Intelligent wearable device energy-saving system based on data analysis

Abstract

The invention relates to the technical field of energy saving, and discloses an intelligent wearable device energy saving system based on data analysis, comprising an intelligent wearable device as a front end of data acquisition and used for acquiring use state data of a user in real time; the data transmission module is used for transmitting the collected use state data to a big data analysis platform for storage; the big data analysis module is used for analyzing the use state data of the big data analysis platform and constructing a user behavior model; and the energy-saving regulation and control module is used for performing energy-saving control on the intelligent wearable equipment according to the prediction result of the user behavior model on the current user behavior. The behavior of each user in the next unit time is predicted through the user behavior model constructed by the big data analysis module, so that the energy consumption of the intelligent wearable device is selectively adjusted according to the prediction result, and the purpose of personalized energy saving is achieved.

Description

Technical Field

[0001] The present invention relates to the technical field of energy saving, and in particular to an energy saving system for a smart wearable device based on data analysis. Background Art

[0002] Smart wearable devices are a general term for wearable devices that are designed and developed using wearable technology to intelligently design daily wearables, such as watches, bracelets, glasses, clothing, etc. These devices integrate sensors, processors, wireless communication and other technologies to achieve monitoring and interaction of human health, exercise, environment and other aspects.

[0003] The energy consumption of smart wearable devices is one of the key concerns of users, because frequent replacement or charging not only affects the normal use of users, but may also affect the normal use of the device due to insufficient battery supply. Through energy-saving design, users can enjoy the convenience of the device for a longer time without worrying about battery problems, thereby improving user satisfaction.

[0004] However, existing smart wearable devices are used by different users, and each user has different usage habits and usage scenarios. As a result, the energy-saving design of existing smart wearable devices cannot be adapted to each user, and the energy utilization efficiency of smart wearable devices has not been effectively improved. Summary of the invention

[0005] The purpose of the present invention is to provide an energy-saving system for smart wearable devices based on data analysis to solve the technical problem that the usage habits and usage scenarios of each user are different, which leads to the energy-saving design of existing smart wearable devices being unable to adapt to each user and the energy utilization efficiency of smart wearable devices not being effectively improved.

[0006] The purpose of the present invention can be achieved through the following technical solutions:

[0007] A smart wearable device energy-saving system based on data analysis, comprising:

[0008] Smart wearable devices, as the front end of data collection, are used to collect user usage status data in real time;

[0009] A data transmission module, used to transmit the collected usage status data to the big data analysis platform for storage;

[0010] Big data analysis module, used to analyze the usage status data of the big data analysis platform and build a user behavior model;

[0011] The energy-saving control module is used to perform energy-saving control on the smart wearable device based on the prediction results of the current user behavior by the user behavior model.

[0012] As a further technical solution, the process of building a user behavior model is:

[0013] Obtain user usage status data, including user personal information, physical condition information, and user interaction information;

[0014] Clean, convert and aggregate user personal information, physical condition information and user interaction information;

[0015] A user behavior model is established based on user personal information, physical condition information and user interaction information; the expression is: ;

[0016] in, is the user behavior index, is the user base coefficient, is the user's physical condition coefficient, is the user interaction coefficient, , , are the deviation values ​​of user basic coefficient, user physical state coefficient and user interaction coefficient in unit time, , , is the weight coefficient, , , is the preset scale factor.

[0017] As a further technical solution, the process of obtaining the user basic coefficient, the user physical state coefficient and the user interaction coefficient is:

[0018] Obtain the current user's personal information, physical condition information, and user interaction information;

[0019] Inputting the current user's personal information, physical condition information, and user interaction information into a pre-trained recurrent neural network model;

[0020] Output user basic coefficient, user physical condition coefficient and user interaction coefficient.

[0021] As a further technical solution, the process of predicting the current user behavior according to the user behavior model is as follows:

[0022] Obtain the current user's behavior prediction index through user behavior model calculation ;

[0023] The current user's behavior prediction index With preset behavior interval To compare:

[0024] when ∈ When the current user's smart wearable device is in the exercise mode in the next unit time;

[0025] when ∈ When the current user's smart wearable device is in the daily mode in the next unit time;

[0026] when ∈ When , it is determined that the smart wearable device of the current user is in the interactive mode in the next unit time;

[0027] in, For the users.

