Battery leasing platform regulation and control system and method based on battery big data dynamic analysis

Abstract

The invention discloses a battery lease platform regulation and control system and method based on battery big data dynamic analysis, and relates to the technical field of battery lease, the regulation and control method comprises the following steps: generating a corresponding lease log for any user, and recording each battery lease process of the user to obtain a corresponding lease record; evaluating the battery performance loss of the lease record; extracting user behaviors monitored in the lease record; performing abnormal behavior identification on each user behavior based on the evaluation condition of the battery performance loss in the selected lease record; for any abnormal behavior, analyzing to obtain a correlation condition of the abnormal behavior to the battery performance loss; setting expected battery distribution of the user corresponding to the lease log; whenever the user has a battery leasing behavior, risk assessment is carried out on the battery leasing behavior of the user, and whether an abnormal behavior early warning is sent to the user is judged.

Description

Technical Field

[0001] The present invention relates to the field of battery rental technology, and in particular to a battery rental platform control system and method based on dynamic analysis of battery big data. Background Art

[0002] A battery rental platform is an online platform that provides users with battery rental services, typically covering a variety of battery types such as electric vehicles, electric bicycles, and mobile devices. With the rapid development of the new energy industry, battery rental has become an important energy application model, and more and more industries and users are currently engaging in battery rental.

[0003] When users use batteries, the battery's operating performance and lifespan will be affected to varying degrees, which is closely related to the user's usage behavior. However, the usage habits and behaviors of different users vary greatly, making it difficult to formulate a unified standard for identifying abnormal behavior. The existing battery rental platform has fixed rental rules. If the battery rental is improperly allocated, it will seriously affect the battery's service life. Summary of the Invention

[0004] The purpose of the present invention is to provide a battery rental platform control system and method based on dynamic analysis of battery big data to solve the problems raised in the prior art.

[0005] To achieve the above objectives, the present invention provides the following technical solution: a battery rental platform control method based on dynamic analysis of battery big data, the control method comprising the following steps:

[0006] Step S100: In the constructed rental data management system, a corresponding rental log is generated for any user. Several sensors are installed inside each battery to record each user's battery rental process, thereby obtaining a corresponding rental record. The battery performance changes presented in any rental record are analyzed to evaluate the battery performance loss of the rental record.

[0007] Step S200: Selecting a rental record from any rental log and extracting user behaviors monitored in the rental record; identifying abnormal behaviors of each user based on the evaluation of battery performance loss in the selected rental record to obtain a set of abnormal behaviors of the user corresponding to the rental log;

[0008] Step S300: For any abnormal behavior, analyze the battery performance changes presented in each rental record in the rental log to obtain the correlation between the abnormal behavior and battery performance loss; based on the occurrence of each abnormal behavior in any rental log, set the expected battery allocation for the user corresponding to the rental log;

[0009] Step S400: Whenever a user rents a battery, a risk assessment is performed on the user's battery rental behavior based on the occurrence of each abnormal behavior in the user's abnormal behavior set; based on the risk assessment result and the user's expected battery allocation, a decision is made as to whether to send an abnormal behavior warning to the user.

[0010] Furthermore, step S100 includes the following steps:

[0011] Step S101: When a user rents a battery on the battery rental platform for the first time, a rental log of the user is generated in the rental data management system after the user authorizes the information; each time the user rents a battery, the battery rental platform randomly allocates a battery to the user and generates a corresponding rental record in the rental log;

[0012] Step S102: Before the retrieved battery is allocated to the user, the battery rental platform reads the battery operating data monitored by various sensors in the battery and records it in the corresponding rental record; when the user returns the battery, the battery operating data monitored by various sensors in the battery are read again and stored in the rental record;

[0013] Step S103: preset evaluation rules for several dimensions of the battery's operating performance, wherein there is a sensor matching any dimension; extract the operating data monitored twice by the same sensor in the rental record, obtain two performance values respectively according to the preset evaluation rules, and calculate the difference between the two performance values of each dimension to obtain the loss evaluation value of each dimension; sum the loss evaluation values of each dimension to obtain the loss evaluation value of the rental record, preset a loss threshold, and if the loss evaluation value of the rental record exceeds the loss threshold, the rental record is set as an abnormal rental record; the sensors inside the battery include voltage sensors, temperature sensors, etc., and the corresponding dimensions include voltage, temperature, etc., which correspond one to one; and the preset evaluation rules include the deviation between the actual voltage value and the expected voltage range, etc., which can reflect the performance changes.

