238 attenuated effect Effects 0.000 description 1
- 238000010280 constant potential charging Methods 0.000 description 1
- 210000004489 deciduous teeth Anatomy 0.000 description 1
- 230000007547 defect Effects 0.000 description 1
- 230000004886 head movement Effects 0.000 description 1
- 230000036541 health Effects 0.000 description 1
- 238000012423 maintenance Methods 0.000 description 1
- 238000012986 modification Methods 0.000 description 1
- 230000004048 modification Effects 0.000 description 1
- 238000012163 sequencing technique Methods 0.000 description 1
- 238000007619 statistical method Methods 0.000 description 1
Abstract
The invention belongs to the technical field of electric toothbrushes, and discloses a charging control method and a charging control system for an electric toothbrush; comprising the following steps: acquiring historical use data of a user, and preprocessing the historical use data to obtain pre-use data; constructing an improved charging naive Bayes model of the user according to the pre-use data; acquiring posterior probability when the electric toothbrush is charged, and predicting a brushing period of a user by using an improved charging naive Bayes model; presetting a charging mode according to a predicted brushing time period; adjusting the posterior probability in real time based on whether the user uses the electric toothbrush during the predicted brushing period; obtaining an adjusted posterior probability; the improved charging naive Bayes model of the user is adjusted in real time according to the adjusted posterior probability; the intelligent and personalized charging control is realized, so that the intelligent and personalized charging control device is convenient for a user to use, and can save energy and protect a battery.
Description
Technical Field
The invention relates to the technical field of electric toothbrushes, in particular to a charging control method and a charging control system of an electric toothbrush.
Background
The patent with the application publication number of CN111682623A discloses a charging control method and device of an electric toothbrush, charging equipment and the electric toothbrush, and relates to the technical field of the electric toothbrush, and the method comprises the following steps: if the prompt signal that the electric toothbrush leaves the charging base is monitored, sending a communication request for establishing communication connection to the electric toothbrush so as to establish communication with the electric toothbrush; acquiring electric quantity information of the electric toothbrush; and determining the working mode of the charging base according to the electric quantity information, and controlling the charging state of the electric toothbrush under the corresponding working mode. The technical problem that a user is not required to manually open the charging base to charge the electric toothbrush is solved, and manual operation is not required in the whole charging process.
However, the existing electric toothbrush has a simpler charging control mode, a fixed charging mode of an ordinary charger is generally adopted, the charging parameters cannot be intelligently adjusted according to the tooth brushing time and habits of different users, excessive and wasteful electric quantity remaining after charging of the users is easily caused, or the charging is insufficient, the complete tooth brushing time cannot be supported, and the actual use requirements of the users cannot be met; the existing charging control lacks algorithms like learning the use habit of a user, the charging strategy cannot be adjusted in real time to better adapt to the change of the user's brushing habit by analyzing the brushing time and pressure preference data of the user, and the charging mode adopted often lacks scientific electric quantity control, so that the user is easy to overcharge or undercharge, the service life of a battery is shortened, and the maintenance and replacement cost of the user is increased.
In view of the above, the present invention proposes a method and system for controlling charging of an electric toothbrush to solve the above-mentioned problems.
Disclosure of Invention
In order to overcome the defects in the prior art, the invention provides the following technical scheme for achieving the purposes: a method of controlling charging of an electric toothbrush, comprising: step 1, acquiring historical use data of a user, and preprocessing the historical use data to obtain pre-use data;
step 2, constructing an improved charging naive Bayes model of a user according to the pre-use data;
step 3, acquiring posterior probability when the electric toothbrush is charged, and predicting the brushing time period of a user by using an improved charging naive Bayes model;
step 4, presetting a charging mode according to a predicted brushing time period; adjusting the posterior probability in real time based on whether the user uses the electric toothbrush during the predicted brushing period; obtaining an adjusted posterior probability;
and step 5, adjusting an improved charging naive Bayesian model of the user in real time according to the adjusted posterior probability.
Further, a storage chip is arranged in the electric toothbrush; the memory chip is used for recording historical use data of not less than k years; historical usage data includes brushing data and charging data;
The brushing data includes a starting time point, a duration and a brushhead pressure of each brushing; the charging data comprises charging start time, charging end time, charging amount and usable time after charging; the starting time point, the duration, the starting time and the ending time of the charging and the time period recorded by the available time after the charging are accurate to seconds.
Further, the preprocessing method for the historical usage data comprises the following steps:
screening and deleting abnormal tooth brushing data and abnormal charging data in the historical use data recorded by the storage chip to obtain effective historical use data; the abnormal tooth brushing data comprise tooth brushing time length less than j seconds or more than h minutes, and the abnormal charging data comprise charging time length less than g minutes;
the values of j, h and g are obtained by analyzing n groups of historical use data, and the obtaining mode comprises the following steps:
selecting f pieces of representative tooth brushing time data from tooth brushing data, and performing K-means cluster analysis; setting a K value in K-means cluster analysis, namely K cluster modes; the tooth brushing time data are the starting time point and the duration of each tooth brushing;
obtaining statistical information for each cluster mode; the statistical information comprises a duration range of a clustering mode, an average duration of the clustering mode and a proportion of total data occupied by the clustering mode;
Judging that 1 to 2 clustering modes represent abnormal tooth brushing conditions of a user according to the statistical information; acquiring an upper limit h and a lower limit j of the brushing time according to abnormal brushing conditions;
the method is the same as the method for acquiring the values of j and h, and the value of g is acquired; selecting f pieces of representative charging time data from the charging data, wherein the charging time data comprises a charging start time and a charging end time;
arranging the effective historical use data into a use sequence according to the sequence of the starting time points; generating a unique ID for each brushing data in the sequence of usage, linking brushing start and end events as a brushing record identifier for subsequent processing; the brushing start and end events include a start time point, a duration, a brushhead pressure;
dividing 24 hours a day into 24 time periods according to 1 hour intervals; traversing the use sequence, and counting the number of starting time points in each time period; obtaining tooth brushing times distribution in different time periods; traversing the use sequence, counting the brushing time in each time period, and carrying out average value to obtain the average brushing time in each time period; analyzing and acquiring brush head pressure distribution in a use sequence;
Clustering the average brushing time periods of all the time periods to obtain brushing time period preference of a user; brushing time preferences include short, medium and long; statistically summarizing the brush head pressure distribution to obtain a brushing pressure preference for the user; brushing pressure preferences include strong, medium and weak;
traversing the use sequence, counting the charge starting time distribution, the average charge duration and the charge quantity distribution, and integrating and calculating the average single-charge usable duration;
the pre-use data includes a brushing number distribution, brushing time preference, brushing pressure preference, charging start time distribution, average charging time duration, charging amount distribution, and average single charging time duration available.