[0028] As a further technical solution, the , , The expression is:

[0029] ;

[0030] ;

[0031] ;

[0032] in, is the starting point and end point of the unit time, , , They are the curves of the current user basic coefficient, user physical condition coefficient, and user interaction coefficient changing over time. , , is a reference curve obtained based on historical data. , , is the reference value.

[0033] As a further technical solution, the process of energy-saving control of smart wearable devices is:

[0034] Go back to multiple consecutive time units from the current moment and obtain the prediction results of user behavior in each time unit;

[0035] Get the total number of times the current user behavior is predicted and the number of times the predicted results matched the actual results ;

[0036] By formula Calculate the prediction accuracy of the current user ;in, ;

[0037] In the formula, is the volatility coefficient, , is the weight factor, is the number of sampling time points, For the Sampling time points, , Respectively The user's physical state coefficient and user interaction coefficient corresponding to each sampling time point, , are the means of the user's physical state coefficient and the user's interaction coefficient, respectively. , is the weight factor;

[0038] The prediction accuracy Compared with the preset accuracy threshold Make comparisons;

[0039] like , then the smart wearable device of the current user is regulated according to energy-saving mode 1;

[0040] Otherwise, the smart wearable device of the current user is regulated according to energy-saving mode 2.

[0041] As a further technical solution, the energy-saving mode 1 is:

[0042] If it is predicted that the next unit of time will be in the daily mode, the smart wearable device is regulated to reduce the data collection frequency, touch voice response speed and screen brightness to a low level;

[0043] If it is predicted that the next unit of time will be in the exercise mode, the smart wearable device is regulated to increase the data collection frequency and screen brightness to a medium level, and reduce the touch voice response speed;

[0044] If it is predicted that the next unit of time will be in interactive mode, the smart wearable device will be regulated to increase the touch voice response speed and screen brightness to a high level, while reducing the frequency of data collection.

[0045] As a further technical solution, the second energy-saving mode is: keeping the current mode of the smart wearable device unchanged.

[0046] Beneficial effects of the present invention:

[0047] The present invention can associate user usage status data with user personal information through the user behavior model, thereby establishing a more accurate user behavior model, predicting user behavior in the next unit time, and then adjusting the current user's smart wearable device to the corresponding energy-saving mode in advance, achieving the purpose of personalized energy saving while taking into account user usage needs, and playing a role in improving user satisfaction. BRIEF DESCRIPTION OF THE DRAWINGS

[0048] The present invention will be further described below in conjunction with the accompanying drawings.

[0049] Figure 1 It is a system structure block diagram of the present invention. DETAILED DESCRIPTION

[0050] 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.

[0051] See also Figure 1 As shown, the present invention is a smart wearable device energy saving system based on data analysis, comprising:

[0052] Smart wearable devices, as the front end of data collection, are used to collect user usage status data in real time;

[0053] A data transmission module, used to transmit the collected usage status data to the big data analysis platform for storage;

[0054] Big data analysis module, used to analyze the usage status data of the big data analysis platform and build a user behavior model;

[0055] The energy-saving control module is used to perform energy-saving control on the smart wearable device based on the prediction results of the current user behavior by the user behavior model.

[0056] The process of building a user behavior model is:

[0057] Obtain user usage status data, including user personal information, physical condition information, and user interaction information;

[0058] Clean, convert and aggregate user personal information, physical condition information and user interaction information;

[0059] A user behavior model is established based on user personal information, physical condition information and user interaction information; the expression is: ;

[0060] in, is the user behavior index, is the user base coefficient, is the user's physical condition coefficient, is the user interaction coefficient, , , are the deviation values ​​of user basic coefficient, user physical state coefficient and user interaction coefficient in unit time, , , is the weight coefficient, which is determined based on historical data and empirical data. , , It is a preset proportional coefficient, which is formulated based on the analysis of historical data and empirical data.

