A power battery monitoring system and method based on data analysis

Abstract

The present invention discloses a power battery monitoring system and method based on data analysis, which relates to the technical field of battery monitoring. By obtaining user authorization to access the on-vehicle information of the power battery, the power battery information is acquired, and the electric vehicle is divided into three states: driving, charging, and stopping according to the power battery information; the charging time of the electric vehicle by the user is determined, and a charging reminder is sent to the user; it is judged whether the health state of the power battery meets the user's requirements. If it does not meet the user's requirements, a processing suggestion is provided for the user; if it meets the user's requirements, the health state of the power battery is continuously monitored; combining the actual situation of the user at the demand side, the cruising range and charging speed of the power battery, the state of charge and health state of the power battery are detected, and a processing suggestion for the power battery is generated, thereby improving the utilization rate of the power battery and reducing the environmental impact caused by the recycling and treatment of the power battery.

Description

Technical Field

[0001] The present invention relates to the technical field of battery monitoring, and particularly to a power battery monitoring system and method based on data analysis. Background Technique

[0002] New energy vehicles in our country have ranked first in the world in terms of production and sales for many consecutive years. It is expected that by 2030, the number of new energy vehicles in use will reach 100 million, and the total installed capacity of power batteries will be 6000 GWh, with a very large capacity. Users' requirements for the product quality of power batteries have also increased accordingly. When consumers purchase new energy vehicles, in addition to paying attention to the quality and performance of the products themselves, they also put forward higher requirements for after-sales services. With the continuous progress of battery technology, the performance and lifespan of power batteries have been significantly improved. However, at the same time, higher requirements have been put forward for the maintenance and servicing of batteries. In the early stage of technological development, the quality of power batteries was uneven, and due to the management problems of battery manufacturers, the product quality problems were serious. Currently, the performance monitoring of power batteries only focuses on the power batteries themselves and does not combine the specific needs of users, making it unclear to users at the demand side when they need to deal with power batteries. Therefore, how to monitor the performance of power batteries and generate treatment suggestions in combination with the actual situation of users at the demand side has become an urgent problem to be solved. Summary of the Invention

[0003] The purpose of the present invention is to provide a power battery monitoring system and method based on data analysis to solve the problems raised in the above background technique.

[0004] In one aspect of the present invention, a power battery monitoring method based on data analysis is provided, including:

[0005] S11, accessing the in-vehicle information of the power battery through user authorization, obtaining the power battery information, and dividing the electric vehicle into three states: driving, charging, and stopped according to the power battery information;

[0006] S12, determining the time when the user charges the electric vehicle based on the power battery information and status information of the electric vehicle, and sending a charging reminder to the user;

[0007] S13, judging whether the health status of the power battery meets the user's needs based on the power battery information, electric vehicle status information, and the time distribution of the user charging the electric vehicle. If it does not meet the user's needs, treatment suggestions are proposed for the user; if it meets the user's needs, the health status of the power battery is continuously monitored.

[0008] In step S11, the step of dividing the electric vehicle into three states: driving, charging, and stopped according to the power battery information further includes the following steps:

[0009] Obtain the state of charge of the power battery from the vehicle information, divide the tram state according to the state of charge sequence. If the state of charge remains unchanged, it is determined that the tram is in a stopped state; if the state of charge continues to increase, it is determined that the tram is in a charging state; if the state of charge continues to decrease, it is determined that the tram is in a driving state;

[0010] Analyze the time of the stopped state directly after the driving state. If the time the tram is in the stopped state is not greater than the time threshold, merge the stopped state with the previous adjacent driving state as part of the driving state. If there are two adjacent driving states, merge the two adjacent driving states into one driving state; the time threshold is set according to the user's situation;

[0011] For the stopped state between the charging state and the driving state, merge the stopped state with the previous adjacent charging state as part of the charging state.

[0012] During the driving process of the user, situations where the vehicle needs to stop may occur, such as waiting for a red light, queuing to enter the charging area for charging, etc. Therefore, when the duration of the stopped state after the driving state is short, it needs to be merged with the previous driving state. After merging, there may be a situation of two consecutive driving states. For example, when the user encounters a red light, the tram state changes according to the process of driving, stopping, and driving. First, merge the stopped state with the previous driving state, and then merge the two adjacent driving states;

[0013] During the charging process of the user's tram, after the tram is fully charged, usually the tram cannot be immediately driven out of the charging area, but will stay in the charging area for a period of time. At this time, the stopped state is part of the charging state.

