Battery state identification method and device of electric vehicle and storage medium

Abstract

The invention discloses a battery state recognition method and device of an electric automobile and a storage medium, and the method comprises the steps: collecting first battery use information and first battery state information of at least one first automobile to a cloud end, and carrying out the recognition of the battery state of the first automobile according to the first battery use information and the first battery state information; performing artificial intelligence model training on the cloud to obtain a battery state recognition model, and using the battery state recognition model to recognize second battery state information corresponding to the second automobile. According to the invention, the artificial intelligence model can be used for training to obtain the battery state recognition model with dynamic adaptability and prediction capability, and the use information of each first battery and the state information of each first battery can be acquired from each first automobile. The first battery use information and the first battery state information can be dynamically collected in real time, so that the battery state recognition model with good real-time performance is obtained through training. The invention is widely applied to the technical field of automobiles.

Description

Technical Field

[0001] The present invention relates to the field of automobile technology, and in particular to a method, a device and a storage medium for identifying the battery status of an electric vehicle. Background Art

[0002] Electric vehicles use power batteries as the energy required for driving. The battery status such as the health and remaining power of the power battery is related to the endurance, performance and safety of electric vehicles. Therefore, it is necessary to monitor and manage the power battery to ensure that the power battery is in a safe and usable state.

[0003] Traditional battery management technologies mostly rely on preset threshold monitoring. For example, a certain voltage threshold is set. When it is detected that the discharge voltage of the power battery is lower than the voltage threshold, it is determined that the remaining power of the power battery is low. Such battery management technologies often lack dynamic adaptability and predictive capabilities. For example, it is difficult to accurately predict the remaining power of the power battery when the discharge voltage is at a certain value (which is not actually reached).

[0004] Battery status recognition based on artificial intelligence models can establish a sophisticated large model, extract features from the raw data of electric vehicles to identify the battery status, and has good recognition accuracy, dynamic adaptability and prediction capabilities. However, the use of artificial intelligence models requires the support of a large amount of training data. It is difficult for individual users to collect enough data to train a usable artificial intelligence model. Although automobile manufacturers and dealers or professional automobile service companies have the strength to organize experiments to collect data for training artificial intelligence models, these data are generally collected under specific experimental conditions and have poor real-time performance. Summary of the invention

[0005] In view of the technical problems of current battery status recognition technology, such as poor dynamic adaptability, prediction ability and real-time performance, the purpose of the present invention is to provide a battery status recognition method, device and storage medium for an electric vehicle.

[0006] In one aspect, an embodiment of the present invention includes a method for identifying a battery status of an electric vehicle, the method comprising the following steps:

[0007] Collecting first battery usage information and first battery status information of at least one first vehicle to the cloud respectively;

[0008] Performing artificial intelligence model training in the cloud according to each of the first battery usage information and each of the first battery status information to obtain a battery status recognition model;

[0009] The battery status identification model is used to identify second battery status information corresponding to the second vehicle.

[0010] Furthermore, the collecting of the first battery usage information and the first battery status information of at least one first vehicle to the cloud includes:

[0011] Set precise information type and fuzzy information type;

[0012] Traversing at least one of the first cars;

[0013] For any of the first vehicles, the precise value of the information belonging to the precise information type is collected from the first vehicle, and the fuzzy value of the information belonging to the fuzzy information type is collected from the first vehicle. The first battery usage information is composed of the precise value of the information belonging to the precise information type and the fuzzy value and transmitted to the cloud. The first battery status information is collected from the first vehicle and transmitted to the cloud.

[0014] Furthermore, the collecting of the fuzzy value of the information belonging to the fuzzy information type from the first automobile includes:

[0015] Setting multiple value ranges of information belonging to the fuzzy information type;

[0016] sending the value range to the first car respectively;

[0017] Instruct the first vehicle to collect a precise value of the information belonging to the fuzzy information type, and determine the value range to which the precise value of the information belonging to the fuzzy information type belongs as the fuzzy value.

[0018] Furthermore, the collecting the first battery usage information and the first battery status information of at least one first vehicle to the cloud separately also includes:

[0019] For different first automobiles, setting different value ranges;

[0020] Clustering the precise values ​​of the information belonging to the precise information type collected from each of the first automobiles to obtain at least one cluster;

[0021] For any of the clusters, the intersection of the fuzzy values ​​corresponding to the first vehicles corresponding to the cluster is obtained to obtain an intersection range, and the fuzzy values ​​before the intersection are replaced with the intersection range as a component of the first battery usage information.

