A health detection method and system for new energy electric vehicles

Abstract

This application discloses a health detection method and system for new energy electric vehicles, relating to the field of big data technology, including: S101, collecting and storing the driver's biometrics and uploading it to a cloud platform; S102, storing and analyzing health data on the cloud platform; S103, viewing health data and health status; Step S102 is implemented in the following way: S201, determining the driver's health status based on the vital sign data; S202, establishing a health feature database, comparing the time series data of the cloud platform with the health feature database, and identifying the current health status of the driver; the health feature database is built into a hierarchical tree based on the time series data, and the hierarchical tree generates cluster superposition centers based on different difference points; the information evaluation rate and similarity rate are obtained based on the cluster superposition centers; it can effectively help drivers manage their own health status and effectively assist driving.

Description

Technical Field

[0001] This invention relates to the field of big data technology, and in particular to a method and system for health detection of new energy electric vehicles. Background Technology

[0002] With the maturity of big data technology, applying big data platforms to various industries has become a development trend. In the development of intelligent transportation systems, big data technology is also indispensable. The rational use of big data processing platforms can effectively solve the problem of analyzing and processing massive traffic flow data.

[0003] Currently, with the development of society and economy, the number of vehicles is gradually increasing, and traffic accidents are also occurring frequently. Driver factors are the main cause of traffic accidents, among which the driver's health status may directly affect the driver's attention while driving.

[0004] The majority of current new energy electric vehicle drivers are middle-aged and elderly people. This group generally suffers from slow reaction times, weak constitutions, forgetfulness, and living alone, making it difficult for them to monitor their health in a timely manner. However, existing new energy electric vehicles generally have low levels of intelligence, failing to detect the driver's health status and thus hindering their ability to manage their own health effectively.

[0005] At the same time, when managing personnel health, it is important to avoid judging changes in drivers' health based on different time-series data, to issue timely warnings, reduce health problems during driving, and improve work efficiency. Summary of the Invention

[0006] This application provides a health detection method and system for new energy electric vehicles, which solves the problem that health detection in the prior art cannot be customized based on individual time-series data, thereby improving the effectiveness and accuracy of health detection.

[0007] This application provides a method and system for health detection of new energy electric vehicles, including:

[0008] S101 collects and stores the driver's biometrics and uploads them to the cloud platform;

[0009] S102, cloud platform for storing and analyzing health data;

[0010] S103, view health data and health status;

[0011] Step S102 is implemented in the following way:

[0012] S201, Determine the health status of the vehicle driver based on vital sign data;

[0013] S202, Establish a health feature database, compare the time series data of the cloud platform with the health feature database, and identify the current health status of the driver;

[0014] The health feature database is built into a hierarchical tree based on time series data. The hierarchical tree generates cluster superposition centers based on different difference point sets. The information evaluation rate and similarity rate are obtained based on the cluster superposition centers.

[0015] The first node is the direct child of the root node, the second node is the child of the first node, the third node is the child of the second node, and the fourth node is the child of the third node.

[0016] S203: If the driver's health status is normal, continue monitoring; if the driver's health status becomes abnormal for a short period of time, issue a warning according to the type of abnormality and provide emotional support; if the driver's health status becomes abnormal for a long period of time, issue a warning according to the type of health status and send health monitoring data to the cloud platform.

[0017] One or more technical solutions provided in the embodiments of this application have at least the following technical effects or advantages:

[0018] This invention collects vital sign data of drivers and establishes a health feature database based on different time series. Different health conditions are expressed through multiple cluster overlay centers, and different labels are set for the cluster overlay centers. Based on the information evaluation rate and similarity rate obtained from the cluster overlay centers, the characteristics and trends of drivers' health status can be effectively identified, and the health status of drivers can be displayed in a timely manner, providing reasonable assistance for healthy driving. This improves the accuracy and effectiveness of personal health detection. Attached Figure Description

[0019] Figure 1 This is a schematic diagram of the process of the present invention;

[0020] Figure 2 This is a flowchart illustrating step S102 of the present invention;

[0021] Figure 3 This is a flowchart illustrating step S103 of the present invention. Detailed Implementation

[0022] To facilitate understanding of the present invention, a more complete description of this application will be given below with reference to the accompanying drawings, which illustrate preferred embodiments of the invention. However, the invention can be implemented in many different forms and is not limited to the embodiments described herein. Rather, these embodiments are provided to enable a more thorough and complete understanding of the disclosure of the present invention.