Further, the method for constructing the improved charging naive bayes model of the user comprises the following steps:
step 201, defining Cj to represent a j-th time period, and j=1, 2,..24; defining Bu to represent the u-th integrated brushing condition at time period Cj, u=1, 2,..9; defining P (Cj) to represent the charging prior probability of the jth time period; definition P (bu|cj) represents u comprehensive conditional probabilities at time period Cj;
step 202, traversing the charge starting time distribution in the pre-use data, counting the charge starting times in each time period, and calculating the proportion of the charge starting times to the total charge times to obtain the charge prior probability P (Cj) of each time period;
Performing conditional probability acquisition operation in each time period; obtaining the probability of brushing conditions in all time periods;
the process of the conditional probability acquisition operation includes:
counting the brushing times of different brushing time preference, and calculating the proportion of the brushing times to the total brushing times in the corresponding time period to obtain the different brushing time preference conditional probability P (B1k|Cj), wherein k=1, 2 and 3; namely, P (b11|cj), P (b12|cj), and P (b13|cj) represent conditional probabilities corresponding to short, medium, and long periods, respectively;
counting the brushing times of different brushing pressure preferences, and calculating the proportion of the brushing times to the total brushing times in the time period to obtain the conditional probability P (B2k|cj) of the different brushing pressure preferences in each time period; namely, P (b21|cj), P (b22|cj), and P (b23|cj) represent conditional probabilities corresponding to strong, medium, and weak voltages, respectively; combining P (B1k|cj) and P (B2k|cj) to obtain a brushing condition probability P (Bu|cj) of the comprehensive brushing condition;
step 203, calculating the ratio of the average usable time length of each time period to the sum of the usable time lengths of the whole use sequence according to the charge amount distribution, the average charge duration and the average usable time length after single charge, and obtaining the probability P (B4|Cj) of the long-term use condition of each time period;
Multiplying P (B4|cj) by P (Bu|cj) to obtain 9 comprehensive conditional probabilities Pu (B|cj) at each time period;
step 204, multiplying P (Cj) of each period Cj by 9 Pu (b|cj) of the corresponding period to form a naive bayes model of each period Cj; the 24-period naive Bayesian model is constructed, and the 24-period naive Bayesian model is combined, namely the improved charging naive Bayesian model of the user is formed.
Further, the method for obtaining the posterior probability includes:
when the electric toothbrush is inserted into the charging seat to start charging, acquiring the current charging start time, and determining a time period of 24 time periods Cj to which the current charging start time belongs;
reading the corresponding prior probability P (Cj) according to the current time period Cj, and reading the conditional probabilities P (B1|cj) to P (B9|cj) of 9 comprehensive brushing conditions in the Cj time period;
multiplying P (Cj) by 9 conditional probabilities P (B1|cj) to P (B9|cj) respectively to obtain posterior probability P of 9 comprehensive brushing conditions in the current time period Cj u (B|Cj)。
Further, the means for predicting a brushing session of a user using the improved charging naive bayes model comprises:
posterior probability P of 9 comprehensive brushing conditions according to current time period Cj u (B|cj), calculating the prediction probability corresponding to each brushing time preference in the current time period Cj, namely short-time prediction probability P (T1|cj), medium-time prediction probability P (T2|cj) and long-time prediction probability P (T3|cj);
the method is the same as that of obtaining the prediction probability corresponding to the brushing time preference, and the prediction probability corresponding to each brushing pressure preference, namely the strong-pressure prediction probability P (F1|Cj), the medium-pressure prediction probability P (F2|Cj) and the weak-pressure prediction probability P (F3|Cj) in the current time period Cj are calculated;
the short-time prediction probability P (T1|cj), the middle-time prediction probability P (T2|cj), the long-time prediction probability P (T3|cj), the strong-pressure prediction probability P (F1|cj), the middle-pressure prediction probability P (F2|cj) and the weak-pressure prediction probability P (F3|cj) are weighted and counted to obtain the comprehensive prediction probability P (Sj|cj) of each brushing time period in the current time period Cj.
Further, the preset charging mode includes:
sorting and dividing the first R time periods with the maximum value of the prediction probability, setting the time period with the maximum value of the prediction probability in the R time periods as a first tooth brushing time period, and pushing the second highest value of the prediction probability in the R time periods as a second time period;
Judging whether the electric quantity required by the primary brushing time period is met or not according to the current electric quantity condition of the electric toothbrush; if the current electric quantity meets the electric quantity required by the primary brushing time period, the preset charging mode is a low-speed charging mode; if the current electric quantity is insufficient to meet the electric quantity required by the primary brushing time period, calculating the charge quantity to be met, wherein the preset charging mode is a quick charging mode;
judging the electric quantity demand of the secondary time period on the premise of meeting the electric quantity demand of the primary brushing time period, if the current electric quantity does not meet the electric quantity demand of the secondary time period, presetting a charging mode of the secondary time period as a quick charging mode, and presetting the ending time of the quick charging mode as p minutes before the primary brushing time period starts;
according to the electric quantity requirement of the user in the tooth brushing time period, an elevator-shaped charging scheme is designed; the elevator-shaped charging scheme is that the charging mode is gradually increased from a low-speed charging mode to a rapid charging mode, and gradually decreased to the low-speed charging mode before the last tooth brushing time period.
Further, the means for adjusting the posterior probability in real time includes:
if the user starts brushing teeth with the electric toothbrush within a certain predicted brushing time period of the user; confirm that the user has brushed for a predicted period of time; enhancing posterior probabilities corresponding to brushing duration preferences and pressure preferences of the time period;
If the user does not brush his or her teeth with the electric toothbrush within any of the predicted user's brushing periods, the posterior probability corresponding to the duration preference and pressure preference of the predicted user's brushing period is reduced.
Further, the real-time adjustment mode of the improved charging naive bayes model of the user includes:
inputting the adjusted posterior probability feedback into an improved charging naive Bayes model, and covering the original posterior probability data;
traversing the adjusted posterior probability, and updating the brushing condition probability under each time period in the model according to the brushing time preference of short time, medium time and long time and the brushing pressure preference of strong pressure, medium pressure and weak pressure, namely improving P (Bu|Cj) in the charging naive Bayes model; multiplying the updated brushing condition probability of each time period with the using long-strip probability of the corresponding time period; obtaining updated comprehensive conditional probability;
according to the Bayesian theorem, the updated comprehensive conditional probability is combined with the charging prior probability P (Cj), and a naive Bayesian model of each time period in the current stage is obtained through recalculation;
reconstructing the recalculated naive Bayes model of 24 time periods into an improved charging naive Bayes prediction model of the user.
A charge control system for an electric toothbrush for implementing the charge control method for an electric toothbrush, comprising: the data acquisition processing module is used for acquiring historical use data of users and preprocessing the historical use data to obtain pre-use data;
the model construction module is used for constructing an improved charging naive Bayes model of the user according to the pre-use data;
a time period prediction module for obtaining posterior probability when the electric toothbrush is charged and predicting a brushing time period of the user by using the improved charging naive bayes model;
the charging mode and adjusting module is used for presetting a charging mode according to the predicted tooth brushing time period; adjusting the posterior probability in real time based on whether the user uses the electric toothbrush during the predicted brushing period; obtaining an adjusted posterior probability;
the model adjusting module is used for adjusting an improved charging naive Bayesian model of a user in real time according to the adjusted posterior probability; all the modules are connected in a wired and/or wireless mode, so that data transmission among the modules is realized.
The charging control method and the charging control system of the electric toothbrush have the technical effects and advantages that:
According to the invention, the reasonable charging strategy is formulated by predicting the brushing time of the user, the elevator-shaped charging mode with combination of quick charging and slow charging is adopted, the use requirement is met, the battery is protected, the optimal charging scheme can be formulated by establishing a model for accurately predicting the brushing time of the user, the model optimization strategy is continuously adjusted by using user feedback to adapt to the user change, the intelligent and personalized charging control is realized, the use of the user is facilitated, the energy is saved, and the battery is protected.
Drawings
FIG. 1 is a schematic diagram of a method of controlling charging of an electric toothbrush according to the present invention;
fig. 2 is a schematic diagram of a charge control system of an electric toothbrush according to the present invention.
Detailed Description
The following description of the embodiments of the present invention will be made clearly and completely with reference to the accompanying drawings, in which it is apparent that the embodiments described are only some embodiments of the present invention, but not all embodiments. All other embodiments, which can be made by those skilled in the art based on the embodiments of the invention without making any inventive effort, are intended to be within the scope of the invention.