[0061] In this embodiment, in order to solve the technical problems raised in the background technology, the user's usage data is first collected through various sensors built into the smart wearable device, and then the collected usage status data is transmitted to the big data analysis platform for storage through the data transmission module. Subsequently, the usage status data of the big data analysis platform is analyzed through the big data analysis module, and a user behavior model is constructed; the expression of the user behavior model is: , it can be seen from the above expression that the larger the user base coefficient, user physical condition coefficient, and user interaction coefficient, the higher the final user behavior index The larger it is, the more frequently the user uses the device and the higher the screen brightness requirement.

[0062] Through the above-mentioned user behavior model, the user usage status data can be associated with the user's personal information, so as to establish a more accurate user behavior model, predict the user behavior in the next unit time, and then adjust the current user's smart wearable device to the corresponding energy-saving mode in advance, so as to achieve the purpose of personalized energy saving while taking into account the user's usage needs, and play a role in improving user satisfaction.

[0063] The process of obtaining the user basic coefficient, user physical condition coefficient and user interaction coefficient is as follows:

[0064] Obtain the current user's personal information, physical condition information, and user interaction information; the user's personal information includes height, weight, age, gender, geographic location, and occupation; physical condition information is physiological information, including heart rate, blood pressure, blood oxygen saturation, sleep quality, body temperature, the user's steps, movement distance, and calorie consumption; user interaction information is the number of touch and voice recognition times;

[0065] Inputting the current user's personal information, physical condition information, and user interaction information into a pre-trained recurrent neural network model;

[0066] Output user basic coefficient, user physical condition coefficient and user interaction coefficient.

[0067] In this embodiment, the user's behavioral habits and operating habits are reflected by the user basic coefficient, user physical state coefficient and user interaction coefficient, among which the user basic coefficient generally remains unchanged, thereby obtaining a more accurate user basic coefficient, user physical state coefficient and user interaction coefficient, providing accurate data support for subsequent user behavior prediction and energy-saving regulation.

[0068] The process of predicting current user behavior based on the user behavior model is as follows:

[0069] Obtain the current user's behavior prediction index through user behavior model calculation ;

[0070] The current user's behavior prediction index With preset behavior interval To compare:

[0071] when ∈ When the current user's smart wearable device is in the exercise mode in the next unit time;

[0072] when ∈ When the current user's smart wearable device is in the daily mode in the next unit time;

[0073] when ∈ , it is determined that the current user's smart wearable device is in the interactive mode in the next unit time; wherein, For the users.

[0074] Said , , The expression is:

[0075] ;

[0076] ;

[0077] ;

[0078] in, is the starting point and end point of the unit time, , , They are the curves of the current user basic coefficient, user physical condition coefficient, and user interaction coefficient changing over time. , , is a reference curve obtained based on historical data. , , is the reference value.

[0079] In this embodiment, a method for predicting the current user behavior is provided. Specifically, first, the user behavior prediction index of the current user is obtained by calculating the user behavior model. , and then the current user's behavior prediction index With preset behavior interval For comparison, when ∈ When ∈ When , it means that the user's user physical state coefficient and user interaction coefficient do not change much, which is within the normal range; combined with the deviation value of the user's basic coefficient, user physical state coefficient, and user interaction coefficient in unit time, if the deviation value is small, it means that the fluctuation is small, so the current behavior prediction index When the user's physical state coefficient changes greatly and the user interaction coefficient does not change much, the deviation value of the user's physical state coefficient and the user interaction coefficient in unit time is combined to judge that the larger the deviation value, the greater the fluctuation. Therefore, the current behavior prediction index is moderate, and the probability of judging that the user is in exercise mode is higher. At the same time, when ∈ , it is judged that the current user's smart wearable device is in the interactive mode in the next unit time; when the user interaction coefficient is larger and the fluctuation is larger, due to the weight coefficient < < , preset scale factor < < Therefore, the user interaction coefficient and the deviation value of the user interaction coefficient in unit time have a significant impact on the overall The impact is greater, so the probability of judging that the user is in interactive mode is higher; through the above technical solution, accurate prediction of user behavior is achieved, so as to facilitate early regulation of smart wearable devices.