[0023] It should be noted that the terms "vertical," "horizontal," "up," "down," "left," "right," and similar expressions used in this article are for illustrative purposes only and do not represent the only possible implementation.

[0024] Unless otherwise defined, all technical and scientific terms used herein have the same meaning as commonly understood by one of ordinary skill in the art to which this invention pertains; the terminology used herein in the description of the invention is for the purpose of describing particular embodiments only and is not intended to limit the invention; the term "and / or" as used herein includes any and all combinations of one or more of the associated listed items.

[0025] like Figure 1 As shown, a health testing method for new energy electric vehicles includes:

[0026] S101 collects and stores the driver's biometrics and uploads them to the cloud platform;

[0027] S102, cloud platform for storing and analyzing health data;

[0028] S103, view health data and health status;

[0029] Specifically, by using devices installed in the vehicle to collect drivers' health data regularly, such as sensors, wristbands, and watches, the data can be collected.

[0030] Specifically, the system collects vital signs data of the driver, including heart rate, blood oxygen saturation, blood pressure, and body temperature. It stores half a month's worth of vital sign data and performs statistical analysis to identify any anomalies. The data and any anomalies are then transmitted to the central control screen via Bluetooth.

[0031] The central control screen receives data from Bluetooth communication for display and emergency alarms, and transmits it to the T-BOX via CAN communication.

[0032] T-BOX uploads health data and other vehicle data to the cloud platform via 4G communication;

[0033] The cloud platform pushes the driver's health data and emergency alarm signals to the driver and their family through a mobile app. The driver and their family can also check the driver's health status through the mobile app.

[0034] Specifically, in one embodiment of the present invention, such as Figure 2 As shown, step S102 can be implemented in the following way:

[0035] S201, Determine the health status of the vehicle driver based on vital sign data;

[0036] Specifically, the vital signs data stored on the cloud platform are obtained and processed in various formats, including 1-day, 3-day, 7-day, and 15-day distributions.

[0037] Different health curves are generated based on different times. The health curve is an image generated by subtracting the standard set data from the collected heart rate, blood oxygen, blood pressure, and body surface temperature. The image contains positive and negative values. If the detected heart rate and other data exceed the set threshold, an alarm will be triggered and sent to the cloud platform for emergency assistance.

[0038] Specifically, the vital sign data is preprocessed by performing a first-level data transformation to remove irrelevant information from the data, thereby obtaining a set of vital sign points.

[0039] The aforementioned primary data change involves classifying vital sign data according to the time of collection to obtain common vital signs within different time intervals; for example, vital signs at this time could be data such as heart rate, blood pressure, and blood oxygen that occur frequently within a day.

[0040] Specifically, the selection of vital sign data should focus on feature points that are invariant and relatively stable.

[0041] It should be noted that the main purpose of the selected feature points is to remove some abrupt or abnormal life feature data from the collected data, and the collected life feature data needs to meet the requirements of a relatively stable state and feature points that can intuitively represent life features.

[0042] After obtaining accurate feature points, the feature points are combined to generate time series, and the life characteristics under different time series are preprocessed to remove outliers.

[0043] S202, Establish a health feature database, compare the time series data of the cloud platform with the health feature database, and identify the current health status of the driver;

[0044] Specifically, each pair of time series is compared using health curve data from the past 1 day, 3 days, 7 days, and 15 days, divided into four time combinations, which are then compared sequentially.

[0045] Specifically, the attributes set in the health characteristic database are value ranges that meet the requirements of a normal health state;

[0046] Specifically, by obtaining the trend or pattern of each time series and comparing the vital characteristics of each group, the long-term health status of drivers can be determined.

[0047] At the same time, each person's health condition is different, and the judgment criteria will also adopt different judgment data. Specifically, the health feature database is based on standard data obtained from the analysis of a large number of people. When judging individual users, basic vital sign data is obtained based on multiple 15-day time series of users, and the health feature database is adjusted according to individual judgment criteria.

[0048] Specifically, the health feature database is built into a hierarchical tree based on time series data, and the hierarchical tree generates cluster superposition centers based on different difference points.