[0014] Furthermore, step S200 includes the following steps:

[0015] Step S201: Extract a rental log at random, select a rental record at random from the rental log, and obtain the battery operation data monitored by each sensor inside the battery at each time point; arbitrarily select a sensor, establish a two-dimensional rectangular coordinate system with the time point as the horizontal coordinate and the operation data as the vertical coordinate to present the operation status of the selected sensor; the battery operation data monitored by each sensor is determined by the sensor type itself, for example, a voltage sensor monitors the battery operation data related to voltage;

[0016] Step S202: Set the operating data at the ath time point to D a , get the difference ΔD between the running data at the ath time point and the a+1th time point a,a+1 ; Get the running data difference between any two adjacent time points, and sort all the running data differences according to the order of the difference time points;

[0017] Step S203: Establish a time window with a window length of L, set the initial length of the time window to L=L0, place one end of the time window on the first running data difference after sorting, and calculate the average value of all running data differences in the time window to obtain the average difference ΔD ave ;

[0018] Step S204: Preset a deviation threshold σ. If any running data difference ΔD within the time window satisfies |ΔD-ΔD ave | / ΔD ave <σ, then adjust the window length of the time window to L=L0+1, and obtain the running data difference ΔD between the L0th time point and the L0+1th time point (L0,L0+1) , if the running data difference ΔD (L0,L0+1) Also satisfies |ΔD (L0,L0+1) -ΔD ave | / ΔD ave <σ, the length of the time window will continue to increase until a difference in operating data does not meet the above conditions, and the time interval corresponding to the entire time window will be set as the time interval of a characteristic behavior; every time the time interval of a characteristic behavior is set, one end of the time window will be moved to the end of the latest set time interval, and the search for the next time interval of the characteristic behavior will continue to obtain the user behavior set corresponding to the selected sensor in the selected rental record;

[0019] By analyzing the changes in battery operating data detected by any sensor within a time interval through a time window, the data change process for a user behavior should tend to be stable. Conversely, the time interval in which the data difference tends to be stable also belongs to the same user behavior, such as continuous charging or continuous discharging. Changing user behavior will inevitably cause significant changes in operating data. Therefore, user behavior can be directly judged based on the internal operating data of the battery;

[0020] Step S205: Obtain the characteristic behavior set corresponding to each sensor in any rental record. Establish a new two-dimensional rectangular coordinate system with the time point as the horizontal coordinate and the type of each sensor as the vertical coordinate. Present the time interval division of the characteristic behavior set corresponding to each sensor, and obtain the characteristic behavior division of any rental record.

[0021] Step S206: Randomly select a rental record. If the selected rental record is not an abnormal rental record, select a characteristic behavior of the maximum time interval from any dimension, obtain the time interval of the corresponding characteristic behavior of other dimensions under the maximum time interval, and set the maximum time interval as ΔT max The time interval corresponding to the i-th dimension is ΔT i , the allowable deviation of the i-th dimension is calculated to be η i =(ΔT max -ΔT i ) / ΔT max , and the maximum time interval ΔT max The corresponding sensors monitor the battery operation data and summarize them to obtain a user behavior of the selected rental record;

[0022] The stability of operational data monitored by different sensors within the same time period may vary. Although fluctuations are classified within a single sensor, sustained and stable changes must be attributed to a single user behavior. Therefore, prioritizing the characteristic behavior with the largest time interval for user behavior identification can help eliminate occasional fluctuations from other sensors and provide more accurate identification of user behavior.

[0023] Step S207: arbitrarily select two user behaviors, preset corresponding similarity comparison rules for the operation data in each dimension, and obtain the similarity in any dimension; calculate the average value of the similarity in each dimension to obtain the similarity between the two selected user behaviors; preset a similarity threshold, and if the similarity between the two user behaviors exceeds the similarity threshold, then the two selected user behaviors are set as similar user behaviors;

[0024] Step S208: Randomly select an abnormal rental record, extract a user behavior from the selected abnormal rental record, obtain the time interval of the corresponding characteristic behavior of the user behavior in each dimension; set the maximum time interval of the corresponding characteristic behavior to ΔT ’ max The time interval corresponding to the i-th dimension is ΔT i ’ , if η i ’ =(ΔT ’ max -ΔT i ’ ) / ΔT ’ max >η i , the extracted user behavior is identified as abnormal behavior; all abnormal behaviors in each abnormal rental record are identified and summarized to obtain the abnormal behavior set of the rental log where each abnormal rental record is located.

[0025] Furthermore, step S300 includes the following steps:

[0026] Step S301: Randomly select a user's rental log, randomly select an abnormal behavior from the abnormal behavior set in the rental log, extract the abnormal rental records containing the selected abnormal behavior from the rental log, and obtain a target abnormal record set; set the number of rental records in the target abnormal record set to N1, and obtain the occurrence frequency of the selected abnormal behavior f = N1 / N, where N is the total number of rental records in the rental log;

[0027] Step S302: randomly select a target abnormal record from the target abnormal record set, and obtain the loss assessment value S of the target abnormal record; set the loss threshold to S th , the loss difference of the target abnormal record is ΔS=SS th ; Obtain the deviation difference of each dimension in the target abnormal record whose time interval does not meet the allowable deviation range, wherein the deviation difference of the i-th dimension is set to Δη=η i ’ -η i , calculate the average deviation difference Δη of the target abnormal record ave , we get the correlation degree of the selected abnormal behavior G = ΔS × f × Δη ave To determine the extent of battery performance loss caused by an abnormal event, it is necessary to consider the extent of loss caused by each abnormal rental record and the deviation under the abnormal dimension. Then, by using the frequency of abnormal behavior, the impact of abnormal behavior on the user's rental battery can be effectively determined.