[0014] In step S12, the step of determining the charging time of the user for the tram based on the power battery information and state information of the tram further includes the following steps:

[0015] For the tram in the charging state, if when the tram enters the driving state from the charging state, the state of charge of the power battery does not reach 100%, the charging time of the user for the tram is the duration of the charging state, and the actual charging time of the power battery is the duration of the charging state; if when the tram enters the driving state from the charging state, the state of charge of the power battery reaches 100%, the charging time of the user for the tram is the duration of the charging state, and obtain the time it takes for the state of charge of the power battery in the charging state to increase to 100% as the actual charging time of the power battery.

[0016] The actual charging time of the power battery is used to analyze the charging speed of the electric vehicle; while the time when the user charges the electric vehicle is used to determine how much electrical energy can be supplemented by the electric vehicle in one charging behavior.

[0017] In step S12, the sending of a charging reminder to the user further includes the following steps:

[0018] S41. Obtain the data of the kth group of adjacent driving states and charging states from the historical vehicle information. Let the state of charge of the power battery at the start of the driving state be XSOC1, the state of health of the power battery at the start of the driving state be XSOH1, the state of charge of the power battery at the start of the charging state be CSOC1, and the state of health of the power battery at the start of the charging state be CSOH1; calculate the energy W of the power battery consumed by the user to travel to the charging area 2j , W 2j = Y×(XSOC1×XSOH1 - CSOC1×CSOH1), where Y is the capacity of the power battery and is determined according to the model of the power battery;

[0019] W 2j reflects the minimum energy required for the electric vehicle to continuously run. As long as the remaining energy of the power battery is not less than W 2j , the user can charge the electric vehicle before the energy of the power battery is exhausted, reducing the trouble brought to the user due to the lack of power of the power battery;

[0020] S42. Perform unsupervised classification on the energy of the power battery consumed by the user to travel to the charging area generated from the data of all groups of adjacent driving states and charging states, determine the classification cluster with the most data, and calculate the average value of W 2j in the classification cluster to obtain W2; 2j

[0021] During the daily use of the electric vehicle by the user, there is usually a fixed charging route. Under the fixed charging route, the value of W 2j is similar and accounts for the majority among all W 2j . Determine the energy required for the electric vehicle to continuously run under the charging route through unsupervised clustering;

[0022] S43. Obtain the current state of charge SOC and state of health SOH of the electric vehicle from the vehicle information, and determine the current energy W of the power battery d , W d = Y×XSOC×XSOH, where XSOC is the current SOC of the electric vehicle and XSOH is the current SOH of the electric vehicle; if the current energy W of the power battery d is not greater than K×W2, then send a charging reminder to the user, where K is a scaling factor.

[0023] In step S13, the determination of whether the health state of the power battery meets the user's requirements further includes the following steps:

[0024] S51, Obtain the data of the a-th group of two consecutive charging states from the historical vehicle information. The two consecutive charging states mean that there is no other charging state between the two charging states. Let the state of charge of the power battery at the end of the previous charging state be XSOC2, the health state of the power battery at the end of the previous charging state be XSOH2, the state of charge of the power battery at the start of the next charging state be CSOC2, and the health state of the power battery at the start of the next charging state be CSOH2. Calculate the energy demand W of the driving range for the power battery 3a , W 3a = Y×(XSOC2×XSOH2 - CSOC2×CSOH2);

[0025] S52, Determine the mode of W 3a among all W 3a and perform clustering on all W 3a to determine the clustering cluster with the most data, and calculate the average value of W

[0026] in the clustering cluster to obtain W3; 3a The mode reflects the most common behavior pattern of the user. Since the values of W 3a are not exactly the same, in order to determine the mode of W 3a , similar W 3a can be regarded as the same value, and the W

[0027] with the most similar values is the mode. min ,

[0028] S53, Calculate the first critical value SOH1 of the health state of the power battery 0 Let the time for the user to charge the electric vehicle be T, and calculate the charging time t 0 required to meet the user's driving range requirement at the current health state SOH where P 0 is the average charging speed of the power battery at the health state SOH 0 ; Combining the time for the user to charge the electric vehicle, calculate the probability that the time for the user to charge the electric vehicle meets the energy demand where N is the number of historical data when the electric vehicle is in the charging state, and n0 is the number of charging states with a duration not less than t 0 ; If P{T≥t 0} is not less than THR, where THR is a preset threshold, then enter step S55, otherwise set the second critical value SOH2 min of the health state of the power battery as the current health state SOH0 ;