[0022] Further, the using the battery status recognition model to identify second battery status information corresponding to the second vehicle includes:

[0023] Deploy the battery status recognition model in the cloud;

[0024] Collecting the second battery usage information from the second car and transmitting it to the cloud;

[0025] Inputting the second battery usage information into the battery status identification model for processing in the cloud;

[0026] The second battery status information output by the battery status recognition model is obtained.

[0027] Further, the using the battery status recognition model to identify second battery status information corresponding to the second vehicle includes:

[0028] According to the battery status recognition model, obtaining a vehicle-mounted recognition model;

[0029] Deploy the vehicle-mounted recognition model on the second vehicle terminal;

[0030] collecting second battery usage information from the second vehicle;

[0031] Inputting the second battery usage information into the vehicle-mounted identification model for processing at the second vehicle end;

[0032] The second battery status information output by the vehicle-mounted identification model is obtained.

[0033] Further, acquiring the vehicle identification model according to the battery status identification model includes:

[0034] Distilling the battery status recognition model in the cloud to obtain the vehicle-mounted recognition model;

[0035] The vehicle-mounted recognition model is sent to the second vehicle terminal.

[0036] Furthermore, the acquiring of the vehicle identification model according to the battery status identification model further includes:

[0037] When the second car belongs to any of the first cars, obtaining a fuzziness degree of the fuzzy value collected from the second car;

[0038] According to the fuzziness level, the deployable scale of the vehicle-mounted recognition model is determined; the deployable scale is negatively correlated with the fuzziness level.

[0039] On the other hand, an embodiment of the present invention also includes a computer device, including a memory and a processor, the memory is used to store at least one program, and the processor is used to load at least one program to execute the battery status identification method of the electric vehicle in the embodiment.

[0040] On the other hand, an embodiment of the present invention further includes a computer-readable storage medium, in which a program executable by a processor is stored. When the program executable by the processor is executed by the processor, it is used to execute the battery status identification method of the electric vehicle in the embodiment.

[0041] The beneficial effect of the present invention is that by executing the battery status identification method of the electric vehicle in the embodiment, a battery status identification model with dynamic adaptability and predictive ability can be obtained by using artificial intelligence model training. Since each first battery usage information and each first battery status information of the trained battery status identification model can be collected from each first automobile actually put into operation, the first battery usage information and the first battery status information can be collected in real time and dynamically, thereby training a battery status identification model with good real-time performance. BRIEF DESCRIPTION OF THE DRAWINGS

[0042] Figure 1 is a schematic diagram of a system to which the battery status identification method of an electric vehicle can be applied in an embodiment;

[0043] Figure 2 is a schematic structural diagram of a first automobile in an embodiment;

[0044] Figure 3 A schematic diagram of the steps of a method for identifying the battery status of an electric vehicle in an embodiment;

[0045] Figure 4 It is a schematic diagram of the principle of obtaining the intersection range of each fuzzy value in the embodiment. DETAILED DESCRIPTION

[0046] In this embodiment, the battery status identification method of an electric vehicle can be applied to Figure 1 In the system shown. Figure 1 The system includes a cloud server and multiple first cars, such as the first car 1, the first car 2, etc. These first cars are electric cars, specifically pure electric cars or hybrid electric cars. Taking one of the first cars as an example, its structure is as follows Figure 2As shown, it includes components such as a processor, a power battery, a battery management system (BMS), a driving state sensor module, an environmental information sensor module, an interaction module, a navigation module and a communication module, wherein the processor is a component with control and data processing functions, the power battery provides power for the first automobile, the battery management system manages the charging, discharging and data collection of the power battery, for example, controls the charging and discharging current of the power battery, detects the real-time voltage, current, operating temperature, internal resistance, health and remaining power of the power battery, the driving state sensor module can detect the driving speed, acceleration and other driving state information of the first automobile, the environmental information sensor module can detect the environment where the vehicle is located to obtain environmental information such as temperature, humidity and wind speed, the interaction module can directly identify the identity of the driver and the identity and number of people on the vehicle through face recognition, fingerprint recognition or voice recognition, and can also allow people on the vehicle to input information such as the load weight of the vehicle, the navigation module can set the departure and destination to navigate the vehicle, and can also detect the location of the vehicle and the degree of traffic congestion in real time, and the communication module can use the wireless communication protocol to communicate with the outside, thereby establishing a communication connection between the first automobile and the cloud server.

