Tablet computer screen health detection and management system based on AI

Abstract

The invention relates to the technical field of screen health management, and particularly discloses an AI-based tablet computer screen health detection and management system, which comprises a user log extraction unit, a power consumption prediction unit and a comparison analysis detection unit, the user log extraction unit is used for extracting screen operation data, an actual power consumption curve and system data in a user log; the power consumption prediction unit is used for predicting the power consumption of the screen according to the screen operation data and the system data to obtain a predicted power consumption curve; and the comparative analysis detection unit is used for comparing the predicted power consumption curve with the actual power consumption curve and judging the health state of the screen according to a comparison result. The predicted power consumption curve and the actual power consumption curve are compared through the comparative analysis detection unit, the accuracy and sensitivity of screen health state detection are improved, corresponding adjustment is made in time according to the judged screen abnormal problem, and the service life of the screen is prolonged.

Description

Technical Field

[0001] The present invention relates to the technical field of screen health management, and specifically to an AI-based tablet computer screen health detection and management system. Background Technique

[0002] With the rapid development of intelligent devices, the performance of each hardware thereof has also increased rapidly. As the screen for displaying the interactive interface, the requirements for parameters such as resolution, refresh rate, and color saturation are also getting higher and higher. Therefore, the health status of the screen directly affects the user experience. Especially for tablet computers, their screen sizes are larger and prices are higher. Therefore, maintaining the health of the screen can effectively extend its service life and reduce the cost of repair or replacement.

[0003] In existing systems, a battery management system is basically equipped. Through the power consumption ratio of the screen in the battery management system, the power consumption status of the screen can be judged. During the process of detecting the screen, the power consumption status of the screen can indirectly reflect the abnormalities existing during the operation of the screen. For example, when the power consumption of the screen is large, it indicates that the risk of its aging is relatively high. Therefore, the existing screen detection and management system will analyze the power consumption status of the screen within a certain period of time, and then judge the screen health risk.

[0004] Since the power consumption status of the screen is affected by many factors, the existing method for judging the screen health status based on the power consumption of the screen mainly realizes the judgment process when there are obvious problems with the screen and the power consumption is abnormal. Although this method realizes the detection of the screen health, the detection result has obvious hysteresis. Therefore, how to timely and accurately realize the health detection of the tablet computer screen is the fundamental problem to be solved by the present invention. Summary of the Invention

[0005] The purpose of the present invention is to provide an AI-based tablet computer screen health detection and management system to solve the following technical problems:

[0006] How to timely and accurately realize the health detection of the tablet computer screen.

[0007] The purpose of the present invention can be achieved through the following technical solutions:

[0008] An AI-based tablet computer screen health detection and management system, the system includes a user log extraction unit, a power consumption prediction unit, and a comparison and analysis detection unit;

[0009] The user log extraction unit is used to extract the screen operation data, the actual power consumption curve, and the system data in the user log;

[0010] The power consumption prediction unit is used to predict the screen power consumption based on the screen operation data and system data, and obtain a predicted power consumption curve;

[0011] The comparison and analysis detection unit is used to compare the predicted power consumption curve with the actual power consumption curve, and judge the screen health status according to the comparison result.

[0012] By adopting the above technical solution, the method for predicting the screen power consumption based on the screen operation data and system data can obtain a standard for comparing with the actual screen power consumption. When the comparison and analysis detection unit compares the predicted power consumption curve with the actual power consumption curve, the influence of normal influencing factors on the judgment result is reduced, and then the problem of large screen power consumption caused by abnormal factors becomes more obvious, improving the accuracy and sensitivity of the screen health status detection, and making corresponding adjustments in a timely manner according to the judged screen abnormal problems to extend the service life of the screen.

[0013] Further, the process of predicting the screen power consumption includes:

[0014] Obtain the actual power consumption curve in the standby state of the device, compare the actual power consumption curve with the standby standard power consumption curve, and obtain the environmental impact factor according to the comparison result;

[0015] Obtain the predicted power consumption curve according to the screen operation data, system key hardware operation data and environmental impact factor.