[0049] Specifically, a hierarchical tree is set up according to different time ranges, with the root node of the tree representing the overall time range and the child nodes representing specific days;

[0050] In this embodiment, data of different number of days are used as child nodes of the root node, and data of 1 day, 3 days, 7 days, and 15 days are used as child nodes of the hierarchical tree; the hierarchy between each child node and the root node directly represents the difference in time range.

[0051] Specifically, the root node time range is set to 15 days, with 1 day's data designated as the first node, 3 days' data as the second node, 7 days' data as the third node, and 15 days' data as the fourth node.

[0052] The first node is the direct child of the root node, the second node is the child of the first node, the third node is the child of the second node, and the fourth node is the child of the third node; in the hierarchical tree, the time ranges of child nodes and parent nodes are mutually inclusive and compatible.

[0053] Specifically, the time series data with the middle mean of each time series is used as the initial data set; different filtering kernels are applied to each initial data set to obtain multiple health curves; the difference between each adjacent health curve is used to obtain an initial difference point set; all initial difference point sets are superimposed to obtain the difference point set for each time series; the first node of the hierarchical tree for each time series is obtained based on the difference point set; each first node corresponds to a difference point set; at the same time, the second, third, and fourth nodes each correspond to a difference point set.

[0054] Hierarchical trees are used to divide different information into multiple nodes, forming multiple cluster centers with different rules. Then, the rule distance between each node is calculated, and similar nodes are merged. This merging process is repeated until all nodes under the hierarchical tree are merged into one node. The resulting node serves as an important credential for the platform to return the health status, representing the commonalities of most health statuses.

[0055] Specifically, between the parent and child nodes of each level of the tree node, the child node contains the feature points of the parent node. When performing superimposed clustering on the child nodes, the corresponding parent nodes of the child nodes are also clustered at the same time. After the final clustering is completed, the vital characteristic data of the driver can be divided into multiple cluster centers with different rules.

[0056] Specifically, for each time series, the feature points of the set with the largest difference and the set of adjacent difference are merged into the set with the largest difference to obtain the first cluster center; for each time series, the corresponding set of the largest adjacent difference and the set of the nearest difference are merged to obtain the second cluster center; the first cluster center and the second cluster center are merged to obtain the cluster superposition center; and so on, continuously superimposing and merging to obtain the cluster superposition center for each time series.

[0057] Generally, the neighboring difference point set refers to the set with the highest similarity to the maximum difference point set, and the maximum neighboring difference point set refers to the difference point set with the highest similarity and the largest number of points; during merging, the points that are farthest apart in the difference point set are merged; the first, second, third, and fourth nodes each have multiple clustering superposition centers;

[0058] Specifically, each cluster overlay center is labeled based on the feature points of the maximum difference set;

[0059] The cluster overlay centers under each time series can represent the changes in the health status of drivers at different times; by evaluating the tightness and separation of the clusters, we can know whether the clustering effect is in a good state, and by displaying the cluster overlay center data, we can know the health status.

[0060] S203: If the driver's health status is normal, continue monitoring; if the driver's health status becomes abnormal for a short period of time, issue a warning according to the type of abnormality and provide emotional relief; if the driver's health status becomes abnormal for a long period of time, issue a warning according to the type of health status and send health monitoring data to the cloud platform.

[0061] If a driver experiences a brief, mild health condition, music can be used to soothe the driver and help them adjust their state. Furthermore, any abnormal vital signs that may be caused by emotional factors can help manage the driver's health by adjusting their emotions.

[0062] Furthermore, when a driver's health condition is abnormal for an extended period, it may indicate a sudden illness or other conditions that could pose a significant danger; or it may indicate that the driver is in a state of sub-health for a long time. In such cases, the driver will be alerted to their health condition.

[0063] Specifically, in the embodiments of the present invention, such as Figure 3 As shown, step S102 also includes the following:

[0064] The information evaluation rate and similarity rate are obtained based on the cluster overlay centers;

[0065] Specifically, information entropy is constructed based on the feature points of each cluster superposition center, and entropy curves are generated. Mutation points are obtained for each entropy curve. The trend of the entropy curve is obtained based on the distance between the feature points of the cluster superposition center corresponding to the difference point set where each mutation point is located and the feature points of the cluster superposition center of the corresponding maximum difference point set. The information evaluation rate of the cluster superposition center is obtained based on the entropy curve trend corresponding to each mutation point.

[0066] At this point, the entropy curve trend represents the changes in health. Based on certain trends, we can see whether the driver's health is good or bad. Furthermore, by merging all the entropy curve trends, we can understand the overall changes and the clustering effect of the cluster superposition centers.