[0047] In this embodiment, the cloud server can be operated by automobile manufacturers or professional automobile service companies, and the first automobile 1, the first automobile 2, etc. can be users who have signed an agreement with the operator of the cloud server and authorized the operator to receive and use the data collected by themselves. For example, the first automobile 1, the first automobile 2, etc. can be a family car, a bus or a truck, etc., which are not dedicated to data collection, and the owners or drivers of the first automobile 1, the first automobile 2, etc. can be ordinary consumers or logistics companies and their staff. The operator of the cloud server will keep confidential the data uploaded to the cloud server by each car, for example, the original data will not be transmitted to others, and the data will only be used to execute steps S1-S3 and other purposes.

[0048] In this embodiment, each step in the method for identifying the battery status of an electric vehicle may be executed by a cloud server, and the cloud server may call components of the first vehicle when executing the steps. Figure 3 , the battery status identification method of the electric vehicle comprises the following steps:

[0049] S1. Collect the first battery usage information and the first battery status information of at least one first vehicle to the cloud;

[0050] S2. Perform artificial intelligence model training in the cloud based on the battery usage information and battery status information to obtain a battery status recognition model;

[0051] S3. Using the battery status recognition model, identify the second battery status information corresponding to the second vehicle.

[0052] In step S1, refer to Figure 1 , the first car 1 collects the first battery usage information 1 and the first battery status information 1 and uploads them to the cloud server, the first car 2 collects the first battery usage information 2 and the first battery status information 2 and uploads them to the cloud server…

[0053] Specifically, taking the first battery usage information 1 and the first battery status information 1 collected by the first automobile 1 as an example, the first battery status information 1 is the battery status of the power battery of the first automobile 1 at a certain moment (for example, moment t), and the first battery status information 1 can specifically be data such as the battery health (State of Health, SOH) or the remaining power (State Of Charge, SOC) collected by the battery management system of the first automobile 1. In this embodiment, the battery health is used as an example of the first battery status information for explanation; the first battery usage information 1 is information related to the usage process of the power battery of the first automobile 1 at a certain moment or a certain time period (for example, the T1 time period before moment t), including information from the power battery of the first automobile 1 itself (reflecting the properties of the power battery itself), information from the use environment of the power battery of the first automobile 1 (reflecting the influence of the use environment on the power battery), and information from the user of the first automobile 1 (reflecting the influence of the user on the power battery), etc. The first battery usage information 1 can specifically include 1, the discharge voltage, discharge current, operating temperature, internal resistance of the power battery of the first automobile 1 (detected by the battery management system of the first automobile 1), the average driving speed and average driving acceleration of the first automobile 1 (detected by the driving state sensing module of the first automobile 1), the departure location and destination location of the first automobile 1 (detected by the navigation module of the first automobile 1), the average temperature, average humidity, average wind speed of the first automobile 1 on the way from the departure location to the destination (detected by the environmental information sensing module of the first automobile 1), and the driving experience of the driver of the first automobile 1, the type of vehicle allowed by the driver's license, the number of people on the vehicle, and the load (detected by the interactive module of the first automobile 1). That is, the contents of the first battery usage information 1 and the first battery status information 1 can be shown in Table 1.

[0054] Table 1

[0055]

[0056] According to Table 1, the first battery usage information 1 can be represented as a high-dimensional vector, which includes discharge voltage, discharge current, operating temperature, internal resistance, departure location, destination location, driving time period, average temperature, average humidity, average wind speed, average driving speed, average driving acceleration, driver's driving experience, driving license type, number of people on board, load capacity and other components, while the first battery status information 1 can specifically be one-dimensional data such as the battery health of the power battery.

[0057] In step S1, for the first car 2, the first car 3, ... and other first cars, their first battery usage information and first battery status information can also be collected according to the data types shown in Table 1, and each first car sends the collected first battery usage information and first battery status information to the cloud server.

[0058] In step S2, the cloud server compiles the battery usage information and battery status information sent by each first automobile into a training data set, and runs an artificial intelligence model. Specifically, an artificial intelligence model with a transformer architecture can be used. The artificial intelligence model is trained using the training data set, and the trained artificial intelligence model becomes a battery status recognition model.