[0080] The process of energy-saving control for smart wearable devices is as follows:

[0081] Go back to multiple consecutive time units from the current moment and obtain the prediction results of user behavior in each time unit;

[0082] Get the total number of times the current user behavior is predicted and the number of times the predicted results matched the actual results ;

[0083] By formula Calculate the prediction accuracy of the current user ;in, ;

[0084] In the formula, is the volatility coefficient, , is the weight factor, is the number of sampling time points, For the Sampling time points, , Respectively The user's physical state coefficient and user interaction coefficient corresponding to each sampling time point, , are the means of the user's physical state coefficient and the user's interaction coefficient, respectively. , is the weight factor, which is formulated based on empirical data;

[0085] The prediction accuracy Compared with the preset accuracy threshold Make comparisons;

[0086] like , then the smart wearable device of the current user is regulated according to energy-saving mode 1;

[0087] Otherwise, the smart wearable device of the current user is regulated according to energy-saving mode 2.

[0088] In this embodiment, according to the historical forecast data, the formula and Calculate the prediction accuracy of the current user , it can be seen from the above formula that the fluctuation coefficient The smaller it is, the and The smaller the integrated standard deviation is, the smaller the impact on the prediction accuracy is. and By comparing the results, we can see whether the prediction accuracy of the current user meets the requirements. If not, it means that there is an error in the current user's behavior prediction model, and it needs to be readjusted before prediction. If it meets the requirements, the smart wearable device will be adjusted in advance according to the prediction results to meet user needs while reducing energy consumption.

[0089] The energy saving mode 1 is:

[0090] If it is predicted that the next unit of time will be in the daily mode, the smart wearable device is regulated to reduce the data collection frequency, touch voice response speed and screen brightness to a low level;

[0091] If it is predicted that the next unit of time will be in the exercise mode, the smart wearable device is regulated to increase the data collection frequency and screen brightness to a medium level, and reduce the touch voice response speed;

[0092] If it is predicted that the next unit of time will be in interactive mode, the smart wearable device will be regulated to increase the touch voice response speed and screen brightness to a high level, while reducing the frequency of data collection.

[0093] The second energy-saving mode is: keeping the current mode of the smart wearable device unchanged.

[0094] Through the above technical solutions, it is only necessary to meet the basic needs of users in daily mode and reduce the power consumption of smart wearable devices. In sports mode, it is necessary to closely monitor the user's physical condition information, so it is necessary to increase the frequency of data collection and allow users to see the content on the screen in real time. In interactive mode, users need to implement interactive actions and voice or touch responses to achieve higher user satisfaction. Relatively speaking, the demand for data collection frequency is relatively low, so smart wearable devices only need to maintain high brightness and fast response. The above control modes are all based on wearable The regulation of the initial value of the device is not based on the real-time value of the smart wearable device; for example, the data collection frequency, touch voice response speed and screen brightness to the lowest initial value are n1, n2, n3 respectively; the real-time value is N1, N2, N3. If it is predicted that it will be in daily mode in the next unit time, n1, n2, n3 will be lowered in turn according to the initial value; if the lowered value is n1①, n2①, n3①, it is determined based on historical data and experience data analysis; then the actual adjustment amount of the smart wearable device this time is N1-n1①, N2-n2①, N3-n3① respectively.

[0095] It should be noted that the calculation formulas and various parameters involved in the calculation in the present invention have been dimensionally processed in advance, and the process of dimensionless processing is well known in the industry and will not be described here.

[0096] The above is a detailed description of an embodiment of the present invention, but the content is only a preferred embodiment of the present invention and cannot be considered to limit the scope of implementation of the present invention. All equivalent changes and improvements made within the scope of the present invention should still fall within the scope of the patent coverage of the present invention.