[0016] By adopting the above technical solution, obtain the actual power consumption curve in the standby state of the device and compare it with the standby standard power consumption curve, and deduce the environmental state according to the difference range between the two, so as to obtain the environmental impact factor. Through the calculation of the environmental impact factor, the accuracy of the power consumption prediction is improved.

[0017] Further, the process of obtaining the environmental impact factor includes:

[0018] Select the starting time point t1, obtain the corresponding value of the starting time point on the actual power consumption curve, and determine the standby standard power consumption curve based on the corresponding value;

[0019] Select the ending time point t2, the time difference between t2 and t1 is a preset fixed value, and obtain the area S enclosed by the actual power consumption curve, the standby standard power consumption curve and the straight line t = t2;

[0020] Compare the area S with multiple preset intervals, and obtain the corresponding environmental impact factor according to the preset interval where the area S is located.

[0021] By adopting the above technical solution, determine the corresponding environmental impact factor according to the deviation condition between the actual power consumption curve and the standby standard power consumption curve.

[0022] Further, the screen operation data includes a brightness change curve, a refresh rate, and a touch frequency change curve over time;

[0023] The system data includes a CPU real-time operation occupancy ratio curve and real-time running software;

[0024] The process of obtaining the predicted power consumption curve includes:

[0025] Analyze the content of the real-time running software based on AI to obtain the display category of the real-time running software, and obtain the dynamic influence factor according to the display type;

[0026] Obtain the temperature influence factor according to the CPU real-time operation occupancy ratio curve;

[0027] Establish a correlation model, which is positively correlated with brightness, refresh rate, touch frequency, dynamic influence factor, temperature influence factor, and environmental influence factor respectively;

[0028] Select an analysis period, and fit the brightness change curve, refresh rate, touch frequency change curve over time, dynamic influence factor, temperature influence factor, and environmental influence factor within the analysis period to obtain the predicted power consumption curve.

[0029] By adopting the above technical solution, the power consumption value is fitted by using the data of the correlation model at a single time point, and the predicted power consumption curve is obtained by connecting the power consumption values at multiple time points. Through the process of obtaining the predicted power consumption curve, a relatively accurate reference can be provided for the actual power consumption range. When comparing the predicted power consumption curve with the actual power consumption curve, the accuracy and sensitivity of the screen health status detection are improved, and corresponding adjustments are made in a timely manner according to the detected screen abnormal problems to extend the service life of the screen.

[0030] Further, the process of obtaining the temperature influence factor includes:

[0031] Through the formula Calculate the heat reference value H(t) at the current time point;

[0032] According to the temperature influence factor corresponding to the numerical range interval where the heat reference value H(t) is located;

[0033] Among them, is the CPU real-time operation occupancy ratio curve, is the CPU operation occupancy ratio reference value, is the device heating model, is the test device heat dissipation rate.

[0034] By adopting the above technical solution, through the acquisition of the temperature influence factor, it is possible to judge the influence of the CPU operation heat generation on the screen power consumption, and thus improve the accuracy of the screen power consumption prediction.

[0035] Furthermore, the process of comparing the predicted power consumption curve with the actual power consumption curve includes:

[0036] Taking the starting point of the analysis period as the origin to establish a coordinate system, and placing the predicted power consumption curve and the actual power consumption curve in the same coordinate system;

[0037] Through the formula Calculate to obtain the power consumption difference , and compare the power consumption difference with the preset warning value. When the power consumption difference exceeds the preset warning value, it is judged that the screen health is abnormal;

[0038] Among them, 0 to t3 is the analysis period, is the predicted power consumption curve, is the actual power consumption curve.

[0039] By adopting the above technical solution, it is possible to judge the abnormal situation of the screen power consumption, and thus indirectly judge the health of the screen. The judgment result is convenient for subsequent management of the screen health; in addition, compared with the method of comparing the actual power consumption with a fixed standard in the prior art, this method can more accurately judge the abnormal situation of the screen power consumption and improve the timeliness of screen processing.

[0040] Furthermore, the system further includes an intelligent management module, and the intelligent management module starts to adjust the management strategy when it judges that the screen health is abnormal.