[0059] Specifically, when executing step S2, the cloud server can perform multiple rounds of training processes. Taking one round of training process as an example, the cloud server inputs the battery usage information sent by one of the first cars (for example, battery usage information 1 sent by the first car 1) into the artificial intelligence model for processing, obtains the actual output data of the artificial intelligence model, and uses the battery status information (for example, battery status information 1) sent by the first car as the expected output data, calculates the error function value between the actual output data and the expected output data, and uses the back propagation algorithm to update the parameters of the artificial intelligence model according to the error function value. In the next round of training, the cloud server will use the battery usage information 2 and battery status information 2 sent by the first car 2 as the input data and expected output data of the artificial intelligence model, respectively. Perform multiple rounds of training processes until the error function converges, or the total number of rounds of the training process reaches the upper limit, and the training of the artificial intelligence model is completed.

[0060] The artificial intelligence model that has undergone the above-mentioned multiple rounds of training has the ability to obtain the characteristics of a car based on its battery usage information and identify the battery status information of the car, and can be used as a battery status recognition model in step S3.

[0061] In step S3, if a second car needs to identify its battery status information, the second car can detect its own power battery through its own battery management system and collect its own second battery usage information. The data format of the second battery usage information can be the same as the data format of the first battery usage information 1 shown in Table 1. The second battery usage information is input into the battery status recognition model for processing. The battery status recognition model will recognize the second battery usage information, and the output value is the second battery status information. The second battery status information can represent the detection or prediction result of the battery status of the power battery of the second car according to the second battery usage information.

[0062] In this embodiment, the second car in step S3 can be any of the first cars in steps S1-S2, that is, the same car not only participates in the training process of the battery status recognition model, but also uses the battery status recognition model to identify its battery status information, or it can be other cars that do not belong to any of the first cars, that is, the second car does not participate in the training process of the battery status recognition model, and only uses the battery status recognition model to identify its battery status information.

[0063] In this embodiment, by executing steps S1-S3, an artificial intelligence model can be used to train a battery status recognition model with dynamic adaptability and predictive ability. Since each first battery usage information and each first battery status information of the trained battery status recognition model can be collected from each first automobile actually put into operation, the first battery usage information and the first battery status information can be collected in real time and dynamically, thereby training a battery status recognition model with good real-time performance.

[0064] In this embodiment, when executing step S1, that is, respectively collecting the first battery usage information and the first battery status information of at least one first vehicle to the cloud, the following steps may be specifically performed:

[0065] S101. Set precise information type and fuzzy information type;

[0066] S102. Traverse at least one first car;

[0067] S103. For any first car, the precise value of information belonging to the precise information type is collected from the first car, the fuzzy value of information belonging to the fuzzy information type is collected from the first car, the precise value and the fuzzy value of the information belonging to the precise information type are combined to form first battery usage information and transmitted to the cloud, and the first battery status information is collected from the first car and transmitted to the cloud.

[0068] In step S101, the precise information type indicates information in the first battery usage information that can or needs to be represented by a precise value, and the fuzzy information type indicates information in the first battery usage information that can or needs to be represented by a fuzzy value. Specifically, the specific contents of the precise information type and the fuzzy information type can be set based on the purpose of privacy protection.

[0069] Taking the first battery usage information 1 shown in Table 1 as an example, it specifically includes data such as discharge voltage, discharge current, operating temperature, internal resistance, departure location, destination location, driving time period, average temperature, average humidity, average wind speed, average driving speed, average driving acceleration, driver's driving experience, driving license type, number of people on board and load capacity, among which data such as discharge voltage, discharge current, operating temperature, internal resistance, average temperature, average humidity, average wind speed, average driving speed, average driving acceleration and load capacity are weakly related to the privacy of the user of the first car 1, that is, the outside world knows the precise values ​​of these data, and cannot infer the privacy information of the user of the first car 1, while the departure location, destination location, driving time period, driver's driving experience, driving license type and number of people on board are strongly related to the privacy of the user of the first car 1, that is, the outside world knows the precise values ​​of these data, and it is easy to infer the privacy information of the user of the first car 1. Therefore, the specific content of the precise information type and the fuzzy information type can be set according to Table 2.

[0070] Table 2

[0071]

[0072] In step S102, the first car 1, the first car 2, ... are traversed, and step S103 is executed for each traversed first car.

[0073] In step S103, taking the first car 1 as an example, the cloud server can request the first car 1 to collect the precise values ​​of the discharge voltage and discharge current when collecting the precise information type, that is, the data directly collected by the first car 1 calling the battery management system and other components. For example, the precise values ​​of the information collected by the first car 1 that belongs to the precise information type include: discharge voltage (450.5V), discharge current (52.56A), operating temperature (27.8℃), internal resistance (30.4mΩ), average temperature (23.6℃), average humidity (60%), average wind speed (11.2m / s), average driving speed (55km / h), average driving acceleration (1.2m·s -2 ) and load capacity (218kg).