Example 1
Referring to fig. 1, a charging control method of an electric toothbrush according to the present embodiment includes:
Step 1, acquiring historical use data of a user, and preprocessing the historical use data to obtain pre-use data;
step 2, constructing an improved charging naive Bayes model of a user according to the pre-use data;
step 3, acquiring posterior probability when the electric toothbrush is charged, and predicting the brushing time period of a user by using an improved charging naive Bayes model;
step 4, presetting a charging mode according to a predicted brushing time period; adjusting the posterior probability in real time based on whether the user uses the electric toothbrush during the predicted brushing period; obtaining an adjusted posterior probability;
and step 5, adjusting an improved charging naive Bayesian model of the user in real time according to the adjusted posterior probability.
The method for acquiring the historical usage data of the user comprises the following steps:
a storage chip is arranged in the electric toothbrush; the memory chip is used for recording historical use data of not less than k years;
the memory chip adopts a nonvolatile memory, such as a flash memory, and even if power is off, the data is not lost; the capacity of the memory chip can store data for at least more than k years; the storage chip writes data related to tooth brushing and charging in real time through a controller in the electric toothbrush; the data is accurate to the order of seconds.
Historical usage data includes brushing data and charging data;
the brushing data includes a starting time point, a duration and a brushhead pressure of each brushing; the charging data comprises charging start time, charging end time, charging amount and usable time after charging; the starting time point, the duration, the starting time and the ending time of charging and the data recorded by the usable time after charging are accurate to the second level;
preferably, the historical use data is transmitted to a preset App of a user mobile phone in a low-power consumption mode of a Bluetooth/wireless local area network and backed up to a cloud.
It should be noted that the starting time point is the accurate time for starting brushing teeth with the electric toothbrush by the user, and is accurate to the second level; the duration is recorded as the tooth brushing start time and the tooth brushing end time, and the tooth brushing duration can be calculated by subtracting the start time from the end time and takes seconds as a unit; the pressure of the brush head is obtained by internally arranging a pressure sensor in the brush head of the electric toothbrush, detecting the pressure value when the brush head is contacted with teeth in real time, and recording the pressure value in a digital quantity format; the charging start time is the accurate time for recording the charging start when a user inserts into the charging port to charge; the charging ending time is the accurate time for recording the ending of the charging when the user pulls out the charging port to end the charging; the charge amount is obtained by detecting and integrating the charging current in the charging period in real time through a sensor which is arranged in the electric toothbrush and can detect the charging current; the obtaining of the usable time length after charging is to obtain the average time length of which the 1mAh electric quantity can be supported for use according to the historical charging data statistics, and calculate the usable time length after charging by combining the charging amount of each time.
The method for preprocessing the historical usage data comprises the following steps:
screening and deleting abnormal tooth brushing data and abnormal charging data in the historical use data recorded by the storage chip to ensure the validity of the data and obtain effective historical use data; the abnormal tooth brushing data comprise tooth brushing time length less than j seconds or more than h minutes, and the abnormal charging data comprise charging time length less than g minutes; j. the values of h and g are obtained by analyzing n sets of historical usage data.
j. The acquisition modes of the values of h and g comprise:
selecting f pieces of representative tooth brushing time data from tooth brushing data, and performing K-means cluster analysis; setting a K value in K-means cluster analysis, namely K cluster modes; for example, if the K value is set to 6, the K value is a cluster mode of 6 different tooth brushing time periods; the tooth brushing time data are the starting time point and the duration of each tooth brushing;
obtaining statistical information for each cluster mode; the statistical information comprises a duration range of a clustering mode, an average duration of the clustering mode and a proportion of total data occupied by the clustering mode;
judging that 1 to 2 clustering modes represent abnormal tooth brushing conditions of a user according to the statistical information; acquiring an upper limit h and a lower limit j of the brushing time according to abnormal brushing conditions;
Selecting f pieces of representative charging time data from the charging data, wherein the charging time data comprises a charging start time and a charging end time; the values of g are obtained in the same manner as the values of j and h.
It should be explained that the representative time data may be understood as data for selecting different time periods, for example, a high frequency period in the morning and evening and a non-high frequency period, to reflect the situation that the user brushes and charges at different times; selecting data representative of brushing and charging durations, for example, selecting data with a shorter part of duration and selecting data with a longer part of duration to reflect the distribution of brushing and charging durations of the user; the amount of data selected is sufficiently large that the more data is selected, the more representative of the overall situation.
Carrying out K-means cluster analysis specifically, and taking the starting time point and the duration of tooth brushing of a user as sample data;
determining a clustering number K when K-means clustering is carried out, and determining the value of the clustering number K to determine the optimal K value through a plurality of test comparison results; for example, initially setting k=5 trial clustering effects; subsequently, gradually determining an optimal K value in the test process; normalizing the starting time point and the duration of the tooth brushing of the user to the same order of magnitude; randomly selecting K pieces of tooth brushing data from the sample data as initial K clustering centers;
Calculating the distance between each sample data and K clustering centers, and distributing each sample to the closest cluster; obtaining a distribution result; re-calculating the cluster center of each cluster (namely the average value of all samples in the cluster) according to the distribution result; the above calculation is repeated until the cluster center is no longer changed.
The method for acquiring the statistical information comprises the following steps:
counting the number of sample data of each of the K clustering modes; finding out the minimum value and the maximum value of the tooth brushing time length in all sample data of each of the K clustering modes, and obtaining a time length range;
averaging the brushing time length of all sample data of each of the K clustering modes to obtain the average time length corresponding to each of the K clustering modes;
dividing a day into different time periods, such as an early-late peak period and an off-peak period, and counting the number of samples in the cluster mode in the different time periods;
drawing a distribution chart or a frequency distribution chart of the tooth brushing starting time points of all sample data of each of the K clustering modes; and calculating the proportion of all sample data of each of the K clustering modes to the total sample number.
The judging mode of abnormal tooth brushing condition comprises the following steps:
judging the clustering mode with too short brushing time as abnormal brushing condition; the clustering mode with overlong tooth brushing time length is also judged to be abnormal tooth brushing condition; if the tooth brushing times in a certain clustering mode are mainly concentrated in abnormal time in the early morning or at night, judging that the abnormal tooth brushing condition of a user is achieved; if the number of samples in a certain cluster mode is small, for example, less than 5%, the abnormal brushing condition is judged.
For an abnormal clustering mode with too short brushing time, finding the maximum value of the brushing time, and adding a tolerance time as the lower limit j of the brushing time; for example, if a certain cluster mode has a duration of 30 seconds at maximum, the lower limit j is set to 45 seconds; for an abnormal clustering mode with overlong tooth brushing time, finding the minimum value of the tooth brushing time, and subtracting a tolerance time to serve as the upper limit h of the tooth brushing time; for example, if a certain clustering pattern has a shortest time of 5 minutes and 30 seconds, the upper limit h is set to 5 minutes.
The process for determining the tolerance time comprises the following steps:
observing the specific duration distribution condition of the clustering mode of abnormal tooth brushing duration and distinguishing the clustering mode from the boundary of normal tooth brushing to determine reasonable tolerance time; in a preferred embodiment, the tolerance time is set to 10-20 seconds.