[0028] Step S303: Obtain the correlation of each abnormal behavior in the rental log, and sum them up to obtain the expected battery loss G of the rental log. total ; According to the evaluation rules, each battery in the battery rental platform is evaluated to obtain the performance value X of each battery, and the performance value of the abnormal battery is set to X yc , if there exists a target battery that satisfies XG total >X yc , the target battery is set as a desired battery of the user.

[0029] Furthermore, step S400 includes the following steps:

[0030] Step S401: When a user rents a battery, the occurrence frequency of each abnormal behavior in the user's abnormal behavior set is obtained, and the correlation degree of each abnormal behavior is obtained; all correlation degrees are summed to obtain the user's expected battery loss G ’ total ;

[0031] Step S402: Obtain the current performance values of each battery on the battery rental platform and calculate the average value to obtain an average performance value X ave , set the battery abnormality performance value to X yc , if X yc +G ’ total >X ave , an abnormal behavior warning will be sent to the user.

[0032] In order to better implement the above method, a battery rental platform control system is also proposed. The control system includes a rental loss analysis module, a user behavior analysis module, a battery rental allocation module, and an abnormal rental analysis module.

[0033] The rental loss analysis module is used to generate a corresponding rental log for any user in the constructed rental data management system. Several sensors are installed inside each battery to record each user's battery rental process, generating a corresponding rental record. The module analyzes the battery performance changes presented in any rental record and evaluates the battery performance loss of the rental record.

[0034] A user behavior analysis module is used to select a rental record from any rental log and extract the user behavior monitored in the rental record; based on the evaluation of battery performance loss in the selected rental record, it identifies abnormal behavior of each user and obtains a set of abnormal behaviors corresponding to the user in the rental log;

[0035] A battery rental allocation module is configured to analyze the battery performance changes presented in each rental record in the rental log for any abnormal behavior, and determine the correlation between the abnormal behavior and battery performance loss; and to set the expected battery allocation for the user corresponding to the rental log based on the occurrence of each abnormal behavior in any rental log;

[0036] The abnormal rental analysis module is used to perform a risk assessment on the user's battery rental behavior based on the occurrence of each abnormal behavior in the user's abnormal behavior set whenever the user rents a battery; based on the risk assessment results and the user's expected battery allocation, it determines whether to send an abnormal behavior warning to the user.

[0037] Furthermore, the lease loss analysis module includes a lease record collection unit and a performance loss evaluation unit;

[0038] The rental record collection unit is used to generate a corresponding rental log for any user in the constructed rental data management system. Several sensors are installed inside each battery to record each battery rental process of the user and obtain a corresponding rental record; the performance loss evaluation unit is used to analyze the changes in battery performance presented in any rental record and evaluate the battery performance loss of the rental record.

[0039] Furthermore, the user behavior analysis module includes a user behavior extraction unit and an abnormal behavior analysis unit;

[0040] The user behavior extraction unit is used to select any rental record from any rental log and extract the user behavior monitored in the rental record; the abnormal behavior analysis unit is used to identify abnormal behavior of each user based on the evaluation of battery performance loss in the selected rental record, and obtain the abnormal behavior set of the user corresponding to the rental log.

[0041] Furthermore, the battery rental allocation module includes an abnormal behavior association unit and a rental allocation analysis unit;

[0042] The abnormal behavior association unit is used to analyze the battery performance changes presented in each rental record in the rental log for any abnormal behavior, and obtain the correlation between the abnormal behavior and battery performance loss; the rental allocation analysis unit is used to set the expected battery allocation for the user corresponding to the rental log based on the occurrence of each abnormal behavior in any rental log.

[0043] Furthermore, the abnormal lease analysis module includes a lease risk assessment unit and an abnormal lease reminder unit;

[0044] The rental risk assessment unit is used to perform a risk assessment on the user's battery rental behavior based on the occurrence of each abnormal behavior in the user's abnormal behavior set whenever the user rents a battery; the abnormal rental reminder unit is used to determine whether to send an abnormal behavior warning to the user based on the risk assessment results and the user's expected battery allocation.

[0045] Compared with the prior art, the present invention has the following beneficial effects:

[0046] 1. This invention monitors a series of user behaviors after renting batteries, analyzes the impact of user behaviors on battery performance, and understands user usage and behavioral habits. It can help rental platforms allocate suitable batteries to users when renting batteries, optimize battery life and user rental efficiency.