[0029] S55, decrease the current state of health (SOH) by a fixed step size to obtain SOH 0 , SOH 1 , SOH 2 , …, SOH max ; and determine the average charging speed P 1 , P 2 , …, P max of the power battery after each decrease in the state of health, as well as the charging time t 1 , t 2 , …, t max required to meet the user's driving range demand. Calculate the probability P{T≥t 1}, P{T≥t 2}, …, P{T≥t max} that the charging time of the electric vehicle by the user meets the energy demand after each decrease in the state of health. When the probability that the charging time of the electric vehicle by the user meets the energy demand is less than THR, stop decreasing the state of health. At this time, the number of times of decreasing the state of health is max, and the second critical value SOH2 min of the power battery state of health is SOH max-1 ; take the minimum value of SOH1 min and SOH2 min to obtain the threshold SOH min of the power battery state of health;

[0030] S56, if the current state of health of the power battery is less than SOH min , remind the user to replace the power battery in time; otherwise, suggest that the user replace the power battery when the state of health of the power battery reaches SOH min ; meanwhile, if SOH min = SOH2 min , remind the user that the reason for replacing the power battery is insufficient charging time; if SOH min = SOH1 min , remind the user that the reason for replacing the power battery is that the driving range of the power battery cannot meet the user's requirements.

[0031] As the health state deteriorates, the chemical activity inside the power battery may decrease, resulting in a weakened ability to receive charging. This means that the power battery may not be able to charge at the previous high power state. Determine whether the power battery can obtain sufficient energy within the time that the user can afford according to the user's charging habits; the decline in the health state of the power battery is associated with the reduction in the available capacity of the power battery. After the health state of the power battery deteriorates, the available capacity of the power battery may not be able to meet the user's mileage requirements. At this time, even if the health state has not declined to the commonly used replacement threshold such as 80%, it is necessary to consider replacing the power battery; when the user's mileage requirement is short, even if the health state has declined to the commonly used replacement threshold, without safety problems, the user can choose not to replace the power battery, thereby improving the utilization rate of the power battery and reducing the environmental impact caused by the recycling and treatment of power batteries. The first critical value determines the health state threshold for replacing the power battery from the perspective of the cruising range, and the second critical value determines the health state threshold for replacing the power battery from the perspective of time.

[0032] In step S54, the average charging speed of the power battery is determined through the following steps:

[0033] S61, obtain the state of health soh at the i-th charging state, the state of charge ksoc at the start of the charging state, the esoc at the end of the charging state, and the actual charging time t of the power battery from the historical vehicle information, and calculate the average charging speed p of the power battery under the state of health soh. i the state of charge ksoc at the start of the charging state i the esoc at the end of the charging state i and the actual charging time t of the power battery i calculate the state of health soh i and the average charging speed p of the power battery under the state of health soh i ,

[0034]

[0035] S62, train the support vector regression model of the average charging speed of the power battery according to the average charging speed of the power battery under different health states; set the hyperparameters of the support vector regression, including the regularization parameter C, the insensitive parameter epsilon, the kernel function and the parameters of the kernel function; divide the health state and the average charging speed of the power battery into a training set and a test set, use the health state in the training set as the input and the average charging speed of the power battery as the output to train the model under the set hyperparameters, and verify the model with the health state and the average charging speed data of the power battery in the test set;

[0036] S63, input the health of the power battery into the support vector regression model to obtain the average charging speed of the power battery under the health state.

[0037] The C value is selected through cross-validation to balance the model complexity and generalization ability; the insensitive parameter epsilon defines the tolerance error range of the regression model. When the gap between the regression value and the actual value is within the epsilon range, the loss is not calculated. A larger epsilon can ignore smaller noises but will lead to a decrease in model accuracy; a smaller epsilon makes the model more sensitive to data points. The selection of the kernel function and its parameters has an important impact on the performance of the support vector regression model. The most suitable kernel function and parameters can be selected through grid search.

[0038] In another aspect of the present invention, a power battery monitoring system based on data analysis is provided, including: a battery data acquisition module, a battery data storage module, a data analysis module, and an output module; the output end of the battery data acquisition module is connected to the input end of the battery data storage module for acquiring power battery information; the output end of the battery data storage module is connected to the input end of the data analysis module for storing power battery information and sending it to the data analysis module; the output end of the data analysis module is connected to the input end of the output module, which divides the electric vehicle into three states: driving, charging, and stopping based on the power battery information, and analyzes the electric vehicle state to generate power battery processing suggestions for the user; the output module is used to transmit the power battery processing suggestions generated by the data analysis module to the user.