[0074] In step S103, taking the first car 1 as an example, the cloud server can request the first car 1 to collect fuzzy values ​​when collecting information such as the departure location, destination location, driver's driving experience, vehicle type permitted by the driver's license, and number of people in the car, which belong to the fuzzy information type.

[0075] For example, the cloud server can set value ranges for the "departure location" such as "distance from positioning point A1>10km", "5km<distance from positioning point A1≤10km", "1km<distance from positioning point A1≤5km" and "distance from positioning point A1≤1km", and send these value ranges to the first car 1. The first car 1 can first detect the precise value of the "departure location" which belongs to the fuzzy information type (such as the specific longitude and latitude), and determine which value range it corresponds to. Assuming it belongs to "5km<distance from positioning point A1≤10km", then for the "departure location" which belongs to the fuzzy information type, the fuzzy value is "5km<distance from positioning point A1≤10km".

[0076] For example, the cloud server can set value ranges such as "driving experience > 10 years", "5 years < driving experience ≤ 10 years", "1 year < driving experience ≤ 5 years" and "driving experience ≤ 1 year" for "driver's driving experience", and send these value ranges to the first car 1. The first car 1 can first detect the exact value of the "driver's driving experience" which belongs to the fuzzy information type (for example, 3 years), and determine which value range it corresponds to, that is, "1 year < driving experience ≤ 5 years". Then, for the "driver's driving experience" which belongs to the fuzzy information type, the fuzzy value is obtained as "1 year < driving experience ≤ 5 years".

[0077] The precise value and the fuzzy value collected in the above manner constitute the first battery usage information 1 of the first car 1 , that is, the form of the first battery usage information 1 can be as shown in Table 3.

[0078] Table 3

[0079]

[0080]

[0081] According to Table 3, for the information of the first battery usage information 1 that belongs to the precise information type, sufficient information can be obtained, thereby providing sufficient support for the cloud server to train a battery status recognition model; for the information of the first battery usage information 1 that belongs to the fuzzy information type, a certain amount of information can be obtained, thereby providing support for the cloud server to train a battery status recognition model, but the departure location of the first car 1 and other information cannot be accurately determined, which is beneficial to protecting the privacy and security of the user of the first car 1.

[0082] In step S103, taking the first car 1 as an example, the cloud server may request the first car 1 to collect the precise value of its first battery status information 1. For example, when the health level is used as the first battery status information, the precise value of the first battery status information 1 may be 85%.

[0083] In step S103, the first car 1 transmits the collected first battery status information 1 and the first battery usage information 1 shown in Table 3 to the cloud.

[0084] In this embodiment, by executing steps S101-S103, the privacy-related part of the first battery usage information collected from the first automobile can be fuzzy processed, thereby providing data support for the cloud server to train a battery status recognition model, while providing privacy protection for the first automobile that provides the first battery usage information.

[0085] In this embodiment, when executing step S1, that is, respectively collecting the first battery usage information and the first battery status information of at least one first vehicle to the cloud, the following steps may be specifically performed:

[0086] S104. For different first cars, set different value ranges;

[0087] S105. Clustering the precise values ​​of the information belonging to the precise information type collected from each first automobile to obtain at least one cluster;

[0088] S106. For any cluster, the intersection of the fuzzy values ​​corresponding to the first vehicles corresponding to the cluster is obtained to obtain an intersection range, and the fuzzy values ​​before the intersection is replaced with the intersection range as a component of the first battery usage information.

[0089] In step S104, different value ranges may be set for the same type of information belonging to the fuzzy information type of different first cars. For example, for the fuzzy information type of "departure location", in the embodiment of the above step S103, the first car 1 is set with value ranges such as "distance from positioning point A1>10km", "5km<distance from positioning point A1≤10km", "1km<distance from positioning point A1≤5km", and "distance from positioning point A1≤1km", while for different first cars, such as the first car 2, a value range different from that of the first car may be set (specifically, it may be different in the location of the positioning point, or different in the upper and lower limits), for example, the first car 2 is set with value ranges such as "distance from positioning point A2>10km", "8km<distance from positioning point A2≤10km", "3km<distance from positioning point A2≤8km", and "distance from positioning point A2≤3km". In this way, even if the precise value (longitude and latitude) of the departure location collected by the first car 2 is the same as the precise value (longitude and latitude) of the departure location of the first car 1, the fuzzy value sent by the first car 2 to the cloud server is different from the fuzzy value sent by the first car 1 to the cloud server.