Arranging the effective historical use data into a use sequence according to the sequence of the starting time points; generating a unique ID for each brushing data in the sequence of usage, linking brushing start and end events as a brushing record identifier for subsequent processing; the brushing start and end events include a start time point, a duration, a brushhead pressure;
24 hours a day are divided into 24 time periods at 1 hour intervals, for example, 7-8 early, 20-21 late, etc.; traversing the use sequence, and counting the number of starting time points in each time period; obtaining tooth brushing times distribution in different time periods; traversing the use sequence, counting the brushing time in each time period, and carrying out average value to obtain the average brushing time in each time period;
Analyzing and acquiring brush head pressure distribution in a use sequence; specifically, dividing the brush head pressure into a plurality of range grades; traversing the use sequence, and counting the occurrence times of each range grade, namely brushing times when the pressure of the brush head falls in a pressure interval; calculating the probability of the total number of samples in each range class in all brushing times; and drawing a frequency distribution map or a probability density distribution map of the pressure of the brush head, and visually displaying the distribution conditions of different pressure intervals.
Clustering the average brushing time periods of all the time periods to obtain brushing time period preference of a user; brushing time preferences include short, medium and long; the clustering mode for obtaining the preference of the tooth brushing time length is the same as the K-means clustering mode for obtaining the clustering mode;
statistically summarizing the brush head pressure distribution to obtain a brushing pressure preference for the user; brushing pressure preferences include strong, medium and weak;
and traversing the use sequence, counting the charge starting time distribution, the average charge duration and the charge quantity distribution, and integrating and calculating the average single-charge usable duration.
The process of statistically summarizing the brush head pressure distribution is to divide the pressure into 3 level intervals, namely strong pressure, medium pressure and weak pressure, according to a frequency distribution map or a probability density distribution map of the brush head pressure;
The charging start time distribution is the number or the duty ratio of charging start time points in each time period; calculating a charging duration in the use sequence, wherein the charging duration is the charging end time minus the charging start time; in each time period, calculating the charging duration of all the charging data to obtain the average charging duration of different time periods; for each charging duration in the usage sequence, the charging amount thereof is correspondingly acquired, and preferably, the charging amount is accurate to 0.1 milliampere hour;
in each time period, measuring the average value of all the charging amounts to obtain the average charging amounts in different time periods; and correspondingly dividing the average charging duration and the average charging amount to obtain the average single-charging usable duration in different time periods.
The pre-use data includes a brushing number distribution, brushing time preference, brushing pressure preference, charging start time distribution, average charging time duration, charging amount distribution, and average single charging time duration available.
Further, the method for constructing the improved charging naive Bayesian model of the user comprises the following steps:
step 201, defining Cj to represent a j-th time period, and j=1, 2,..24; defining Bu to represent the u-th integrated brushing condition at time period Cj, u=1, 2,..9; defining P (Cj) to represent the charging prior probability of the jth time period; definition P (bu|cj) represents u comprehensive conditional probabilities at time period Cj;
Step 202, traversing the charge starting time distribution in the pre-use data, counting the charge starting times in each time period, and calculating the proportion of the charge starting times to the total charge times to obtain the charge prior probability P (Cj) of each time period;
performing conditional probability acquisition operation in each time period; the brushing condition probabilities for all time periods were obtained.
The process of the conditional probability acquisition operation includes:
counting the brushing times of different brushing time preference (short time, medium time and long time), and calculating the proportion of the brushing times to the total brushing times in the corresponding time period to obtain 3 conditional probabilities P (B1k|cj) of different brushing time preference (short time, medium time and long time) in each time period, wherein k=1, 2 and 3; namely, P (b11|cj), P (b12|cj), and P (b13|cj) represent conditional probabilities corresponding to short, medium, and long periods, respectively;
counting the brushing times of different brushing pressure preferences, and calculating the proportion of the brushing times to the total brushing times in the time period to obtain the conditional probability P (B2k|cj) of the different brushing pressure preferences in each time period; namely, P (b21|cj), P (b22|cj), and P (b23|cj) represent conditional probabilities corresponding to strong, medium, and weak voltages, respectively;
combining P (B1k|cj) and P (B2k|cj) to obtain a brushing condition probability P (Bu|cj) of the comprehensive brushing condition;
Step 203, calculating the ratio of the average usable time length of each time period to the sum of the usable time lengths of the whole use sequence according to the charge amount distribution, the average charge duration and the average usable time length after single charge, and obtaining the probability P (B4|Cj) of the long-term use condition of each time period.
Specifically, for each period Cj, calculating an average charge amount for the period according to the charge amount distribution statistics; counting according to historical data to obtain average duration of about 1 milliamp-hour electric quantity which can be used in a supporting way, and marking the average duration as a; dividing the average charge amount of the time period Cj by a to obtain the average usable time length of the time period, and marking the average usable time length as b; calculating average usable time length of all time periods Cj; calculating the sum of the usable time lengths of all the time periods Cj, and recording the sum as a total usable time length B; for each time period Cj, calculating the ratio of the average usable time length bj to the total usable time length B, namely P (B4|cj); the history data is electric toothbrush usage data of a user over a period of time (e.g., 1 year), including a start time, an end time, a charge start time, a charge end time, and a charge amount per charge of each brushing.
P (b4|cj) is multiplied by P (bu|cj) to obtain 9 comprehensive conditional probabilities Pu (b|cj) for each time period.
The method for combining P (b1k|cj) and P (b2k|cj) includes:
p (B1k|cj) and P (B2k|cj) under the same time period Cj are correspondingly combined, namely, the probability of combining the P (B1k|cj) and the P (B2k|cj) under the same comprehensive tooth brushing condition is combined; for example, at time period C1, P (B11|C1) for short duration preference is combined with P (B21|C1) for strong pressure preference to P1 (B|C1), representing the probability of short duration and strong pressure brushing at time period C1; the probability P2 (B|C1) of the medium time and the strong pressure is obtained by combining P (B12|C1) and P (B21|C1); the probabilities of long-term and strong pressures P3 (B|C1) are combined by P (B13|C1) and P (B21|C1); similarly, the probabilities Pu (B|Cj) for the 9 combined brushing conditions over time period Cj are obtained by combining.
Step 204, multiplying P (Cj) of each period Cj by 9 Pu (b|cj) of the corresponding period to form a naive bayes model of each period Cj; the 24-period naive Bayesian model is constructed, and the 24-period naive Bayesian model is combined, namely the improved charging naive Bayesian model of the user is formed.
In a preferred embodiment, the constructed improved charged naive Bayes model is stored in Json format data in a memory chip of the electric toothbrush, so that the data can be read and used in subsequent prediction.
Further, the method for obtaining the posterior probability includes:
when the electric toothbrush is inserted into the charging seat to start charging, acquiring the current charging start time, determining a time period in 24 time periods Cj to which the current charging start time belongs, for example, determining that the current time period is C7 when the current time period is 7 a.m. for 20 minutes;
reading the corresponding prior probability P (Cj) according to the current time period Cj, and reading the brushing condition probability P (Bu|cj) of 9 comprehensive brushing conditions in the Cj time period;
multiplying P (Cj) by 9 conditional probabilities P (B1|cj) to P (B9|cj) respectively to obtain posterior probability P of 9 comprehensive brushing conditions in the current time period Cj u (b|cj); posterior probability P u (B|cj) represents the probability that the user will brush with different duration and pressure for the current time period Cj; p (P) u (B|cj) will be used to predict the brushing session for the user subsequently.
Further, the method for predicting a brushing session of a user using the improved charging naive bayes model comprises:
posterior probability P of 9 comprehensive brushing conditions according to current time period Cj u (B|cj), the prediction probability corresponding to each brushing time preference in the current time period Cj, namely, short-time prediction probability P (T1|cj), medium-time prediction probability P (T2|cj) and long-time prediction probability P (T3|cj) are calculated.