[0039] The battery data acquisition module further includes a permission acquisition unit, a vehicle information acquisition unit, and a data transmission unit; the permission acquisition unit is used to obtain from the user the access permission to the power battery part of the vehicle information; the vehicle information acquisition unit is used to acquire the state of charge data, health state data, and the time when the data is generated of the power battery; the data transmission unit is used to send the state of charge data, health state data, and the time when the data is generated of the power battery to the battery data storage module. The data analysis module further includes a tram state segmentation unit, a charging speed determination unit, a time determination unit, a state of charge monitoring unit, and a health state monitoring unit; the tram state segmentation unit divides the tram into three states: driving, charging, and stopping based on the state of charge change information of the power battery; the charging speed determination unit calculates the charging speed of the power battery based on the actual charging time and state of charge information of the power battery, trains a support vector regression model based on the charging speed and health state information of the power battery, and determines the average charging speed of the power battery under a specified health state; the time determination unit is used to determine the actual charging time of the power battery and the time when the user charges the tram; the state of charge monitoring unit issues a charging reminder to the user based on the current state of charge of the power battery; the health state monitoring unit is used to provide the user with power battery processing suggestions and reasons, and continuously monitor the health state of the power battery. The output module further includes a display unit, which displays the analysis results of the data analysis module to the user in the form of a chart and displays the power battery processing suggestions generated by the data analysis module.

[0040] Compared with the prior art, the beneficial effects achieved by the present invention are as follows: By combining the actual situation of the user on the demand side, the cruising range and charging speed of the power battery, the state of charge and health state of the power battery are detected, and power battery processing suggestions are generated, thereby improving the utilization rate of the power battery and reducing the impact on the environment caused by the recycling and treatment of the power battery. BRIEF DESCRIPTION OF THE DRAWINGS

[0041] The drawings are used to provide a further understanding of the present invention, and constitute a part of the specification. They are used together with the embodiments of the present invention to explain the present invention, and do not constitute a limitation to the present invention.

[0042] In the drawings:

[0043] Figure 1 is a schematic structural diagram of a power battery monitoring system based on data analysis according to an embodiment of the present invention. DETAILED DESCRIPTION OF THE EMBODIMENTS

[0044] Next, the technical solutions in the embodiments of the present invention will be clearly and completely described in conjunction with 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.

[0045] In an embodiment of the present invention, please refer to Figure 1 , a power battery monitoring system based on data analysis is provided, including: a battery data acquisition module, a battery data storage module, a data analysis module, and an output module; the output end of the battery data acquisition module is connected to the input end of the battery data storage module for obtaining power battery information; the output end of the battery data storage module is connected to the input end of the data analysis module for storing power battery information and sending it to the data analysis module; the output end of the data analysis module is connected to the input end of the output module, and based on the power battery information, the electric vehicle is divided into three states: driving, charging, and stopping, and the electric vehicle state is analyzed to generate power battery processing suggestions for the user; the output module is used to transmit the power battery processing suggestions generated by the data analysis module to the user.

[0046] The battery data acquisition module further includes an authority acquisition unit, a vehicle-mounted information acquisition unit, and a data transmission unit; the authority acquisition unit is used to obtain from the user the access authority to the part of the vehicle-mounted information related to the power battery; the vehicle-mounted information acquisition unit is used to obtain the state of charge data, health state data, and the time when the data is generated of the power battery; the data transmission unit is used to send the state of charge data, health state data, and the time when the data is generated of the power battery to the battery data storage module. The data analysis module further includes an electric vehicle state segmentation unit, a charging speed determination unit, a time determination unit, a state of charge monitoring unit, and a health state monitoring unit; the electric vehicle state segmentation unit divides the electric vehicle into three states: driving, charging, and stopping based on the change information of the state of charge of the power battery; the charging speed determination unit calculates the charging speed of the power battery based on the actual charging time and state of charge information of the power battery, trains a support vector regression model based on the charging speed and health state information of the power battery, and determines the average charging speed of the power battery under a specified health state; the time determination unit is used to determine the actual charging time of the power battery and the time when the user charges the electric vehicle; the state of charge monitoring unit issues a charging reminder to the user based on the current state of charge of the power battery; the health state monitoring unit is used to propose power battery processing suggestions and reasons to the user and continuously monitor the health state of the power battery. The output module further includes a display unit, which displays the analysis results of the data analysis module to the user in the form of a chart and displays the power battery processing suggestions generated by the data analysis module.

[0047] In an embodiment of the present invention, a power battery monitoring method based on data analysis is provided, including:

[0048] S11, accessing the vehicle information of the power battery through user authorization, obtaining the power battery information, and dividing the electric vehicle into three states: driving, charging, and stopping according to the power battery information;

[0049] Obtain the state of charge of the power battery from the vehicle information, divide the state of the electric vehicle according to the state of charge sequence. If the state of charge remains unchanged, it is determined that the electric vehicle is in the stopped state; if the state of charge continues to increase, it is determined that the electric vehicle is in the charging state; if the state of charge continues to decrease, it is determined that the electric vehicle is in the driving state;

[0050] Analyze the time of the stopped state directly after the driving state. If the time that the electric vehicle is in the stopped state is not greater than the time threshold, merge the stopped state with the previous adjacent driving state as part of the driving state. If there are two adjacent driving states, merge the two adjacent driving states into one driving state; the time threshold is set according to the user's situation;

[0051] For the stopped state between the charging state and the driving state, merge the stopped state with the previous adjacent charging state as part of the charging state.