[0090] The principle of steps S105-S106 is as follows Figure 4 shown.

[0091] Reference Figure 4 , the precise value 1 (including discharge voltage 1, discharge current 1, etc.) collected from the first car 1, the precise value 2 (including discharge voltage 2, discharge current 2, etc.) collected from the first car 2, etc. can be used as vectors, and clustering algorithms such as kmeans are used to cluster these precise values. Assume that Figure 4 As shown, a cluster consisting of precise value 1, precise value 3 and precise value 4 is obtained, which means that precise value 1, precise value 3 and precise value 4 are relatively close and can replace each other within the error range for predicting battery status information.

[0092] Reference Figure 4 In step S106, the precise value 1 in the same cluster is collected from the first car 1, and its corresponding fuzzy value 1 is also collected from the first car 1 (including the departure location 1, the destination location 1...). Similarly, the precise value 3 is collected from the first car 3, and its corresponding fuzzy value 3 is also collected from the first car 3 (including the departure location 3, the destination location 3...). The precise value 4 is collected from the first car 4, and its corresponding fuzzy value 4 is also collected from the first car 4 (including the departure location 4, the destination location 4...). Therefore, the intersection of fuzzy value 1, fuzzy value 3 and fuzzy value 4 can be obtained to obtain the intersection range. The intersection range is as follows: Figure 4The shaded area in the figure is shown.

[0093] For example, departure location 1 in fuzzy value 1, departure location 3 in fuzzy value 3, and departure location 4 in fuzzy value 4 respectively correspond to certain location ranges, and their intersection range can be calculated, that is, intersection range = departure location 1 ∩ departure location 3 ∩ departure location 4.

[0094] Corresponding processing can also be performed on other components in the fuzzy values ​​such as destination position 1, destination position 3, etc.

[0095] In this embodiment, when the intersection range is not an empty set, the intersection range is used to replace each fuzzy value before the intersection is obtained. For example, the intersection range of departure location 1, departure location 3 and departure location 4 is used to replace departure location 1, departure location 3 and departure location 4 as a component of the first battery usage information. Another component of the first battery usage information is any one of the precise value 1, precise value 3 and precise value 4 (because they are clustered into the same cluster, they can replace each other) or the average value.

[0096] In this embodiment, the principle of executing steps S104-S106 is that by finding the intersection range, a relatively more accurate numerical value can be determined based on multiple fuzzy values, so that the cloud server can obtain more accurate data to train the battery status recognition model; from the cloud server side, the intersection range of multiple fuzzy values ​​and the clustering cluster composed of multiple precise values ​​indicate that "there is at least one first car that has detected such precise values ​​and relatively more accurate information belonging to the fuzzy information type", which is sufficient for artificial intelligence model training, and it is impossible to know which specific first car has detected such information, so the privacy and security of the first car can still be protected.

[0097] In this embodiment, when executing step S3, that is, using the battery status recognition model to recognize the second battery status information corresponding to the second vehicle, the following steps may be specifically performed:

[0098] S301A. Deploy the battery status recognition model in the cloud;

[0099] S302A. Collect the second battery usage information from the second car and transmit it to the cloud;

[0100] S303A. The second battery usage information is input into the battery status recognition model for processing in the cloud;

[0101] S304A. Obtain second battery status information output by the battery status recognition model.

[0102] Steps S301A-S304A are the first execution mode of step S3.

[0103] In steps S301A-S304A, the battery status recognition model is deployed on the cloud server, and the second vehicle uploads the collected second battery status information to the cloud server and inputs it into the battery status recognition model for processing, and then the cloud server sends the second battery status information to the second vehicle. By deploying the battery status recognition model on the cloud server, the advantage of the large computing power of the cloud server can be utilized, which is conducive to more accurate recognition of the second battery status information.

[0104] In this embodiment, when executing step S3, that is, using the battery status recognition model to recognize the second battery status information corresponding to the second vehicle, the following steps may be specifically performed:

[0105] S301B obtains the vehicle identification model according to the battery status identification model;

[0106] S302B. Deploy the vehicle recognition model on the second vehicle end;

[0107] S303B collects the second battery usage information from the second car;

[0108] S304B. The second battery usage information is input into the vehicle identification model for processing at the second vehicle end;

[0109] S305B. Obtain the second battery status information output by the vehicle-mounted recognition model.

[0110] Steps S301B-S305B are a second execution mode of step S3.