The calculation mode of the prediction probability corresponding to each brushing time preference comprises the following steps:
the posterior probability is grouped correspondingly according to the preference of brushing time, wherein the corresponding grouping is that the posterior probability with short time is grouped into a group with short time; in particular, wherein P 1 (B|Cj)、P 2 (B|Cj)、P 3 (B|cj) represents a short time; p (P) 4 (B|Cj)、P 5 (B|Cj)、P 6 (B|cj) represents; p (P) 7 (B|Cj)、P 8 (B|Cj)、P 9 (B|cj) represents a long term;
averaging the posterior probability belonging to the short time to obtain the prediction probability P (T1|Cj) of the short time in the current time period Cj;
averaging the posterior probability belonging to the middle time to obtain the prediction probability P (T2|Cj) of the middle time in the current time period Cj;
the posterior probability belonging to a long term is averaged to obtain the predicted probability P (t3|cj) for the long term in the current time period Cj.
The method is the same as the mode of the prediction probability corresponding to each brushing time preference, and the prediction probability corresponding to each brushing pressure preference, namely the strong prediction probability P (F1|cj), the medium prediction probability P (F2|cj) and the weak prediction probability P (F3|cj) in the current time period Cj are calculated;
weighting and counting the short-time prediction probability P (T1|cj), the middle-time prediction probability P (T2|cj), the long-time prediction probability P (T3|cj), the strong-pressure prediction probability P (F1|cj), the middle-pressure prediction probability P (F2|cj) and the weak-pressure prediction probability P (F3|cj) to obtain the comprehensive prediction probability P (Sj|cj) of each brushing time period in the current time period Cj; namely P (s1|cj), P (s2|cj)..p (s24|cj).
The calculation mode of the weighted statistics is as follows:
adding the short, medium and long prediction probabilities of the time period Cj, and multiplying the short, medium and long prediction probabilities by the weight w11 of the time preference in the time period Cj obtained through experimental training to obtain a time duration product value;
adding the prediction probabilities of strong pressure, medium pressure and weak pressure in the time period Cj, and multiplying the prediction probabilities with the weight w12 of the brushing pressure preference obtained through experimental training in the time period Cj to obtain a pressure product value;
accumulating and summing the duration product value and the pressure product value to obtain the comprehensive prediction probability of the time period Cj;
the same calculation is carried out on each time period S1 to S24, and the comprehensive prediction probability P (Sj|Cj) of each time period is obtained;
sequencing the predictive probabilities P (S1|cj) to P (S24|cj) of 24 time periods, and selecting the first R time periods with the maximum predictive probability value as the brushing time periods of the user;
preferably, after each prediction is completed, the actual tooth brushing data is used for feeding back and adjusting the numerical values of the weights w11 and w12, so that the prediction accuracy of the model is continuously optimized.
It should be noted that, the process of acquiring the weights w11 and w12 includes:
the method comprises the steps of firstly, collecting a large amount of user tooth brushing data as training data, wherein the training data comprise tooth brushing time, tooth brushing pressure and corresponding tooth brushing time period;
Secondly, carrying out statistical analysis on the training data to obtain distribution proportions of different brushing time preference and brushing pressure preference in each time period;
thirdly, setting the values of initial weights w11 and w12, for example, the values of w11 and w12 are taken;
fourthly, predicting tooth brushing time in different time periods by using training data and weights w11 and w12, and calculating errors of a prediction result and actual tooth brushing time of the training data;
fifthly, adjusting the values of w11 and w12 through a back propagation algorithm to minimize a prediction error; repeating the fourth step and the fifth step until the error meets the requirement or the number of turns reaches a set value; the error meeting requirement and the set value can be determined according to actual conditions;
sixth, values of w11 and w12 are obtained to minimize the prediction error, and recorded as final weights.
Further, the mode of presetting the charging mode includes:
sorting and dividing the first R time periods with the maximum value of the prediction probability, setting the time period with the maximum value of the prediction probability in the R time periods as a first tooth brushing time period, and pushing the second highest value of the prediction probability in the R time periods as a second time period;
judging whether a charging mode of the electric quantity required by the primary brushing time period is met or not according to the current electric quantity condition of the electric toothbrush; if the current electric quantity meets the electric quantity required by the primary brushing time period, the preset charging mode is a low-speed charging mode; for maintaining battery health of the rechargeable toothbrush; if the current electric quantity is insufficient to meet the electric quantity required by the primary brushing time period, calculating the charge quantity to be met, wherein the preset charging mode is a quick charging mode;
Judging the electric quantity demand of the secondary time period on the premise of meeting the electric quantity demand of the primary tooth brushing time period, and adding the secondary time period into a quick charging mode if the current electric quantity does not meet the electric quantity demand of the secondary time period; the end time of the quick charge mode is preset to be p minutes before the first brushing time period starts, so that the automatic switching to low-speed charge is prevented from being performed by overcharging;
according to the electric quantity requirement of the user in the tooth brushing time period, an elevator-shaped charging scheme is designed; the elevator-shaped charging scheme is that the charging mode is gradually increased from the low-speed charging mode to the fast charging mode, and then gradually decreased to the low-speed charging mode before the last brushing period.
Preferably, the charging condition of the user is monitored in real time, if the user interrupts charging, the available time corresponding to the residual electric quantity is reevaluated, and the scheme required by charging is updated.
Preferably, the means for calculating the amount of charge to be satisfied includes:
according to the historical use data of the user, counting the average brushing time length of the first brushing time period, and recording as t1; inquiring a product manual of the electric toothbrush to obtain rated power P of the product; according to the brushing habit of a user, estimating the average power P in the brushing process, wherein the average power P is generally slightly smaller than the rated power P; multiplying the average brushing time t1 by the average power p to obtain the total energy E1 required by the primary brushing time period; inquiring the battery capacity C of the electric toothbrush and the rated voltage U of the battery;
The total energy E1 is converted into the unit of electric quantity Q1, wherein the unit is q1=E1/(U×3600) and mAh; acquiring the residual electric quantity of the current electric toothbrush as Q2; the charge amount q=q1-Q2 to be satisfied is calculated.
In a preferred embodiment, the implementation of the low-speed charging mode and the high-speed charging mode includes:
the output voltage of the charger is set to be about 0.8 times of the rated voltage of the battery, for example, the rated voltage of the battery is 3.7V, and then the output voltage of the charger is set to be about 3V; the charger output current was set to about 0.5C, C being the battery capacity, and 0.5C representing 50% of the battery capacity. For example, when the battery capacity is 2000mAh, the output current is set to about 1000 mA; adopting a constant-voltage constant-current charging mode, firstly carrying out constant-current charging to a set voltage value, and then carrying out constant-voltage charging; the temperature of the battery is properly detected during charging and controlled below 40 ℃ to realize a low-speed charging mode;
setting the charger output voltage to a battery rated voltage, for example 3.7V; the charger output current is set to about 1C, i.e., 100% of the battery capacity; the method adopts a voltage-stabilizing fixed current mode to rapidly charge; monitoring the temperature of the battery during charging, and if the temperature exceeds 45 ℃ during charging, reducing the current to control the temperature; after the charge quantity reaches a set value, the quick charge mode is exited; a fast charge mode is achieved.
Further, the means for adjusting the posterior probability in real time includes:
if the user starts brushing teeth with the electric toothbrush within a certain predicted brushing time period of the user; (judged by the brush head movement and the pressure sensor), confirming that the user brushes teeth for a predicted period of time; enhancing posterior probabilities corresponding to brushing duration preferences and pressure preferences of the time period; for example, if the posterior probability of the original short time is 0.3, the time is increased to 0.35; the posterior probability of the original strong pressure is 0.2, and then the original strong pressure is enhanced to 0.25;
if the user does not brush his or her teeth with the electric toothbrush within any predicted brushing period of the user, the posterior probability corresponding to the duration preference and the pressure preference of the predicted brushing period of the user is reduced, for example, the prior short posterior probability is 0.3, which is reduced to 0.25; the posterior probability of the original strong pressure is 0.2, and the posterior probability is reduced to 0.15;
preferably, the posterior probabilities of different durations and pressure preferences are more finely enhanced or attenuated depending on the specific duration and pressure of the user brushing.