[0067] Specifically, the process of obtaining abrupt change points from the entropy curve includes:

[0068] Construct an entropy sequence based on the entropy curve. Obtain the entropy difference by taking the difference between the previous extreme point and the next extreme point in the entropy sequence. Take the smaller of the two entropy values ​​that are greater than the average of the entropy difference as the abrupt change point of the entropy curve.

[0069] Specifically, the entropy difference between the cluster superposition center corresponding to each mutation point and the cluster superposition center under the corresponding time series is obtained, and the ratio of the entropy difference to the cluster superposition center is used as the information evaluation rate.

[0070] The similarity rate is calculated as the ratio of the feature points of the cluster superposition center corresponding to each mutation point to the feature points of the largest cluster superposition center in the corresponding time series.

[0071] At this point, the cluster superposition center corresponding to each mutation point refers to the cluster superposition center within a unit time period, while the cluster superposition center under the corresponding time series refers to the cluster superposition center within a specified time period that is longer than a unit time period.

[0072] The driver's health status is classified and labeled based on the information assessment rate, and health characteristics are obtained based on the similarity rate.

[0073] Based on the ratio of entropy difference to cluster center, the health status of each mutation point is assessed, a health status report is generated, and sent to the cloud platform.

[0074] The health report is generated. Based on the information evaluation rate corresponding to each mutation point, the current health status is determined to be of what health category. Furthermore, based on the similarity rate, the characteristics that the current health status is close to are judged, thereby obtaining the driver's health status.

[0075] In this embodiment, a hierarchical tree is constructed from the vital characteristic data of the drivers, and cluster analysis is performed on the nodes of the hierarchical tree to obtain the characteristic information of each different driver. The cluster overlay center can directly reflect the health status of the drivers. The clustering effect can be effectively understood based on the information evaluation rate of the cluster overlay center, thereby ensuring the accuracy of the clustering analysis and enabling drivers to accurately identify their own health changes.

[0076] To improve the ability to identify the driver's health status, S103 also includes:

[0077] S301, obtain the device data channel for each device;

[0078] Specifically, confirm the connection status of each mobile device bound to the vehicle, and confirm whether each data channel has connection requests to check the driver's health.

[0079] S302, Confirm that there is a target device data channel with a connection request;

[0080] When a mobile device requests a connection, the device data channel corresponding to the connection request is marked as the target device data channel;

[0081] S303, confirms connection permissions by connecting to the target device's data channel;

[0082] Based on the target device's data channel, confirm whether the current device has connection permissions, and simultaneously monitor the current driver's working status;

[0083] S304: Obtain the driver's health monitoring data from each device's data channel and issue an alarm based on the health monitoring data;

[0084] Get real-time health monitoring data and view health analysis reports for the past 15 days to confirm the driver's health status;

[0085] S305 periodically acquires alarm information and sends it back to monitoring personnel;

[0086] If a driver has an abnormal health condition within the past 15 days, the cloud platform will automatically record the time and location of the abnormal data, the characteristics of the abnormal data, and suggestions for improving the driver's health.

[0087] The technical solutions described in the embodiments of this application above have at least the following technical effects or advantages:

[0088] This invention collects vital sign data of drivers and establishes a health feature database based on different time series. Different health conditions are expressed through multiple cluster overlay centers, and different labels are set for the cluster overlay centers. Based on the information evaluation rate and similarity rate obtained from the cluster overlay centers, the characteristics and trends of drivers' health status can be effectively identified, and the health status of drivers can be displayed in a timely manner to reasonably assist healthy driving.

[0089] This invention also provides a health detection system for new energy electric vehicles, comprising:

[0090] The data acquisition module is used to collect vital sign data.

[0091] The communication module is used to send vital sign data to the cloud platform;

[0092] The data analysis module is used to analyze vital sign data and generate health analysis reports;

[0093] The above description is merely a preferred embodiment of the present invention and is not intended to limit the invention. For those skilled in the art, the present invention can have various modifications and variations. Any modifications, equivalent substitutions, improvements, etc., made within the spirit and principles of the present invention should be included within the scope of protection of the present invention.