[0111] When executing step S301B, the battery status recognition model running on the cloud server can be distilled to obtain the vehicle-mounted recognition model, and the vehicle-mounted recognition model can be sent to the second vehicle end for deployment. The battery status recognition model is a large model, and the vehicle-mounted recognition model obtained by rectifying it is a small model, which can be run by the second vehicle end with relatively less computing power resources. If the computing power resources of the second vehicle end are sufficient, in step S301B, the cloud server can also directly send the battery status recognition model as the vehicle-mounted recognition model to the second vehicle end.

[0112] In steps S301B-S305B, the vehicle identification model is deployed on the second vehicle side. After the second vehicle inputs the collected second battery status information into the locally running vehicle identification model for processing, the second battery status information output by the vehicle identification model is directly obtained. By deploying the vehicle identification model on the second vehicle side, the second battery status information can still be identified in the event of a cloud server failure or a communication network failure.

[0113] In this embodiment, if the second car belongs to any first car, that is, the first car that participated in the training of the battery status recognition model in steps S1-S3 subsequently acts as the second car and uses the trained battery status recognition model to identify the second battery status information, then when executing step S301B, the deployable scale of the on-board recognition model sent by the cloud server to the second car can be determined based on the degree of fuzziness of the fuzzy value uploaded to the cloud server when the second car executes steps S101-S103 as the first car.

[0114] Specifically, the degree of ambiguity of the fuzzy value can be measured by the size of the value range. The larger the value range, the greater the degree of ambiguity. For example, if the value range corresponding to a fuzzy value is "5km<distance from positioning point A1≤10km", and the value range corresponding to another fuzzy value is "6km<distance from positioning point A1≤12km", then the latter has a greater degree of ambiguity than the former.

[0115] In this embodiment, if the second car has a smaller degree of fuzziness in the fuzzy value uploaded to the cloud server when executing steps S101-S103 as the first car, then when executing step S301B, the cloud server sets a larger deployable scale for the second car. For example, the cloud server uses the battery status recognition model as the teacher model and outputs more soft labels (such as probability distribution, intermediate layer features, etc.) to distill and obtain the student model as the vehicle-mounted recognition model.

[0116] Since running a larger deployable on-board recognition model is conducive to obtaining more accurate second battery status information when the computing power resources of the second car are sufficient, this can encourage each first car to provide less ambiguous first battery usage information, that is, to provide more refined training data for training a more complete battery status recognition model.

[0117] After obtaining the second battery status information, the second vehicle may generate a health report according to the second battery status information, the health report including information such as the current health status, the estimated remaining life, maintenance recommendations, and determine whether the second battery status information is lower than a warning threshold. If the second battery status information is lower than the warning threshold, the second vehicle may generate an alarm message and provide the user with the battery health report and the alarm message through a mobile application, a vehicle display screen, or a text message, so as to remind the user to take timely measures to ensure the stable performance of the power battery of the second vehicle, and to achieve the following effects:

[0118] 1. Improve safety: Through real-time monitoring and early warning, effectively prevent safety accidents caused by power battery failure and ensure the safe use of power batteries;

[0119] 2. Enhanced reliability: Combining the advantages of cloud and local systems to ensure that battery health information can be provided to users in various situations;

[0120] 4. Extend battery life: Maintenance suggestions based on data analysis help users arrange charging and usage strategies reasonably and slow down battery degradation;

[0121] 5. Optimize user experience: Provide intuitive and easy-to-understand battery health information to enhance user confidence and satisfaction with electric vehicles.

[0122] A computer program for executing the method for identifying the battery status of an electric vehicle in this embodiment can be written and written into a computer device or a storage medium. When the computer program is read out and executed, the method for identifying the battery status of an electric vehicle in this embodiment is executed, thereby achieving the same technical effect as the method for identifying the battery status of an electric vehicle in the embodiment.

[0123] It should be noted that, unless otherwise specified, when a feature is referred to as being "fixed" or "connected" to another feature, it may be directly fixed or connected to the other feature, or it may be indirectly fixed or connected to the other feature. In addition, the descriptions of up, down, left, right, etc. used in the present disclosure are only relative to the relative positional relationship of the components of the present disclosure in the accompanying drawings. The singular forms of "a", "" and "the" used in the present disclosure are also intended to include the plural forms, unless the context clearly indicates other meanings. In addition, unless otherwise defined, all technical and scientific terms used in this embodiment have the same meaning as those generally understood by those skilled in the art. The terms used in the specification of this embodiment are only for describing specific embodiments and are not intended to limit the present invention. The term "and / or" used in this embodiment includes any combination of one or more related listed items.