[0047] 2. The present invention identifies abnormal user behavior by analyzing changes in battery operating data across various dimensions. This can be accomplished solely through the battery's built-in sensors, eliminating the need for more complex monitoring equipment. This allows for more efficient resource utilization while ensuring accuracy.

[0048] 3. The present invention significantly reduces the occurrence of abnormalities during battery use caused by random rentals, and can help users match the most suitable battery when renting a battery. At the same time, it can also issue early warnings to users who frequently exhibit abnormal behaviors, thereby increasing the battery life and ensuring the efficiency of both the platform and users. BRIEF DESCRIPTION OF THE DRAWINGS

[0049] Figure 1 A schematic diagram of the steps of a battery rental platform control method based on dynamic analysis of battery big data;

[0050] Figure 2 This is a structural diagram of the battery rental platform control system based on dynamic analysis of battery big data. DETAILED DESCRIPTION

[0051] The following will clearly and completely describe the technical solutions in the embodiments of the present invention in conjunction with the accompanying drawings. 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 making creative efforts are within the scope of protection of the present invention.

[0052] Example: Figures 1 to 2 As shown, the present invention provides a battery rental platform control method based on dynamic analysis of battery big data, and the control method includes the following steps:

[0053] Step S100: In the constructed rental data management system, a corresponding rental log is generated for any user. Several sensors are installed inside each battery to record each user's battery rental process, thereby obtaining a corresponding rental record. The battery performance changes presented in any rental record are analyzed to evaluate the battery performance loss of the rental record.

[0054] Wherein, step S100 includes the following steps:

[0055] Step S101: When a user rents a battery on the battery rental platform for the first time, a rental log of the user is generated in the rental data management system after the user authorizes the information; each time the user rents a battery, the battery rental platform randomly allocates a battery to the user and generates a corresponding rental record in the rental log;

[0056] Step S102: Before the retrieved battery is allocated to the user, the battery rental platform reads the battery operating data monitored by various sensors in the battery and records it in the corresponding rental record; when the user returns the battery, the battery operating data monitored by various sensors in the battery are read again and stored in the rental record;

[0057] Step S103: Preset evaluation rules for several dimensions of the battery's operating performance, where there is a sensor that matches any dimension; extract the operating data of the same sensor monitored twice in the rental record, obtain two performance values respectively according to the preset evaluation rules, and calculate the difference between the two performance values of each dimension to obtain the loss evaluation value of each dimension; sum the loss evaluation values of each dimension to obtain the loss evaluation value of the rental record, preset a loss threshold, and if the loss evaluation value of the rental record exceeds the loss threshold, the rental record is set as an abnormal rental record.

[0058] Step S200: Selecting a rental record from any rental log and extracting user behaviors monitored in the rental record; identifying abnormal behaviors of each user based on the evaluation of battery performance loss in the selected rental record to obtain a set of abnormal behaviors of the user corresponding to the rental log;

[0059] Wherein, step S200 includes the following steps:

[0060] Step S201: extracting a rental log at random, selecting a rental record at random from the rental log, and acquiring the battery operation data monitored by various sensors inside the battery at various time points; randomly selecting a sensor, using the time point as the horizontal coordinate and the operation data as the vertical coordinate, to establish a two-dimensional rectangular coordinate system to present the operation status of the selected sensor;

[0061] Step S202: Set the operating data at the ath time point to D a , get the difference ΔD between the running data at the ath time point and the a+1th time point a,a+1 ; Get the running data difference between any two adjacent time points, and sort all the running data differences according to the order of the difference time points;

[0062] Example 1: The difference in running data between the first time point and the second time point is D 1,2 , the difference in running data between the second time point and the third time point is D 2,3 , the difference in running data between the 3rd time point and the 4th time point is D 3,4 , so the sorting of running data difference is D 1,2 、D 2,3 、D3,4 ;

[0063] Step S203: Establish a time window with a window length of L, set the initial length of the time window to L=L0, place one end of the time window on the first running data difference after sorting, and calculate the average value of all running data differences in the time window to obtain the average difference ΔD ave ;

[0064] Step S204: Preset a deviation threshold σ. If any running data difference ΔD within the time window satisfies |ΔD-ΔD ave | / ΔD ave <σ, then adjust the window length of the time window to L=L0+1, and obtain the running data difference ΔD between the L0th time point and the L0+1th time point (L0,L0+1) , if the running data difference ΔD (L0,L0+1) Also satisfies |ΔD (L0,L0+1) -ΔD ave | / ΔD ave <σ, the length of the time window will continue to increase until a difference in operating data does not meet the above conditions, and the time interval corresponding to the entire time window will be set as the time interval of a characteristic behavior; every time the time interval of a characteristic behavior is set, one end of the time window will be moved to the end of the latest set time interval, and the search for the next time interval of the characteristic behavior will continue to obtain the user behavior set corresponding to the selected sensor in the selected rental record;

[0065] Step S205: Obtain the characteristic behavior set corresponding to each sensor in any rental record. Establish a new two-dimensional rectangular coordinate system with the time point as the horizontal coordinate and the type of each sensor as the vertical coordinate. Present the time interval division of the characteristic behavior set corresponding to each sensor, and obtain the characteristic behavior division of any rental record.