Claims

1. A health detection method for new energy electric vehicles, characterized in that, include: S101 collects and stores the driver's biometrics and uploads them to the cloud platform; S102, cloud platform for storing and analyzing health data; S103, view health data and health status; Step S102 is implemented in the following manner: S201, Determine the health status of the vehicle driver based on vital sign data; S202, Establish a health feature database, compare the time series data of the cloud platform with the health feature database, and identify the current health status of the driver; The health feature database is based on a time series to build a hierarchical tree, which divides different information into multiple nodes and forms multiple cluster centers with different rules. The rule distance between each node is calculated, and similar nodes are merged. This merging process is repeated until all nodes under the hierarchical tree are merged into one node. Between the parent and child nodes of each hierarchical tree node, the child node contains the feature points of the parent node. When performing superimposed clustering on the child nodes, the corresponding parent nodes of the child nodes are also clustered at the same time. The information evaluation rate and similarity rate are obtained based on the cluster superimposed center. The root node is set to a time range of 15 days. Data from 1 day is designated as the first node, data from 3 days as the second node, data from 7 days as the third node, and data from 15 days as the fourth node. The first node is the direct child of the root node, the second node is the child of the first node, the third node is the child of the second node, and the fourth node is the child of the third node. The time series data with the middle mean of each time series is used as the initial data set; different filtering kernels are applied to each initial data set to obtain multiple health curves; and the initial difference point set is obtained by differencing each adjacent health curve. The difference point set for each time series is obtained by superimposing all the initial difference point sets; The first node of the hierarchical tree for each time series is obtained from the difference point set for each time series; each first node corresponds to a difference point set; each second, third, and fourth node also corresponds to a difference point set; For each time series, the feature points of the set with the largest difference and the set of adjacent difference points are merged into the set with the largest difference to obtain the first cluster center; for each time series, the corresponding set of the largest adjacent difference points is merged with the set of the nearest difference points to obtain the second cluster center; the first cluster center and the second cluster center are merged to obtain the cluster superposition center; and so on, continuously superimposing and merging to obtain the cluster superposition center for each time series. S203: If the driver's health status is normal, continue monitoring; if the driver's health status becomes abnormal for a short period of time, issue a warning according to the type of abnormality and provide emotional support; if the driver's health status becomes abnormal for a long period of time, issue a warning according to the type of health status and send health monitoring data to the cloud platform.

2. The health detection method for a new energy electric vehicle as described in claim 1, characterized in that, Each cluster overlay center is labeled based on the feature points of the maximum difference set.

3. The health detection method for a new energy electric vehicle as described in claim 1, characterized in that, Information entropy is constructed based on the feature points of each cluster superposition center, and entropy curves are generated. Change points are obtained for each entropy curve. The trend of the entropy curve is obtained based on the distance between the feature points of the cluster superposition center corresponding to the difference point set where each change point is located and the feature points of the cluster superposition center of the corresponding maximum difference point set. Based on the entropy curve trend corresponding to each mutation point, the information evaluation rate of the cluster superposition center is obtained; The method for obtaining the mutation point includes: constructing an entropy value sequence based on the entropy value curve; obtaining the entropy value difference by taking the difference between the previous extreme point and the next extreme point in the entropy value sequence; and taking the smaller of the two entropy values ​​that are greater than the average of the entropy value difference as the mutation point of the entropy value curve.

4. The health detection method for a new energy electric vehicle as described in claim 3, characterized in that, The entropy difference between the cluster superposition center corresponding to each mutation point and the cluster superposition center under the corresponding time series is obtained, and the ratio of the entropy difference to the cluster superposition center is used as the information evaluation rate.

5. The health detection method for a new energy electric vehicle as described in claim 3, characterized in that, The similarity rate is calculated as the ratio of the feature points of the cluster superposition center corresponding to each mutation point to the feature points of the largest cluster superposition center in the corresponding time series.

6. The health detection method for a new energy electric vehicle as described in claim 1, characterized in that, The driver's health status is classified and labeled based on the information assessment rate, and health characteristics are obtained based on the similarity rate.

7. The health detection method for a new energy electric vehicle as described in claim 1, characterized in that, Step S103 is achieved in the following manner: S301, obtain the device data channel for each device; S302, Confirm that there is a target device data channel with a connection request; S303, confirms connection permissions by connecting to the target device's data channel; S304: Obtain the driver's health monitoring data from each device's data channel and issue an alarm based on the health monitoring data; S305 periodically acquires alarm information and sends it back to monitoring personnel.

Images