[0041] By adopting the above technical solution, it is possible to timely handle the screen health problems and extend the service life of the screen.

[0042] Furthermore, the adjustment management strategy includes multiple adjustment levels, and the corresponding adjustment level is determined according to the range where the difference between the power consumption difference and the preset warning value is located.

[0043] By adopting the above technical solution, determining the corresponding adjustment level according to the range where the difference between the power consumption difference and the preset warning value is located can adaptively meet the problem handling in different situations of the screen, improve the screen life while reducing the impact on users.

[0044] The beneficial effects of the present invention:

[0045] The method of predicting the screen power consumption according to the screen operation data and system data in the present invention can obtain a standard for comparing with the actual screen power consumption. When comparing the predicted power consumption curve with the actual power consumption curve through the comparison and analysis detection unit, the influence of normal influencing factors on the judgment result is reduced, and thus the problem of large screen power consumption caused by abnormal factors becomes more obvious, improving the accuracy and sensitivity of the detection of the screen health state, and making corresponding adjustments in a timely manner according to the judged screen abnormal problems to extend the service life of the screen. BRIEF DESCRIPTION OF THE DRAWINGS

[0046] The present invention will be further described below with reference to the accompanying drawings.

[0047] Figure 1 It is a logic block diagram of an AI-based tablet computer screen health detection and management system in the present invention. DETAILED DESCRIPTION OF THE EMBODIMENTS

[0048] The technical solutions in the embodiments of the present invention will be clearly and completely described below with reference to the accompanying drawings in the embodiments of the present invention. Obviously, the described embodiments are only a part of the embodiments of the present invention, rather than all the embodiments. All other embodiments obtained by those of ordinary skill in the art based on the embodiments of the present invention without creative efforts shall fall within the protection scope of the present invention.

[0049] In one embodiment, an AI-based tablet computer screen health detection and management system is provided. Please refer to Figure 1 As shown, the system includes a user log extraction unit, a power consumption prediction unit, and a comparison and analysis detection unit. The user log extraction unit is used to extract the screen operation data, actual power consumption curve, and system data in the user log. The screen operation data includes parameter data such as screen brightness data, refresh rate data, and touch frequency data that affect the screen power consumption. The system data includes data related to the operation of the CPU and data related to the running software. The power consumption prediction unit is used to predict the screen power consumption according to the screen operation data and system data to obtain a predicted power consumption curve. Finally, the comparison and analysis detection unit compares the predicted power consumption curve with the actual power consumption curve, and judges the screen health state according to the comparison result. Through the above solution, the method of predicting the screen power consumption according to the screen operation data and system data can obtain a standard for comparing with the actual screen power consumption. When comparing the predicted power consumption curve with the actual power consumption curve through the comparison and analysis detection unit, the influence of normal influencing factors (such as screen brightness, refresh rate, etc.) on the judgment result is reduced, and thus the problem of large screen power consumption caused by abnormal factors becomes more obvious, thereby improving the accuracy and sensitivity of the detection of the screen health state, and making corresponding adjustments in a timely manner according to the judged screen abnormal problems to extend the service life of the screen.

[0050] In one embodiment, the process of predicting the screen power consumption includes: obtaining the actual power consumption curve of the device in the standby state, comparing the actual power consumption curve with the standby standard power consumption curve, and obtaining the environmental impact factor according to the comparison result; obtaining the predicted power consumption curve based on the screen operation data, the system key hardware operation data, and the environmental impact factor; since the battery power consumption is affected by the environmental temperature, and the temperature of the tablet computer cannot be directly measured, in this embodiment, by obtaining the actual power consumption curve of the device in the standby state, comparing it with the standby standard power consumption curve, and inversely deducing the environmental state according to the difference range between the two, and then obtaining the environmental impact factor. In the process of predicting the screen power consumption, the environmental factor has an important impact. Therefore, by calculating the environmental impact factor, the accuracy of the power consumption prediction is improved; it should be noted that the above-mentioned device standby state means that the screen is turned off and the CPU occupancy rate of all background running software is less than the preset ratio, and the preset ratio is selected according to the accuracy requirement. In this embodiment, the preset ratio is selected as 10%; the standby standard power consumption curve is obtained by fitting the measurement data of the same model tablet computer under the standard environmental temperature. Therefore, when the actual power consumption curve differs greatly from the standby standard power consumption curve, it indicates that the current tablet computer is greatly affected by environmental factors, and at this time, the environmental impact factor is also large.