[0066] Step S206: Randomly select a rental record. If the selected rental record is not an abnormal rental record, select a characteristic behavior of the maximum time interval from any dimension, obtain the time interval of the corresponding characteristic behavior of other dimensions under the maximum time interval, and set the maximum time interval as ΔT max The time interval corresponding to the i-th dimension is ΔT i , the allowable deviation of the i-th dimension is calculated to be η i =(ΔT max -ΔT i ) / ΔT max , and the maximum time interval ΔT max The corresponding sensors monitor the battery operation data and summarize them to obtain a user behavior of the selected rental record;

[0067] Step S207: arbitrarily select two user behaviors, preset corresponding similarity comparison rules for the operation data in each dimension, and obtain the similarity in any dimension; calculate the average value of the similarity in each dimension to obtain the similarity between the two selected user behaviors; preset a similarity threshold, and if the similarity between the two user behaviors exceeds the similarity threshold, then the two selected user behaviors are set as similar user behaviors;

[0068] Step S208: Randomly select an abnormal rental record, extract a user behavior from the selected abnormal rental record, obtain the time interval of the corresponding characteristic behavior of the user behavior in each dimension; set the maximum time interval of the corresponding characteristic behavior to ΔT ’ max The time interval corresponding to the i-th dimension is ΔT i ’ , if η i ’ =(ΔT ’ max -ΔT i ’ ) / ΔT ’ max >η i , the extracted user behavior is identified as abnormal behavior; all abnormal behaviors in each abnormal rental record are identified and summarized to obtain the abnormal behavior set of the rental log where each abnormal rental record is located.

[0069] Step S300: For any abnormal behavior, analyze the battery performance changes presented in each rental record in the rental log to obtain the correlation between the abnormal behavior and battery performance loss; based on the occurrence of each abnormal behavior in any rental log, set the expected battery allocation for the user corresponding to the rental log;

[0070] Wherein, step S300 includes the following steps:

[0071] Step S301: Randomly select a user's rental log, randomly select an abnormal behavior from the abnormal behavior set in the rental log, extract the abnormal rental records containing the selected abnormal behavior from the rental log, and obtain a target abnormal record set; set the number of rental records in the target abnormal record set to N1, and obtain the occurrence frequency of the selected abnormal behavior f = N1 / N, where N is the total number of rental records in the rental log;

[0072] Step S302: randomly select a target abnormal record from the target abnormal record set, and obtain the loss assessment value S of the target abnormal record; set the loss threshold to S th, the loss difference of the target abnormal record is ΔS=SS th ; Obtain the deviation difference of each dimension in the target abnormal record whose time interval does not meet the allowable deviation range, wherein the deviation difference of the i-th dimension is set to Δη=η i ’ -η i , calculate the average deviation difference Δη of the target abnormal record ave , we get the correlation degree of the selected abnormal behavior G = ΔS × f × Δη ave ;

[0073] Example 2: Assume that one of the user's abnormal behaviors is high-temperature usage, and the user's rental log contains 8 abnormal rental records of high-temperature usage, and there are a total of 10 rental records in the rental log. The occurrence frequency f = 0.8; the loss difference of a target abnormal record is set to 10, and the average deviation difference is 0.2. The correlation G = 10 × 0.2 × 0.8 = 1.6;

[0074] Step S303: Obtain the correlation of each abnormal behavior in the rental log, and sum them up to obtain the expected battery loss G of the rental log. total ; According to the evaluation rules, each battery in the battery rental platform is evaluated to obtain the performance value X of each battery, and the performance value of the abnormal battery is set to X yc , if there exists a target battery that satisfies XG total >X yc , the target battery is set as a desired battery of the user.

[0075] Step S400: Whenever a user rents a battery, a risk assessment is performed on the user's battery rental behavior based on the occurrence of each abnormal behavior in the user's abnormal behavior set; based on the risk assessment result and the user's expected battery allocation, a decision is made as to whether to send an abnormal behavior warning to the user;

[0076] Step S400 includes the following steps:

[0077] Step S401: When a user rents a battery, the occurrence frequency of each abnormal behavior in the user's abnormal behavior set is obtained, and the correlation degree of each abnormal behavior is obtained; all correlation degrees are summed to obtain the user's expected battery loss G ’ total ;

[0078] Step S402: Obtain the current performance values of each battery on the battery rental platform and calculate the average value to obtain an average performance value X ave , set the battery abnormality performance value to X yc , if X yc +G’ total >X ave , an abnormal behavior warning will be sent to the user.