Specifically, the brushing time period is divided into a plurality of ranges, such as 0-1 min, 1-2 min, 2-3 min, etc.; after each brushing, determining the type of brushing duration preference according to the range of the brushing duration, and enhancing the middle-time brushing posterior probability of the corresponding brushing period, wherein the increasing amplitude is in direct proportion to the time exceeding the lower limit of the range, for example, exceeding 1 minute for 20 seconds, and increasing by 0.02; while reducing the posterior probability of the brushing for a period other than the time period, e.g., reducing the posterior probability for short and long periods by an amount proportional to the time period exceeding the lower limit.
Monitoring and recording the pressure of the brush head of each brushing of a user, and judging whether the pressure of the brush head belongs to strong pressure, medium pressure or weak pressure; enhancing posterior probability corresponding to brushing pressure preference according to the judging result, and weakening posterior probability not corresponding to the pressure preference; the amplitude of the enhancement and the weakening is adjusted according to the difference of the pressure;
after a period of monitoring and adjustment, the posterior probability distribution can more accurately reflect the brushing habit of the user.
Further, the real-time adjustment method of the improved charging naive bayes model of the user includes:
inputting the adjusted posterior probability feedback into an improved charging naive Bayes model, and covering the original posterior probability data;
traversing the adjusted posterior probability, and updating the brushing condition probability under each time period in the model according to the brushing time preference of short time, medium time and long time and the brushing pressure preference of strong pressure, medium pressure and weak pressure, namely improving P (Bu|Cj) in the charging naive Bayes model; multiplying the updated brushing condition probability of each time period with the using long-strip probability of the corresponding time period; obtaining updated comprehensive conditional probability;
according to the Bayesian theorem, the updated comprehensive conditional probability is combined with the charging prior probability P (Cj), and the naive Bayesian model of each time period in the current stage is obtained through recalculation.
Reconstructing the recalculated naive Bayes model of 24 time periods into an improved charging naive Bayes prediction model of the user.
According to the method, the device and the system, reasonable charging strategies are formulated for predicting the brushing time of the user, a quick charging and slow charging combined elevator-shaped charging mode is adopted, the use requirement is met, the battery is protected, an optimal charging scheme can be formulated by accurately predicting the brushing time of the user, the model optimization strategies are continuously adjusted by user feedback to adapt to user changes, intelligent and personalized charging control is achieved, the user can use conveniently, energy sources can be saved, and the battery is protected.
Example 2
Referring to fig. 2, the embodiment is not described in detail in embodiment 1, and in the process of controlling the charging of the electric toothbrush, a charging control method of the electric toothbrush is implemented by a charging control program of the electric toothbrush, and a charging control system of the electric toothbrush includes: the data acquisition processing module is used for acquiring historical use data of users and preprocessing the historical use data to obtain pre-use data;
the model construction module is used for constructing an improved charging naive Bayes model of the user according to the pre-use data;
A time period prediction module for obtaining posterior probability when the electric toothbrush is charged and predicting a brushing time period of the user by using the improved charging naive bayes model;
the charging mode and adjusting module is used for presetting a charging mode according to the predicted tooth brushing time period; adjusting the posterior probability in real time based on whether the user uses the electric toothbrush during the predicted brushing period; obtaining an adjusted posterior probability;
the model adjusting module is used for adjusting an improved charging naive Bayesian model of a user in real time according to the adjusted posterior probability; all the modules are connected in a wired and/or wireless mode, so that data transmission among the modules is realized.
In addition, according to embodiments of the present application, a charging control method of an electric toothbrush, the process described in the drawings may be implemented as a computer software program. For example, the present application provides a non-transitory machine-readable storage medium storing machine-readable instructions executable by a processor to perform instructions corresponding to the method steps provided herein, of course, the architecture shown in the figures of a method of controlling the charging of an electric toothbrush is merely exemplary, and when implementing different devices, adaptive selection or adjustment may be made as desired.
The above formulas are all formulas with dimensionality removed and numerical calculation, the formulas are formulas with the latest real situation obtained by software simulation through collecting a large amount of data, and preset parameters and threshold selection in the formulas are set by those skilled in the art according to the actual situation.
The above description is only a preferred embodiment of the present invention, and the protection scope of the present invention is not limited to the above examples, and all technical solutions belonging to the concept of the present invention belong to the protection scope of the present invention. It should be noted that modifications and adaptations to those skilled in the art without departing from the principles of the present invention are intended to be comprehended within the scope of the present invention.
Claims (10)
1. A charging control method of an electric toothbrush, comprising:
step 1, acquiring historical use data of a user, and preprocessing the historical use data to obtain pre-use data;
step 2, constructing an improved charging naive Bayes model of a user according to the pre-use data;
step 3, acquiring posterior probability when the electric toothbrush is charged, and predicting the brushing time period of a user by using an improved charging naive Bayes model;
Step 4, presetting a charging mode according to a predicted brushing time period; adjusting the posterior probability in real time based on whether the user uses the electric toothbrush during the predicted brushing period; obtaining an adjusted posterior probability;
and step 5, adjusting an improved charging naive Bayesian model of the user in real time according to the adjusted posterior probability.
2. The method for controlling charge of an electric toothbrush according to claim 1, wherein the electric toothbrush has a built-in memory chip; the memory chip is used for recording historical use data of not less than k years; historical usage data includes brushing data and charging data;
the brushing data includes a starting time point, a duration and a brushhead pressure of each brushing; the charging data includes a charging start time, a charging end time, a charging amount, and a usable time period after charging.
3. The method of claim 2, wherein the preprocessing the historical usage data comprises:
screening and deleting abnormal tooth brushing data and abnormal charging data in the historical use data recorded by the storage chip to obtain effective historical use data; the abnormal tooth brushing data comprise tooth brushing time length less than j seconds or more than h minutes, and the abnormal charging data comprise charging time length less than g minutes;
The values of j, h and g are obtained by analyzing n groups of historical use data, and the obtaining mode comprises the following steps:
selecting f pieces of representative tooth brushing time data from tooth brushing data, and performing K-means cluster analysis; setting a K value in K-means cluster analysis, namely K cluster modes; the tooth brushing time data are the starting time point and the duration of each tooth brushing;
obtaining statistical information for each cluster mode; the statistical information comprises a duration range of a clustering mode, an average duration of the clustering mode and a proportion of total data occupied by the clustering mode;
judging that 1 to 2 clustering modes represent abnormal tooth brushing conditions of a user according to the statistical information; acquiring an upper limit h and a lower limit j of the brushing time according to abnormal brushing conditions;
the method is the same as the method for acquiring the values of j and h, and the value of g is acquired; selecting f pieces of representative charging time data from the charging data, wherein the charging time data comprises a charging start time and a charging end time;
arranging the effective historical use data into a use sequence according to the sequence of the starting time points; generating a unique ID for each brushing data in the sequence of usage, linking brushing start and end events as a brushing record identifier for subsequent processing; the brushing start and end events include a start time point, a duration, a brushhead pressure;
Dividing 24 hours a day into 24 time periods according to 1 hour intervals; traversing the use sequence, and counting the number of starting time points in each time period; obtaining tooth brushing times distribution in different time periods; traversing the use sequence, counting the brushing time in each time period, and carrying out average value to obtain the average brushing time in each time period; analyzing and acquiring brush head pressure distribution in a use sequence;
clustering the average brushing time periods of all the time periods to obtain brushing time period preference of a user; brushing time preferences include short, medium and long; statistically summarizing the brush head pressure distribution to obtain a brushing pressure preference for the user; brushing pressure preferences include strong, medium and weak;
traversing the use sequence, counting the charge starting time distribution, the average charge duration and the charge quantity distribution, and integrating and calculating the average single-charge usable duration;
the pre-use data includes a brushing number distribution, brushing time preference, brushing pressure preference, charging start time distribution, average charging time duration, charging amount distribution, and average single charging time duration available.