[0051] In one embodiment, the process of obtaining the environmental impact factor includes: selecting a starting time point t1, obtaining the corresponding value of the starting time point on the actual power consumption curve, determining the standby standard power consumption curve based on the corresponding value. In the test data, the slope of the standby standard power consumption curve approaches a fixed value. Therefore, the slope k is fitted according to the test data, and the standby standard power consumption curve is determined according to the slope k. Select an ending time point t2, and the time difference between t2 and t1 is a preset fixed value. The preset fixed value can be set to different sizes. When the preset fixed value is larger, the obtained environmental impact factor is more accurate. After establishing the coordinate system, the actual power consumption curve and the standby standard power consumption curve are placed in the same coordinate system. Since the starting values are the same, the area S enclosed by the actual power consumption curve, the standby standard power consumption curve, and the straight line t = t2 is obtained. The area S reflects the deviation between the actual power consumption curve and the standby standard power consumption curve during the period from t1 to t2. Obviously, the larger the area S, the greater the deviation between the two, that is, the greater the influence of environmental factors. Therefore, the area S is compared with multiple preset intervals, and the corresponding environmental impact factor is obtained according to the preset interval where the area S is located. It should be noted that the multiple preset intervals are divided according to the test data during the period from t1 to t2 under different temperature states, and considering the influence of actual error factors, and different preset intervals correspond to environmental impact factors. The specific size of the environmental impact factor will vary according to the quantization standard of the influence degree of different environmental factors on power consumption. Therefore, it is necessary to ensure that the quantization standard of the environmental impact factor is the same during the power consumption prediction process.

[0052] In one embodiment, the screen operation data includes the brightness change curve, the refresh rate, and the touch frequency change curve over time; the system data includes the CPU real-time operation occupancy ratio curve and the real-time running software; among the above parameters, the brightness change curve, the refresh rate, and the touch frequency change curve over time are all important factors affecting the screen power consumption, and the real-time running software will have different effects on the screen power consumption due to different software types. For example, games and videos have a relatively high frequency of screen switching, so the power consumption is relatively high, while the screen power consumption during the operation of software such as browsers and music players is relatively low. For the CPU real-time operation occupancy ratio curve, when its real-time operation occupancy ratio is relatively high and lasts for a long time, it is likely to cause the device to heat up, and further lead to a relatively high temperature. Therefore, by predicting the screen power consumption through the above parameters, a relatively accurate result can be obtained; the process of obtaining the predicted power consumption curve includes: analyzing the content of the real-time running software based on AI to obtain the display category of the real-time running software, and different display categories set corresponding dynamic influence factors according to the screen power consumption rate in the big data. Then, the temperature influence factor is obtained according to the CPU real-time operation occupancy ratio curve. Like the environmental influence factor, the specific magnitudes of the dynamic influence factor and the temperature influence factor will vary according to different quantization standards. Therefore, during the power consumption prediction process, it is necessary to ensure that the quantization standards of the dynamic influence factor and the temperature influence are the same respectively. It should also be noted that both the environmental influence factor and the temperature influence factor are the influence of temperature on the screen power consumption, but the environmental influence factor is the influence of the environmental temperature, and the temperature influence factor is the influence of the device temperature caused by the CPU operation, and the two do not overlap; then a correlation model is established, and the correlation model is positively correlated with the brightness, refresh rate, touch frequency, dynamic influence factor, temperature influence factor, and environmental influence factor respectively. It should be noted that the correlation model can be established by training a large model to establish the relationship between the power consumption and different influencing factors, and the trained large model is used as the correlation model. It can also determine the corresponding relationship based on the test data through mathematical modeling, and the mathematical model is used as the correlation model. The specific establishment method is not further described in this embodiment. After obtaining the correlation model, an analysis period is selected, and the predicted power consumption curve is obtained by fitting the brightness change curve, refresh rate, touch frequency change curve over time, dynamic influence factor, temperature influence factor, and environmental influence factor within the analysis period. The fitting process is to fit the power consumption value through the data of the brightness change curve, refresh rate, touch frequency change curve over time, dynamic influence factor, temperature influence factor, and environmental influence factor at a single time point, and the predicted power consumption curve is obtained by connecting the power consumption values at multiple time points. Through the process of obtaining the predicted power consumption curve, a relatively accurate reference can be provided for the actual power consumption range. When comparing the predicted power consumption curve with the actual power consumption curve, the accuracy and sensitivity of the screen health status detection are improved, and corresponding adjustments are made in a timely manner according to the detected screen abnormal problems to extend the service life of the screen.