[0079] The battery rental platform control system includes a rental loss analysis module, a user behavior analysis module, a battery rental allocation module, and an abnormal rental analysis module;

[0080] The rental loss analysis module is used to generate a corresponding rental log for any user in the constructed rental data management system. Several sensors are installed inside each battery to record each user's battery rental process, generating a corresponding rental record. The module analyzes the battery performance changes presented in any rental record and evaluates the battery performance loss of the rental record.

[0081] A user behavior analysis module is used to select a rental record from any rental log and extract the user behavior monitored in the rental record; based on the evaluation of battery performance loss in the selected rental record, it identifies abnormal behavior of each user and obtains a set of abnormal behaviors corresponding to the user in the rental log;

[0082] A battery rental allocation module is configured to analyze the battery performance changes presented in each rental record in the rental log for any abnormal behavior, and determine the correlation between the abnormal behavior and battery performance loss; and to set the expected battery allocation for the user corresponding to the rental log based on the occurrence of each abnormal behavior in any rental log;

[0083] The abnormal rental analysis module is used to perform a risk assessment on the user's battery rental behavior based on the occurrence of each abnormal behavior in the user's abnormal behavior set whenever the user rents a battery; based on the risk assessment results and the user's expected battery allocation, it determines whether to send an abnormal behavior warning to the user.

[0084] Among them, the lease loss analysis module includes a lease record collection unit and a performance loss evaluation unit;

[0085] The rental record collection unit is used to generate a corresponding rental log for any user in the constructed rental data management system. Several sensors are installed inside each battery to record each battery rental process of the user and obtain a corresponding rental record; the performance loss evaluation unit is used to analyze the changes in battery performance presented in any rental record and evaluate the battery performance loss of the rental record.

[0086] Among them, the user behavior analysis module includes a user behavior extraction unit and an abnormal behavior analysis unit;

[0087] The user behavior extraction unit is used to select any rental record from any rental log and extract the user behavior monitored in the rental record; the abnormal behavior analysis unit is used to identify abnormal behavior of each user based on the evaluation of battery performance loss in the selected rental record, and obtain the abnormal behavior set of the user corresponding to the rental log.

[0088] Among them, the battery rental allocation module includes an abnormal behavior association unit and a rental allocation analysis unit;

[0089] The abnormal behavior association unit is used to analyze the battery performance changes presented in each rental record in the rental log for any abnormal behavior, and obtain the correlation between the abnormal behavior and battery performance loss; the rental allocation analysis unit is used to set the expected battery allocation for the user corresponding to the rental log based on the occurrence of each abnormal behavior in any rental log.

[0090] Among them, the abnormal lease analysis module includes a lease risk assessment unit and an abnormal lease reminder unit;

[0091] The rental risk assessment unit is used to perform a risk assessment on the user's battery rental behavior based on the occurrence of each abnormal behavior in the user's abnormal behavior set whenever the user rents a battery; the abnormal rental reminder unit is used to determine whether to send an abnormal behavior warning to the user based on the risk assessment results and the user's expected battery allocation.

[0092] It will be apparent to those skilled in the art that the present invention is not limited to the details of the exemplary embodiments described above and that the invention can be embodied in other specific forms without departing from the spirit or essential characteristics of the invention. Therefore, the embodiments should be considered in all respects as illustrative and non-restrictive, and the scope of the invention is defined by the appended claims, not the foregoing description, and all variations within the meaning and range of equivalents of the claims are intended to be included therein. Any reference sign in a claim should not be construed as limiting the claim to which it relates.

Claims

1. A battery rental platform control method based on dynamic analysis of battery big data, characterized by: The control method comprises the following steps: Step S100: In the constructed rental data management system, a corresponding rental log is generated for any user. Several sensors are installed inside each battery to record each user's battery rental process, thereby obtaining a corresponding rental record. The battery performance changes presented in any rental record are analyzed to evaluate the battery performance loss of the rental record. Step S200: Selecting a rental record from any rental log and extracting user behaviors monitored in the rental record; identifying abnormal behaviors of each user based on the evaluation of battery performance loss in the selected rental record to obtain a set of abnormal behaviors of the user corresponding to the rental log; Step S300: For any abnormal behavior, analyze the battery performance changes presented in each rental record in the rental log to obtain the correlation between the abnormal behavior and battery performance loss; based on the occurrence of each abnormal behavior in any rental log, set the expected battery allocation for the user corresponding to the rental log; Step S400: Whenever a user rents a battery, a risk assessment is performed on the user's battery rental behavior based on the occurrence of each abnormal behavior in the user's abnormal behavior set; based on the risk assessment result and the user's expected battery allocation, a decision is made as to whether to send an abnormal behavior warning to the user.