[0052] The time threshold can be set according to the user's waiting time for a red light, waiting time for charging, etc. When the maximum waiting time for the user to wait for a red light is 90s, the time threshold can be set to 90s.

[0053] S12, based on the power battery information and state information of the electric vehicle, determine the time for the user to charge the electric vehicle and send a charging reminder to the user;

[0054] For an electric vehicle in the charging state, if the state of charge of the power battery does not reach 100% when the electric vehicle enters the driving state from the charging state, the time for the user to charge the electric vehicle is the duration of the charging state, and the actual charging time of the power battery is the duration of the charging state; if the state of charge of the power battery reaches 100% when the electric vehicle enters the driving state from the charging state, the time for the user to charge the electric vehicle is the duration of the charging state, and obtain the time it takes for the state of charge of the power battery in the charging state to increase to 100% as the actual charging time of the power battery.

[0055] Obtain the data of the k-th group of adjacent driving states and charging states from historical vehicle information. Let the state of charge of the power battery at the start of the driving state be XSOC1, the state of health of the power battery at the start of the driving state be XSOH1, the state of charge of the power battery at the start of the charging state be CSOC1, and the state of health of the power battery at the start of the charging state be CSOH1; calculate the energy W of the power battery consumed by the user when going to the charging area 2j , W 2j = Y×(XSOC1×XSOH1 - CSOC1×CSOH1), where Y is the capacity of the power battery and is determined according to the model of the power battery;

[0056] Based on the energy of the power battery consumed by the user when going to the charging area generated from the data of all groups of adjacent driving states and charging states, for all W 2j perform unsupervised classification, determine the classification cluster with the most data, and calculate the average value of W 2j in the classification cluster to obtain W2;

[0057] For all W 2j perform unsupervised classification, and methods such as kmeans clustering, hierarchical clustering, DBSCAN clustering, etc. can be used;

[0058] S43, obtain the current state of charge SOC and state of health SOH of the electric vehicle from the vehicle information, and determine the energy W of the current power battery d , W d = Y×XSOC×XSOH, where XSOC is the current SOC of the electric vehicle and XSOH is the current SOH of the electric vehicle; if the energy W of the current power battery d is not greater than K×W2, then send a charging reminder to the user, where K is a scaling factor. K can be set to a number greater than 1 according to the actual situation, such as 1.2, 1.3, etc., to cope with unexpected situations when the user goes to the charging area.

[0059] S13, based on the power battery information of the electric vehicle, the electric vehicle state information, and the time distribution of the user charging the electric vehicle, judge whether the state of health of the power battery meets the user's needs. If it does not meet the user's needs, then put forward a processing suggestion for the user; if it meets the user's needs, then continuously monitor the state of health of the power battery;

[0060] Obtain the data of the a-th group of two consecutive charging states from the historical vehicle information, where the two consecutive charging states mean that there are no other charging states between the two charging states; let the state of charge of the power battery at the end of the previous charging state be XSOC2, the state of health of the power battery at the end of the previous charging state be XSOH2, the state of charge of the power battery at the start of the next charging state be CSOC2, and the state of health of the power battery at the start of the next charging state be CSOH2, and calculate the energy demand W of the driving range for the power battery 3a , W 3a = Y×(XSOC2×XSOH2 - CSOC2×CSOH2);

[0061] Determine the mode of W 3a among all W 3a Perform clustering, determine the clustering cluster with the most data, and calculate the average value of W 3a in the clustering cluster to obtain W3;

[0062] Calculate the first critical value SOH1 of the state of health of the power battery min ,

[0063] Let the time for the user to charge the electric vehicle be T, and calculate the charging time t required to meet the user's driving range demand at the current state of health SOH 0 where P 0 , in the formula is the average charging speed of the power battery at the state of health SOH 0 ; combined with the time for the user to charge the electric vehicle, calculate the probability that the time for the user to charge the electric vehicle meets the energy demand 0 where N is the number of historical data when the electric vehicle is in the charging state, and n0 is the number of charging states with a duration of not less than t ; if P{T≥t 0} is not less than THR, where THR is a preset threshold, then enter step S55, otherwise set the second critical value SOH2 0 of the state of health of the power battery as the current state of health SOH min ; 0 ;