[0053] In one embodiment, the process of obtaining the temperature influence factor includes: through the formula calculate to obtain the heat reference value H(t) at the current time point; where is the real-time operation occupancy curve of the CPU, is the operation occupancy reference value of the CPU, is the device heating model, is the heat dissipation rate of the test device, and the device heating model and the heat dissipation rate of the test device are both obtained by fitting according to the temperature detection status under different CPU operation test data of this model tablet computer device. Therefore, the calculated heat reference value H(t) can reflect the heat generation status of the device. According to the temperature influence factor corresponding to the numerical range interval where the heat reference value H(t) is located, the numerical range interval is obtained by dividing according to the result range calculated by substituting the test data, and the corresponding temperature influence factor is set according to the influence degree of different heat generations on the power consumption of the screen. Its specific numerical value is determined according to the selected quantization standard. For example, taking the ratio of the power consumption corresponding to the middle value of each numerical range interval to the power consumption corresponding to the middle value of the smallest range interval every 30 minutes as the quantization standard. Through the acquisition of the temperature influence factor, the influence status of CPU operation heating on the screen power consumption can be judged, and thus the accuracy of screen power consumption prediction can be improved.

[0054] In one embodiment, the process of comparing the predicted power consumption curve with the actual power consumption curve includes: establishing a coordinate system with the starting point of the analysis period as the origin, and placing the predicted power consumption curve and the actual power consumption curve in the same coordinate system; through the formula calculate to obtain the power consumption difference , where 0 to t3 is the analysis period, and the duration of the analysis period is set according to the selection, is the predicted power consumption curve, is the actual power consumption curve. Therefore, the power consumption difference reflects the deviation status between the actual power consumption and the predicted power consumption. When the power consumption difference is larger, it indicates that the degree to which the actual power consumption is lower than the predicted power consumption is higher. At this time, the risk of abnormal screen power consumption is higher. Therefore, compare the power consumption difference with the preset warning value. The preset warning value is set by fitting according to the duration of the analysis period and the test data of normal screens and abnormally power-consuming screens. When the power consumption difference When it exceeds the preset warning value, it indicates that there is an abnormality in the screen. Therefore, it is determined that the screen is unhealthy. Through the above comparison process, the abnormal power consumption situation of the screen can be judged, and then the health of the screen can be judged indirectly. The judgment result is convenient for subsequent management of the screen health. In addition, compared with the method of comparing the actual power consumption with a fixed standard in the prior art, this method can more accurately judge the abnormality of the screen power consumption situation and improve the timeliness of screen processing.

[0055] In one embodiment, the system further includes an intelligent management module. The intelligent management module starts to adjust the management strategy when it determines that the screen is unhealthy. The adjustment management strategy includes multiple groups of adjustment levels, including adjusting the refresh rate, adjusting the brightness, reminding the user to turn off the device, etc., which are set according to requirements and will not be further limited here. When the difference is larger, it indicates that the severity of the abnormal screen power consumption problem is higher. Adopting a higher-level adjustment level can prevent the screen health problem from further expanding. Therefore, according to the power consumption difference Determining the corresponding adjustment level according to the range where the difference from the preset warning value is located can adaptively meet the problem handling in different situations of the screen, improve the screen life while reducing the impact on users.