4. The method of claim 3, wherein the means for constructing the improved charging naive bayes model of the user comprises:
Step 201, defining Cj to represent a j-th time period, and j=1, 2,..24; defining Bu to represent the u-th integrated brushing condition at time period Cj, u=1, 2,..9; defining P (Cj) to represent the charging prior probability of the jth time period; definition P (bu|cj) represents u comprehensive conditional probabilities at time period Cj;
step 202, traversing the charge starting time distribution in the pre-use data, counting the charge starting times in each time period, and calculating the proportion of the charge starting times to the total charge times to obtain the charge prior probability P (Cj) of each time period;
performing conditional probability acquisition operation in each time period; obtaining the probability of brushing conditions in all time periods;
the process of the conditional probability acquisition operation includes:
counting the brushing times of different brushing time preference, and calculating the proportion of the brushing times to the total brushing times in the corresponding time period to obtain the different brushing time preference conditional probability P (B1k|Cj), wherein k=1, 2 and 3; namely, P (b11|cj), P (b12|cj), and P (b13|cj) represent conditional probabilities corresponding to short, medium, and long periods, respectively;
counting the brushing times of different brushing pressure preferences, and calculating the proportion of the brushing times to the total brushing times in the time period to obtain the conditional probability P (B2k|cj) of the different brushing pressure preferences in each time period; namely, P (b21|cj), P (b22|cj), and P (b23|cj) represent conditional probabilities corresponding to strong, medium, and weak voltages, respectively; combining P (B1k|cj) and P (B2k|cj) to obtain a brushing condition probability P (Bu|cj) of the comprehensive brushing condition;
Step 203, calculating the ratio of the average usable time length of each time period to the sum of the usable time lengths of the whole use sequence according to the charge amount distribution, the average charge duration and the average usable time length after single charge, and obtaining the probability P (B4|Cj) of the long-term use condition of each time period;
multiplying P (B4|cj) by P (Bu|cj) to obtain 9 comprehensive conditional probabilities Pu (B|cj) at each time period;
step 204, multiplying P (Cj) of each period Cj by 9 Pu (b|cj) of the corresponding period to form a naive bayes model of each period Cj; the 24-period naive Bayesian model is constructed, and the 24-period naive Bayesian model is combined, namely the improved charging naive Bayesian model of the user is formed.
5. The method for controlling charging of an electric toothbrush according to claim 4, wherein the means for obtaining the posterior probability comprises:
when the electric toothbrush is inserted into the charging seat to start charging, acquiring the current charging start time, and determining a time period of 24 time periods Cj to which the current charging start time belongs;
reading the corresponding prior probability P (Cj) according to the current time period Cj, and reading the conditional probabilities P (B1|cj) to P (B9|cj) of 9 comprehensive brushing conditions in the Cj time period;
Multiplying P (Cj) by 9 conditional probabilities P (B1|cj) to P (B9|cj) respectively to obtain posterior probability P of 9 comprehensive brushing conditions in the current time period Cj u (B|Cj)。
6. The method of claim 5, wherein predicting the brushing session of the user using the modified charging naive bayes model comprises:
posterior probability P of 9 comprehensive brushing conditions according to current time period Cj u (B|cj), calculating the prediction probability corresponding to each brushing time preference in the current time period Cj, namely short-time prediction probability P (T1|cj), medium-time prediction probability P (T2|cj) and long-time prediction probability P (T3|cj);
the method is the same as that of obtaining the prediction probability corresponding to the brushing time preference, and the prediction probability corresponding to each brushing pressure preference, namely the strong-pressure prediction probability P (F1|Cj), the medium-pressure prediction probability P (F2|Cj) and the weak-pressure prediction probability P (F3|Cj) in the current time period Cj are calculated;
the short-time prediction probability P (T1|cj), the middle-time prediction probability P (T2|cj), the long-time prediction probability P (T3|cj), the strong-pressure prediction probability P (F1|cj), the middle-pressure prediction probability P (F2|cj) and the weak-pressure prediction probability P (F3|cj) are weighted and counted to obtain the comprehensive prediction probability P (Sj|cj) of each brushing time period in the current time period Cj.
7. The method of claim 6, wherein the preset charging mode comprises:
sorting and dividing the first R time periods with the maximum value of the prediction probability, setting the time period with the maximum value of the prediction probability in the R time periods as a first tooth brushing time period, and pushing the second highest value of the prediction probability in the R time periods as a second time period;
judging whether the electric quantity required by the primary brushing time period is met or not according to the current electric quantity condition of the electric toothbrush; if the current electric quantity meets the electric quantity required by the primary brushing time period, the preset charging mode is a low-speed charging mode; if the current electric quantity is insufficient to meet the electric quantity required by the primary brushing time period, calculating the charge quantity to be met, wherein the preset charging mode is a quick charging mode;
judging the electric quantity demand of the secondary time period on the premise of meeting the electric quantity demand of the primary brushing time period, if the current electric quantity does not meet the electric quantity demand of the secondary time period, presetting a charging mode of the secondary time period as a quick charging mode, and presetting the ending time of the quick charging mode as p minutes before the primary brushing time period starts;
According to the electric quantity requirement of the user in the tooth brushing time period, an elevator-shaped charging scheme is designed; the elevator-shaped charging scheme is that the charging mode is gradually increased from a low-speed charging mode to a rapid charging mode, and gradually decreased to the low-speed charging mode before the last tooth brushing time period.
8. The method for controlling charge of an electric toothbrush according to claim 7, wherein the means for adjusting the posterior probability in real time comprises:
if the user starts brushing teeth with the electric toothbrush within a certain predicted brushing time period of the user; confirm that the user has brushed for a predicted period of time; enhancing posterior probabilities corresponding to brushing duration preferences and pressure preferences of the time period;
if the user does not brush his or her teeth with the electric toothbrush within any of the predicted user's brushing periods, the posterior probability corresponding to the duration preference and pressure preference of the predicted user's brushing period is reduced.
9. The method for controlling charging of an electric toothbrush according to claim 8, wherein the real-time adjustment of the improved charging naive bayes model of the user comprises:
inputting the adjusted posterior probability feedback into an improved charging naive Bayes model, and covering the original posterior probability data;
Traversing the adjusted posterior probability, and updating the brushing condition probability under each time period in the model according to the brushing time preference of short time, medium time and long time and the brushing pressure preference of strong pressure, medium pressure and weak pressure, namely improving P (Bu|Cj) in the charging naive Bayes model; multiplying the updated brushing condition probability of each time period with the using long-strip probability of the corresponding time period; obtaining updated comprehensive conditional probability;
according to the Bayesian theorem, the updated comprehensive conditional probability is combined with the charging prior probability P (Cj), and a naive Bayesian model of each time period in the current stage is obtained through recalculation;
reconstructing the recalculated naive Bayes model of 24 time periods into an improved charging naive Bayes prediction model of the user.