Claims

1. A smart wearable device energy saving system based on data analysis, characterized in that: include: Smart wearable devices, as the front end of data collection, are used to collect user usage status data in real time; A data transmission module, used to transmit the collected usage status data to the big data analysis platform for storage; Big data analysis module, used to analyze the usage status data of the big data analysis platform and build a user behavior model; The energy-saving control module is used to perform energy-saving control on the smart wearable device based on the prediction results of the current user behavior by the user behavior model.

2. The smart wearable device energy saving system based on data analysis according to claim 1 is characterized in that: The process of building a user behavior model is: Obtain user usage status data, including user personal information, physical condition information, and user interaction information; Clean, convert and aggregate user personal information, physical condition information and user interaction information; A user behavior model is established based on user personal information, physical condition information and user interaction information; the expression is: ; in, is the user behavior index, is the user base coefficient, is the user's physical condition coefficient, is the user interaction coefficient, , , are the deviation values ​​of user basic coefficient, user physical state coefficient and user interaction coefficient in unit time, , , is the weight coefficient, , , is the preset scale factor.

3. The smart wearable device energy saving system based on data analysis according to claim 2 is characterized in that: The process of obtaining the user basic coefficient, user physical condition coefficient and user interaction coefficient is as follows: Obtain the current user's personal information, physical condition information, and user interaction information; Inputting the current user's personal information, physical condition information, and user interaction information into a pre-trained recurrent neural network model; Output user basic coefficient, user physical condition coefficient and user interaction coefficient.

4. The smart wearable device energy saving system based on data analysis according to claim 2 or 3, characterized in that: The process of predicting current user behavior based on the user behavior model is as follows: Obtain the current user's behavior prediction index through user behavior model calculation ; The current user's behavior prediction index With preset behavior interval To compare: when ∈ When the current user's smart wearable device is in the exercise mode in the next unit time; when ∈ When the current user's smart wearable device is in the daily mode in the next unit time; when ∈ When , it is determined that the smart wearable device of the current user is in the interactive mode in the next unit time; in, For the users.

5. The smart wearable device energy saving system based on data analysis according to claim 2 is characterized in that: Said , , The expression is: ； ； ； in, is the starting point and end point of the unit time, , , They are the curves of the current user basic coefficient, user physical condition coefficient, and user interaction coefficient changing over time. , , is a reference curve obtained based on historical data. , , is the reference value.

6. The smart wearable device energy saving system based on data analysis according to claim 1 is characterized in that: The process of energy-saving control for smart wearable devices is as follows: Go back to multiple consecutive time units from the current moment and obtain the prediction results of user behavior in each time unit; Get the total number of times the current user behavior is predicted and the number of times the predicted results matched the actual results ; By formula Calculate the prediction accuracy of the current user ;in, ; In the formula, is the volatility coefficient, , is the weight factor, is the number of sampling time points, For the Sampling time points, , Respectively The user's physical state coefficient and user interaction coefficient corresponding to each sampling time point, , are the means of the user's physical state coefficient and the user's interaction coefficient, respectively. , is the weight factor; The prediction accuracy Compared with the preset accuracy threshold Make comparisons; like , then the smart wearable device of the current user is regulated according to energy-saving mode 1; Otherwise, the smart wearable device of the current user is regulated according to energy-saving mode 2.

7. The smart wearable device energy saving system based on data analysis according to claim 6 is characterized in that: The energy saving mode 1 is: If it is predicted that the next unit of time will be in the daily mode, the smart wearable device is regulated to reduce the data collection frequency, touch voice response speed and screen brightness to a low level; If it is predicted that the next unit of time will be in the exercise mode, the smart wearable device is regulated to increase the data collection frequency and screen brightness to a medium level, and reduce the touch voice response speed; If it is predicted that the next unit of time will be in interactive mode, the smart wearable device will be regulated to increase the touch voice response speed and screen brightness to a high level, while reducing the frequency of data collection.

8. The smart wearable device energy saving system based on data analysis according to claim 6 is characterized in that: The second energy-saving mode is: keeping the current mode of the smart wearable device unchanged.

Images