[0064] The average charging speed is determined through the following steps:

[0065] Obtain the state of health soh, the state of charge ksoc at the start of the charging state, the esoc at the end of the charging state i and the actual charging time t of the power battery in the i-th charging state from the historical vehicle information i , calculate the state of health soh i and the state of charge esoc at the end of the charging state i at the start of the charging statei Average charging speed p of the lower power battery i , Train the average charging speed support vector regression model of the power battery according to the average charging speed of the power battery under different health states; set the hyperparameters of support vector regression, including the regularization parameter C, the insensitive parameter epsilon, the kernel function and the parameters of the kernel function; divide the health state and the average charging speed of the power battery into a training set and a test set, use the health state in the training set as the input and the average charging speed of the power battery as the output, train the model under the set hyperparameters, and verify the model with the health state and the average charging speed data of the power battery in the test set; input the power battery health into the support vector regression model to obtain the average charging speed of the power battery under the health state. Commonly used kernel functions can be selected from linear kernel, polynomial kernel, RBF kernel, etc.

[0066] Reduce the current health state SOH 0 in a fixed step size to obtain SOH 1 , SOH 2 , …, SOH max ; and determine the average charging speed P of the power battery after each reduction of the health state 1 , P 2 , …, P max , the charging time t required to meet the user's driving mileage demand 1 , t 2 , …, t max , calculate the probability P{T≥t that the charging time of the user for the electric vehicle meets the energy demand after each reduction of the health state 1}, P{T≥t 2}, …, P{T≥t max}, when the probability that the charging time of the user for the electric vehicle meets the energy demand is less than THR, stop reducing the health state. At this time, the number of times of reducing the health state is max, and the second critical value SOH2 of the power battery health state min is SOH max-1 ; take the minimum value of SOH1 min and SOH2 min to obtain the threshold SOH of the power battery health state min ;

[0067] If the current power battery health state is less than SOH min , remind the user to replace the power battery in time; otherwise, suggest that the user replace the power battery when the power battery health state reaches SOH min ; at the same time, if SOH min =SOH2 min, remind the user to replace the power battery because the charging time is insufficient; if the SOH min = SOH1 min , remind the user to replace the power battery because the battery life of the power battery cannot meet the user's requirements.

[0068] Optionally, THR is set to 90%, that is, it is required that the charging amount meets the driving range requirements of the user in 90% of the cases; the fixed step size can be set to 1%, and the current health state is reduced by 1% each time; when the SOH 0 is 80%, if the probability that the charging time of the user's electric vehicle meets the energy requirement is greater than 90%, then the SOH 0 is reduced to obtain SOH 1 , SOH 1 = 79%. If under the SOH 1 , the probability that the charging time of the user's electric vehicle meets the energy requirement is still greater than 90%, then continue to reduce the SOH 1 to obtain SOH 2 ; when the probability that the charging time of the user's electric vehicle meets the energy requirement under the SOH 2 is not greater than 90%, max is 2, and the second critical value SOH2 min is SOH 1 , and so on.

[0069] It should be noted that in this article, relational terms such as first and second are only used to distinguish one entity or operation from another entity or operation, and do not necessarily require or imply any actual relationship or order between these entities or operations. Moreover, the term "comprising", "including" or any other variant thereof is intended to cover non-exclusive inclusion, so that a process, method, article or device including a series of elements not only includes those elements, but also includes other elements not expressly listed, or further includes elements inherent to such process, method, article or device.

[0070] Finally, it should be noted that the above are only the preferred embodiments of the present invention and are not used to limit the present invention. Although the present invention has been described in detail with reference to the foregoing embodiments, those skilled in the art can still modify the technical solutions described in the foregoing embodiments, or perform equivalent replacements for some of the technical features. Any modifications, equivalent replacements, improvements, etc. made within the spirit and principles of the present invention shall be included within the protection scope of the present invention.