10. A charge control system of an electric toothbrush for realizing a charge control method of an electric toothbrush according to any one of claims 1 to 9, characterized by comprising:
the data acquisition processing module is used for acquiring historical use data of users and preprocessing the historical use data to obtain pre-use data;
the model construction module is used for constructing an improved charging naive Bayes model of the user according to the pre-use data;
A time period prediction module for obtaining posterior probability when the electric toothbrush is charged and predicting a brushing time period of the user by using the improved charging naive bayes model;
the charging mode and adjusting module is used for presetting a charging mode according to the predicted tooth brushing time period; adjusting the posterior probability in real time based on whether the user uses the electric toothbrush during the predicted brushing period; obtaining an adjusted posterior probability;
the model adjusting module is used for adjusting an improved charging naive Bayesian model of a user in real time according to the adjusted posterior probability; all the modules are connected in a wired and/or wireless mode, so that data transmission among the modules is realized.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202410126358.2A CN117691719B (en) | 2024-01-30 | 2024-01-30 | Charging control method and system for electric toothbrush |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202410126358.2A CN117691719B (en) | 2024-01-30 | 2024-01-30 | Charging control method and system for electric toothbrush |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN117691719A true CN117691719A (en) | 2024-03-12 |
| CN117691719B CN117691719B (en) | 2024-04-05 |
Family
ID=90126779
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN202410126358.2A Active CN117691719B (en) | 2024-01-30 | 2024-01-30 | Charging control method and system for electric toothbrush |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN117691719B (en) |
Citations (9)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| US20120133337A1 (en) * | 2010-11-30 | 2012-05-31 | Jan-Willem Rombouts | Method and system for charging a fleet of batteries |
| US20130132388A1 (en) * | 2013-01-22 | 2013-05-23 | Ashtel Studios, Inc. | Apparatus for capturing brushing habits |
| CN103278777A (en) * | 2013-05-24 | 2013-09-04 | 杭州电子科技大学 | Method for estimating health status of lithium battery on basis of dynamic Bayesian network |
| CN107577522A (en) * | 2017-09-30 | 2018-01-12 | 广东欧珀移动通信有限公司 | Application control method, apparatus, storage medium and electronic equipment |
| WO2019062462A1 (en) * | 2017-09-30 | 2019-04-04 | Oppo广东移动通信有限公司 | Application control method and apparatus, storage medium and electronic device |
| WO2019096296A1 (en) * | 2017-11-20 | 2019-05-23 | 蔚来汽车有限公司 | Method, device and system for calculating soc of battery on basis of charging status of charging side |
| US20220311061A1 (en) * | 2019-12-27 | 2022-09-29 | Guangdong Oppo Mobile Telecommunications Corp., Ltd. | Charging method, electronic device and storage medium |
| CN116629409A (en) * | 2023-05-06 | 2023-08-22 | 南昌大学 | SOFC system fault probability prediction method based on naive Bayes algorithm |
| CN116738333A (en) * | 2023-06-09 | 2023-09-12 | 北京航空航天大学 | Electrical signal multi-classification and prediction method for naive Bayes of small sample of aircraft |
-
2024
- 2024-01-30 CN CN202410126358.2A patent/CN117691719B/en active Active
Patent Citations (9)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| US20120133337A1 (en) * | 2010-11-30 | 2012-05-31 | Jan-Willem Rombouts | Method and system for charging a fleet of batteries |
| US20130132388A1 (en) * | 2013-01-22 | 2013-05-23 | Ashtel Studios, Inc. | Apparatus for capturing brushing habits |
| CN103278777A (en) * | 2013-05-24 | 2013-09-04 | 杭州电子科技大学 | Method for estimating health status of lithium battery on basis of dynamic Bayesian network |
| CN107577522A (en) * | 2017-09-30 | 2018-01-12 | 广东欧珀移动通信有限公司 | Application control method, apparatus, storage medium and electronic equipment |
| WO2019062462A1 (en) * | 2017-09-30 | 2019-04-04 | Oppo广东移动通信有限公司 | Application control method and apparatus, storage medium and electronic device |
| WO2019096296A1 (en) * | 2017-11-20 | 2019-05-23 | 蔚来汽车有限公司 | Method, device and system for calculating soc of battery on basis of charging status of charging side |
| US20220311061A1 (en) * | 2019-12-27 | 2022-09-29 | Guangdong Oppo Mobile Telecommunications Corp., Ltd. | Charging method, electronic device and storage medium |
| CN116629409A (en) * | 2023-05-06 | 2023-08-22 | 南昌大学 | SOFC system fault probability prediction method based on naive Bayes algorithm |
| CN116738333A (en) * | 2023-06-09 | 2023-09-12 | 北京航空航天大学 | Electrical signal multi-classification and prediction method for naive Bayes of small sample of aircraft |
Non-Patent Citations (1)
| Title |
|---|
| 戴志辉;谢军;陈曦;王增平;: "基于动态贝叶斯网络的智能变电站监控系统可靠性分析", 电力系统保护与控制, no. 23, 5 December 2018 (2018-12-05), pages 68 - 76 * |
Also Published As
| Publication number | Publication date |
|---|---|
| CN117691719B (en) | 2024-04-05 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| JP3887048B2 (en) | Monitoring system | |
| CN103499794A (en) | Method and device for evaluating state of charge (SOC) of energy storage battery | |
| CN101031810A (en) | Method of estimating the state-of-charge and of the use time left of a rechargeable battery, and apparatus for executing such a method | |
| CN101535827A (en) | Apparatus and method for determination of the state-of-charge of a battery when the battery is not in equilibrium | |
| CN110198042B (en) | Dynamic optimization method for power grid energy storage and storage medium | |
| JPWO2012002429A1 (en) | Charge / discharge control device | |
| CN112368904A (en) | Action generating device, storage element evaluation device, computer program, learning method, and evaluation method | |
| CN116231795B (en) | Comprehensive management control system for distributed storage battery | |
| CN113452056A (en) | Charging control method and system of battery exchange cabinet, equipment and storage medium | |
| CN107565585A (en) | Energy storage device peak regulation report-back time Forecasting Methodology and its model creation method | |
| CN108631348B (en) | Energy storage power station control method based on energy storage unit state evaluation | |
| Lu et al. | Modeling discharge characteristics for predicting battery remaining life | |
| CN116436008A (en) | Power dispatching method and terminal for optical storage charging station | |
| CN110118937A (en) | The storage battery charge state edge calculations optimizing detection method of adaptive prediction model | |
| CN115811135A (en) | Intelligent monitoring and regulation system for charging and battery replacing equipment based on power grid peak regulation and frequency modulation | |
| US11835589B2 (en) | Method and apparatus for machine-individual improvement of the lifetime of a battery in a battery-operated machine | |
| CN117691719B (en) | Charging control method and system for electric toothbrush | |
| CN116353402B (en) | Electric automobile fills electric pile period control system and fills electric pile | |
| CN116488219B (en) | Distributed energy storage acquisition strategy control method based on multi-level architecture | |
| US20230231396A1 (en) | Method and apparatus for executing a charging operation of a device battery | |
| CN114400746B (en) | Charging wake-up control method of electric bicycle BMS | |
| CN111428903A (en) | Interruptible load optimization method based on deep reinforcement learning | |
| CN113466718B (en) | SOC correction method for mobile phone battery | |
| JP2004191150A (en) | Remaining capacity operating device and its method for secondary battery | |
| EP4224656A1 (en) | Method and energy management system for preventing high energy usage peaks |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| PB01 | Publication | ||
| PB01 | Publication | ||
| SE01 | Entry into force of request for substantive examination | ||
| SE01 | Entry into force of request for substantive examination | ||
| GR01 | Patent grant | ||
| GR01 | Patent grant |