2. The battery rental platform control method based on dynamic analysis of battery big data according to claim 1 is characterized by: The step S100 includes the following steps: Step S101: When a user rents a battery on the battery rental platform for the first time, a rental log of the user is generated in the rental data management system after the user authorizes the information; each time the user rents a battery, the battery rental platform randomly allocates a battery to the user and generates a corresponding rental record in the rental log; Step S102: Before the retrieved battery is allocated to the user, the battery rental platform reads the battery operating data monitored by various sensors in the battery and records it in the corresponding rental record; when the user returns the battery, the battery operating data monitored by various sensors in the battery are read again and stored in the rental record; Step S103: Preset evaluation rules for several dimensions of the battery's operating performance, where there is a sensor that matches any dimension; extract the operating data of the same sensor monitored twice in the rental record, obtain two performance values respectively according to the preset evaluation rules, and calculate the difference between the two performance values of each dimension to obtain the loss evaluation value of each dimension; sum the loss evaluation values of each dimension to obtain the loss evaluation value of the rental record, preset a loss threshold, and if the loss evaluation value of the rental record exceeds the loss threshold, the rental record is set as an abnormal rental record.

3. The battery rental platform control method based on dynamic analysis of battery big data according to claim 2 is characterized by: The step S200 includes the following steps: Step S201: extracting a rental log at random, selecting a rental record at random from the rental log, and acquiring the battery operation data monitored by various sensors inside the battery at various time points; randomly selecting a sensor, using the time point as the horizontal coordinate and the operation data as the vertical coordinate, to establish a two-dimensional rectangular coordinate system to present the operation status of the selected sensor; Step S202: Set the operating data at the ath time point to D a , get the difference ΔD between the running data at the ath time point and the a+1th time point a,a+1 ; Get the running data difference between any two adjacent time points, and sort all the running data differences according to the order of the difference time points; Step S203: Establish a time window with a window length of L, set the initial length of the time window to L=L0, place one end of the time window on the first running data difference after sorting, and calculate the average value of all running data differences in the time window to obtain the average difference ΔD ave ; Step S204: Preset a deviation threshold σ. If any running data difference ΔD within the time window satisfies |ΔD-ΔD ave | / ΔD ave <σ, then adjust the window length of the time window to L=L0+1, and obtain the running data difference ΔD between the L0th time point and the L0+1th time point (L0,L0+1) , if the running data difference ΔD (L0,L0+1) Also satisfies |ΔD (L0,L0+1) -ΔD ave | / ΔD ave <σ, the length of the time window will continue to increase until a difference in operating data does not meet the above conditions, and the time interval corresponding to the entire time window will be set as the time interval of a characteristic behavior; every time the time interval of a characteristic behavior is set, one end of the time window will be moved to the end of the latest set time interval, and the search for the next time interval of the characteristic behavior will continue to obtain the user behavior set corresponding to the selected sensor in the selected rental record; Step S205: Obtain the characteristic behavior set corresponding to each sensor in any rental record. Establish a new two-dimensional rectangular coordinate system with the time point as the horizontal coordinate and the type of each sensor as the vertical coordinate. Present the time interval division of the characteristic behavior set corresponding to each sensor, and obtain the characteristic behavior division of any rental record. Step S206: Randomly select a rental record. If the selected rental record is not an abnormal rental record, select a characteristic behavior of the maximum time interval from any dimension, obtain the time interval of the corresponding characteristic behavior of other dimensions under the maximum time interval, and set the maximum time interval as ΔT max The time interval corresponding to the i-th dimension is ΔT i , the allowable deviation of the i-th dimension is calculated to be η i =(ΔT max -ΔT i ) / ΔT max , and the maximum time interval ΔT max The corresponding sensors monitor the battery operation data and summarize them to obtain a user behavior of the selected rental record; Step S207: arbitrarily select two user behaviors, preset corresponding similarity comparison rules for the operation data in each dimension, and obtain the similarity in any dimension; calculate the average value of the similarity in each dimension to obtain the similarity between the two selected user behaviors; preset a similarity threshold, and if the similarity between the two user behaviors exceeds the similarity threshold, then the two selected user behaviors are set as similar user behaviors; Step S208: Randomly select an abnormal rental record, extract a user behavior from the selected abnormal rental record, obtain the time interval of the corresponding characteristic behavior of the user behavior in each dimension; set the maximum time interval of the corresponding characteristic behavior to ΔT ’ max The time interval corresponding to the i-th dimension is ΔT i ’ , if η i ’ =(ΔT ’ max -ΔT i ’ ) / ΔT ’ max >η i , the extracted user behavior is identified as abnormal behavior; all abnormal behaviors in each abnormal rental record are identified and summarized to obtain the abnormal behavior set of the rental log where each abnormal rental record is located.