Claims

1. A power battery monitoring method based on data analysis, characterized in that, It includes the following steps: S11, access the vehicle information of the power battery through user authorization, obtain the power battery information, and divide the electric vehicle into three states: driving, charging, and stopping according to the power battery information; S12, based on the power battery information and status information of the electric vehicle, determine the time for the user to charge the electric vehicle, and send a charging reminder to the user; S13, based on the power battery information of the electric vehicle, the electric vehicle status information, and the time distribution of the user charging the electric vehicle, judge whether the health status of the power battery meets the user's needs. If it does not meet the user's needs, propose a treatment suggestion for the user; if it meets the user's needs, continuously monitor the health status of the power battery; In step S13, the judgment of whether the health status of the power battery meets the user's needs further includes the following steps: S51. Obtain the data of the a-th group of consecutive two charging states from historical vehicle-mounted information. The consecutive two charging states mean that there is no other charging state between the two charging states. Let the state of charge of the power battery at the end of the previous charging state be XSOC2, the state of health of the power battery at the end of the previous charging state be XSOH2, the state of charge of the power battery at the start of the next charging state be CSOC2, and the state of health of the power battery at the start of the next charging state be CSOH2. Calculate the energy demand W of the driving range for the power battery. 3a , W 3a = Y×(XSOC2×XSOH2 - CSOC2×CSOH2), where Y is the capacity of the power battery and is determined according to the model of the power battery. S52, determine W 3a Find the mode in, for all W 3a Perform clustering, determine the cluster with the most data, and calculate the average value of W 3a in the cluster to obtain W3; S53, calculate the first critical value SOH1 of the state of health of the power battery min , S54. Let the charging time of the electric vehicle by the user be T, and calculate the current state of health (SOH). 0 The charging time t required to meet the user's driving range requirement. 0 , In the formula, P 0 Is the state of health SOH 0 The average charging speed of the power battery under; Combine the charging time of the electric vehicle by the user to calculate the probability that the charging time of the electric vehicle by the user meets the energy requirement. In the formula, N is the number of historical data when the electric vehicle is in the charging state, and n0 is the number of charging states with a duration not less than t 0 ; If P{T≥t 0}} is not less than THR, where THR is a preset threshold, then go to step S55; otherwise, set the second critical value SOH2 of the power battery health state min As the current state of health SOH 0 ; S55, decrease the current state of health (SOH) 0 in fixed step sizes to obtain SOH 1 , SOH 2 , …, SOH max ; and determine the average charging speed P 1 , P 2 , …, P max of the power battery after each decrease in the state of health, the charging time t 1 , t 2 , …, t max required to meet the user's driving range demand, and calculate the probability P{T≥t 1}, P{T≥t 2}, …, P{T≥t max} that the charging time of the electric vehicle by the user meets the energy demand after each decrease in the state of health. When the probability that the charging time of the electric vehicle by the user meets the energy demand is less than THR, stop decreasing the state of health. At this time, the number of times of decreasing the state of health is max, and the second critical value SOH2 min of the power battery state of health is SOH max-1 ; take the minimum value of SOH1 min and SOH2 min to obtain the threshold SOH min of the power battery state of health; S56, if the current state of health of the power battery is less than SOH min , the user will be reminded to replace the power battery in time; otherwise, it is recommended that the user replace the power battery when the state of health of the power battery reaches SOH min ; at the same time, if SOH min = SOH2 min , the user will be reminded that the reason for replacing the power battery is insufficient charging time; if SOH min = SOH1 min , the user will be reminded that the reason for replacing the power battery is that the driving range of the power battery cannot meet the user's requirements.

2. The battery power monitoring method based on data analysis according to claim 1, wherein In step S11, the division of the electric vehicle into three states of driving, charging, and stopping according to the power battery information further includes the following steps: Obtain the state of charge of the power battery from the vehicle information, divide the electric vehicle status according to the state of charge sequence. If the state of charge remains unchanged, judge that the electric vehicle is in the stopping state; if the state of charge continuously increases, judge that the electric vehicle is in the charging state; if the state of charge continuously decreases, judge that the electric vehicle is in the driving state; Analyze the time of the stopping state directly after the driving state. If the time of the electric vehicle in the stopping state is not greater than the time threshold, merge the stopping state with the previous adjacent driving state as part of the driving state. If there are two adjacent driving states, merge the two adjacent driving states into one driving state; the time threshold is set according to the user's situation; For the stopping state between the charging state and the driving state, merge the stopping state with the previous adjacent charging state as part of the charging state.

3. The method for monitoring a power battery based on data analysis according to claim 2, wherein, In step S12, the determination of the time for the user to charge the electric vehicle based on the power battery information and status information of the electric vehicle further includes the following steps: For the electric vehicle in the charging state, if the state of charge of the power battery does not reach 100% when the electric vehicle enters the driving state from the charging state, the time for the user to charge the electric vehicle is the duration of the charging state, and the actual charging time of the power battery is the duration of the charging state; if the state of charge of the power battery reaches 100% when the electric vehicle enters the driving state from the charging state, the time for the user to charge the electric vehicle is the duration of the charging state, and obtain the time it takes for the state of charge of the power battery in the charging state to increase to 100% as the actual charging time of the power battery.