[0056] The above has described an embodiment of the present invention in detail, but the content described is only the preferred embodiment of the present invention and cannot be considered as limiting the scope of implementation of the present invention. All equivalent changes and improvements made according to the scope of the present invention application should still fall within the scope covered by the patent of the present invention.

Claims

1. An AI-based tablet computer screen health detection and management system, characterized in that, The system includes a user log extraction unit, a power consumption prediction unit, and a comparison and analysis detection unit; The user log extraction unit is used to extract the screen operation data, the actual power consumption curve, and the system data from the user log; The power consumption prediction unit is used to predict the screen power consumption based on the screen operation data and the system data to obtain a predicted power consumption curve; The comparison and analysis detection unit is used to compare the predicted power consumption curve with the actual power consumption curve and determine the screen health status according to the comparison result.

2. The AI-based tablet computer screen health detection and management system according to claim 1, wherein, The process of predicting the screen power consumption includes: Obtain the actual power consumption curve in the device standby state, compare the actual power consumption curve with the standby standard power consumption curve, and obtain the environmental impact factor according to the comparison result; Obtain the predicted power consumption curve based on the screen operation data, the system key hardware operation data, and the environmental impact factor.

3. The AI-based tablet computer screen health detection and management system according to claim 2, wherein The process of obtaining the environmental impact factor includes: Select the starting time point t1, obtain the corresponding value of the starting time point on the actual power consumption curve, and determine the standby standard power consumption curve based on the corresponding value; Select the ending time point t2, the time difference between t2 and t1 is a preset fixed value, and obtain the area S enclosed by the actual power consumption curve, the standby standard power consumption curve, and the line t = t2; Compare the area S with multiple preset intervals, and obtain the corresponding environmental impact factor according to the preset interval where the area S is located.

4. The AI-based tablet computer screen health detection and management system according to claim 3, wherein, The screen operation data includes the brightness change curve, the refresh rate, and the touch frequency change curve over time; The system data includes the CPU real-time operation occupancy ratio curve and the real-time running software; The process of obtaining the predicted power consumption curve includes: Analyze the content of the real-time running software based on AI, obtain the display category of the real-time running software, and obtain the dynamic impact factor according to the display type; Obtain the temperature impact factor according to the CPU real-time operation occupancy ratio curve; Establish a correlation model, and the correlation model is positively correlated with the brightness, the refresh rate, the touch frequency, the dynamic impact factor, the temperature impact factor, and the environmental impact factor respectively; Select the analysis period, and fit the predicted power consumption curve according to the brightness change curve, the refresh rate, the touch frequency change curve over time, the dynamic impact factor, the temperature impact factor, and the environmental impact factor within the analysis period.

5. The AI-based tablet computer screen health detection and management system according to claim 4, wherein, The process of obtaining the temperature impact factor includes: Obtained by the formula Calculate the heat reference value H(t) at the current time point; According to the temperature impact factor corresponding to the numerical range interval where the heat reference value H(t) is located; Among them, is the real-time running occupancy ratio curve of the CPU, is the baseline value of the CPU running occupancy ratio, is the device heating model, is the heat dissipation rate of the test device.

6. The AI-based tablet computer screen health detection and management system according to claim 4, wherein, The process of comparing the predicted power consumption curve with the actual power consumption curve includes: Establish a coordinate system with the starting point of the analysis period as the origin, and place the predicted power consumption curve and the actual power consumption curve in the same coordinate system; Obtained through the formula Calculate the power consumption difference , and compare the power consumption difference with a preset warning value. When the power consumption difference exceeds the preset warning value, it is determined that the screen health is abnormal; Among them, 0 to t3 is the analysis period, is the predicted power consumption curve, is the actual power consumption curve.

7. The AI-based tablet computer screen health detection and management system according to claim 6, characterized in that, The system further includes an intelligent management module, and the intelligent management module starts to adjust the management strategy when it determines that the screen health is abnormal.

8. The AI-based tablet computer screen health detection and management system according to claim 7, wherein The adjustment management policy includes multiple groups of adjustment levels, and determines the corresponding adjustment level according to the range where the difference between the power consumption difference and the preset warning value lies.

Images