4. The battery rental platform control method based on dynamic analysis of battery big data according to claim 3 is characterized by: The step S300 includes the following steps: Step S301: Randomly select a user's rental log, randomly select an abnormal behavior from the abnormal behavior set in the rental log, extract the abnormal rental records containing the selected abnormal behavior from the rental log, and obtain a target abnormal record set; set the number of rental records in the target abnormal record set to N1, and obtain the occurrence frequency of the selected abnormal behavior f = N1 / N, where N is the total number of rental records in the rental log; Step S302: randomly select a target abnormal record from the target abnormal record set, and obtain the loss assessment value S of the target abnormal record; set the loss threshold to S th , the loss difference of the target abnormal record is ΔS=SS th ; Obtain the deviation difference of each dimension in the target abnormal record whose time interval does not meet the allowable deviation range, wherein the deviation difference of the i-th dimension is set to Δη=η i ’ -η i , calculate the average deviation difference Δη of the target abnormal record ave , we get the correlation degree of the selected abnormal behavior G = ΔS × f × Δη ave ; Step S303: Obtain the correlation of each abnormal behavior in the rental log, and sum them up to obtain the expected battery loss G of the rental log. total ; According to the evaluation rules, each battery in the battery rental platform is evaluated to obtain the performance value X of each battery, and the performance value of the abnormal battery is set to X yc , if there exists a target battery that satisfies XG total >X yc , the target battery is set as a desired battery of the user.

5. The battery rental platform control method based on dynamic analysis of battery big data according to claim 4 is characterized by: The step S400 includes the following steps: Step S401: When a user rents a battery, the occurrence frequency of each abnormal behavior in the user's abnormal behavior set is obtained, and the correlation degree of each abnormal behavior is obtained; all correlation degrees are summed to obtain the user's expected battery loss G ’ total ; Step S402: Obtain the current performance values of each battery on the battery rental platform and calculate the average value to obtain an average performance value X ave , set the battery abnormality performance value to X yc , if X yc +G ’ total >X ave , an abnormal behavior warning will be sent to the user.

6. A battery rental platform control system, configured to execute the battery rental platform control method based on dynamic analysis of battery big data according to any one of claims 1 to 5, characterized in that: The control system includes a rental loss analysis module, a user behavior analysis module, a battery rental allocation module and an abnormal rental analysis module; The rental loss analysis module is used to generate a corresponding rental log for any user in the constructed rental data management system. Several sensors are installed in each battery to record each battery rental process of the user, thereby obtaining a corresponding rental record; the battery performance changes presented in any rental record are analyzed to evaluate the battery performance loss of the rental record; The user behavior analysis module is configured to select a rental record from any rental log and extract the user behavior monitored in the rental record; based on the evaluation of battery performance loss in the selected rental record, identify abnormal behavior of each user to obtain a set of abnormal behaviors corresponding to the user in the rental log; The battery rental allocation module is configured to analyze the battery performance changes presented in each rental record in the rental log for any abnormal behavior, and determine the correlation between the abnormal behavior and battery performance loss; and to set the expected battery allocation for the user corresponding to the rental log based on the occurrence of each abnormal behavior in any rental log; The abnormal rental analysis module is used to perform a risk assessment on the user's battery rental behavior based on the occurrence of each abnormal behavior in the user's abnormal behavior set whenever the user rents a battery; Based on the risk assessment results and the user's expected battery allocation, determine whether to send an abnormal behavior warning to the user.

7. The battery rental platform control system according to claim 6, characterized in that: The lease loss analysis module includes a lease record collection unit and a performance loss evaluation unit; The rental record collection unit is used to generate a corresponding rental log for any user in the constructed rental data management system. Several sensors are installed inside each battery to record each battery rental process of the user and obtain a corresponding rental record; The performance loss evaluation unit is used to analyze the battery performance changes presented in any rental record and evaluate the battery performance loss of the rental record.

8. The battery rental platform control system according to claim 6, characterized in that: The user behavior analysis module includes a user behavior extraction unit and an abnormal behavior analysis unit; The user behavior extraction unit is used to select any rental record from any rental log and extract the user behavior monitored in the rental record; the abnormal behavior analysis unit is used to identify abnormal behavior of each user based on the evaluation of battery performance loss in the selected rental record, and obtain a set of abnormal behaviors of the user corresponding to the rental log.

9. The battery rental platform control system according to claim 6, characterized in that: The battery rental allocation module includes an abnormal behavior association unit and a rental allocation analysis unit; The abnormal behavior association unit is used to analyze the battery performance changes presented in each rental record in the rental log for any abnormal behavior, and obtain the association between the abnormal behavior and battery performance loss; the rental allocation analysis unit is used to set the expected battery allocation for the user corresponding to the rental log based on the occurrence of each abnormal behavior in any rental log.

10. The battery rental platform control system according to claim 6, characterized in that: The abnormal lease analysis module includes a lease risk assessment unit and an abnormal lease reminder unit; The rental risk assessment unit is configured to perform a risk assessment on the user's battery rental behavior based on the occurrence of each abnormal behavior in the user's abnormal behavior set whenever the user rents the battery; The abnormal rental reminder unit is used to determine whether to send an abnormal behavior warning to the user based on the risk assessment result and the user's expected battery allocation.

Images