4. The method for monitoring a power battery based on data analysis according to claim 2, wherein In step S12, the sending of the charging reminder to the user further includes the following steps: S41. Obtain the data of the k-th group of adjacent driving states and charging states from the historical vehicle information. Let the state of charge of the power battery at the start of the driving state be XSOC1, the state of health of the power battery at the start of the driving state be XSOH1, the state of charge of the power battery at the start of the charging state be CSOC1, and the state of health of the power battery at the start of the charging state be CSOH1; calculate the energy W consumed by the power battery for the user to travel to the charging area 2j , W 2j = Y×(XSOC1×XSOH1 - CSOC1×CSOH1); S42, based on the energy of the power battery consumed by the user to go to the charging area generated from the data of all groups' adjacent driving states and charging states, for all W 2j perform unsupervised classification, determine the classification cluster with the most data, and calculate the average value of W 2j in the classification cluster to obtain W2; S43. Obtain the current state of charge (SOC) and state of health (SOH) of the electric vehicle from the vehicle information, and determine the energy W of the current power battery d , W d = Y × XSOC × XSOH, where XSOC is the current SOC of the electric vehicle and XSOH is the current SOH of the electric vehicle; if the energy W of the current power battery d is not greater than K × W2, a charging reminder is sent to the user, where K is a scaling factor 5. The method for monitoring a power battery based on data analysis according to claim 4, wherein In step S54, the average charging speed of the power battery is determined through the following steps: S61, obtain the state of health (SOH) at the i-th charging state, the state of charge (KSOC) at the start of the charging state i , the end-of-charge state of charge (ESOC) i , and the actual charging time (t) of the power battery i from historical vehicle information, and calculate the average charging speed (p) of the power battery at the state of health (SOH) i i i ​​​ S62. Train the average charging speed support vector regression model of the power battery according to the average charging speed of the power battery under different health states; set the hyperparameters of support vector regression, including the regularization parameter C, the insensitive parameter epsilon, the kernel function and the parameters of the kernel function; divide the health state and the average charging speed of the power battery into a training set and a test set, use the health state in the training set as the input and the average charging speed of the power battery as the output, train the model under the set hyperparameters, and verify the model with the health state and the average charging speed data of the power battery in the test set; S63. Input the power battery health into the support vector regression model to obtain the average charging speed of the power battery under the health state.

6. A power battery monitoring system based on data analysis, which uses a power battery monitoring method based on data analysis according to any one of claims 1-5, characterized in that Including: A battery data acquisition module, a battery data storage module, a data analysis module and an output module; The output end of the battery data acquisition module is connected to the input end of the battery data storage module, and is used to obtain power battery information; the output end of the battery data storage module is connected to the input end of the data analysis module, and is used to store the power battery information and send it to the data analysis module; the output end of the data analysis module is connected to the input end of the output module, divides the electric vehicle into three states of driving, charging and stopping based on the power battery information, and analyzes the electric vehicle state to generate a power battery processing suggestion for the user; The output module is used to transmit the power battery processing suggestion generated by the data analysis module to the user.

7. The battery monitoring system based on data analysis according to claim 6, wherein The battery data acquisition module further includes an access right acquisition unit, a vehicle-mounted information acquisition unit and a data transmission unit; the access right acquisition unit is used to obtain from the user the access right to the part of the vehicle-mounted information related to the power battery; the vehicle-mounted information acquisition unit is used to obtain the state of charge data, health state data and the time when the data is generated of the power battery; the data transmission unit is used to send the state of charge data, health state data and the time when the data is generated of the power battery to the battery data storage module.

8. The battery power monitoring system based on data analysis according to claim 7, characterized in that, The data analysis module further includes an electric vehicle state segmentation unit, a charging speed determination unit, a time determination unit, a state of charge monitoring unit and a health state monitoring unit; the electric vehicle state segmentation unit divides the electric vehicle into three states of driving, charging and stopping based on the change information of the state of charge of the power battery; the charging speed determination unit calculates the charging speed of the power battery based on the actual charging time and the state of charge information of the power battery, trains the support vector regression model based on the charging speed and the health state information of the power battery, and determines the average charging speed of the power battery under the specified health state; The time determination unit is used to determine the actual charging time of the power battery and the time when the user charges the electric vehicle. The state of charge monitoring unit issues a charging reminder to the user based on the current state of charge of the power battery; the health state monitoring unit is used to propose a power battery processing suggestion and the reason to the user, and continuously monitor the health state of the power battery.

9. The battery power monitoring system based on data analysis according to claim 8, wherein The output module further includes a display unit, which displays the analysis results of the data analysis module in the form of icons to the user and shows the battery power processing suggestions generated by the data analysis module.

Images