[0124] It should be understood that, although the term first, second, third etc. may be adopted to describe various elements in the present disclosure, these elements should not be limited to these terms. These terms are only used to distinguish the same type of elements from each other. For example, without departing from the scope of the present disclosure, the first element may also be referred to as the second element, and similarly, the second element may also be referred to as the first element. The use of any and all examples or exemplary language ("for example", "such as" etc.) provided by the present embodiment is only intended to better illustrate embodiments of the present invention, and unless otherwise required, the scope of the present invention will not be limited.

[0125] It should be recognized that embodiments of the present invention can be implemented or carried out by computer hardware, a combination of hardware and software, or by computer instructions stored in a non-transitory computer-readable memory. The methods can be implemented in a computer program using standard programming techniques - including a non-transitory computer-readable storage medium configured with the computer program, wherein the storage medium so configured causes the computer to operate in a specific and predefined manner - in accordance with the methods and drawings described in the specific embodiments. Each program can be implemented in a high-level procedural or object-oriented programming language to communicate with the computer system. However, if desired, the program can be implemented in assembly or machine language. In any case, the language can be a compiled or interpreted language. In addition, for this purpose the program is capable of running on a programmed application-specific integrated circuit.

[0126] In addition, the operations of the processes described in this embodiment can be performed in any suitable order, unless this embodiment otherwise indicates or is otherwise clearly contradicted by the context. The processes described in this embodiment (or variations and / or combinations thereof) can be executed under the control of one or more computer systems configured with executable instructions, and can be implemented as code (e.g., executable instructions, one or more computer programs, or one or more applications) executed jointly on one or more processors, by hardware, or a combination thereof. A computer program includes a plurality of instructions executable by one or more processors.

[0127] Furthermore, the method can be implemented in any type of computing platform operably connected, including but not limited to personal computers, minicomputers, mainframes, workstations, network or distributed computing environments, separate or integrated computer platforms, or communicating with charged particle tools or other imaging devices, etc. Aspects of the present invention can be implemented in machine-readable code stored on a non-transitory storage medium or device, whether removable or integrated into the computing platform, such as a hard disk, optical read and / or write storage medium, RAM, ROM, etc., such that it can be read by a programmable computer and, when the storage medium or device is read by the computer, can be used to configure and operate the computer to perform the processes described herein. In addition, the machine-readable code, or portions thereof, can be transmitted via a wired or wireless network. When such media include instructions or programs that implement the above steps in conjunction with a microprocessor or other data processor, the invention of this embodiment includes these and other different types of non-transitory computer-readable storage media. When programmed according to the methods and techniques of the present invention, the present invention also includes the computer itself.

[0128] The computer program can be applied to input data to perform the functions of the present embodiment, thereby converting the input data to generate output data stored in a non-volatile memory. The output information can also be applied to one or more output devices such as a display. In a preferred embodiment of the present invention, the converted data represents a physical and tangible object, including a specific visual depiction of the physical and tangible object produced on the display.

[0129] The above are only preferred embodiments of the present invention. The present invention is not limited to the above embodiments. As long as the technical effects of the present invention are achieved by the same means, any modifications, equivalent substitutions, improvements, etc. made within the spirit and principles of the present invention shall be included in the scope of protection of the present invention. Within the scope of protection of the present invention, its technical solutions and / or implementation methods may have various modifications and changes.

Claims

1. A method for identifying the battery status of an electric vehicle, characterized in that: The battery status identification method of the electric vehicle comprises: collecting first battery usage information and first battery status information of at least one first vehicle respectively to the cloud; Performing artificial intelligence model training in the cloud according to each of the first battery usage information and each of the first battery status information to obtain a battery status recognition model; The battery status identification model is used to identify second battery status information corresponding to the second vehicle.

2. The method for identifying the battery status of an electric vehicle according to claim 1, characterized in that: The collecting the first battery usage information and the first battery status information of at least one first vehicle to the cloud includes: Set precise information type and fuzzy information type; Traversing at least one of the first cars; For any of the first vehicles, the precise value of the information belonging to the precise information type is collected from the first vehicle, and the fuzzy value of the information belonging to the fuzzy information type is collected from the first vehicle. The first battery usage information is composed of the precise value of the information belonging to the precise information type and the fuzzy value and transmitted to the cloud. The first battery status information is collected from the first vehicle and transmitted to the cloud.

3. The method for identifying the battery status of an electric vehicle according to claim 2, characterized in that: The collecting of the fuzzy value of the information belonging to the fuzzy information type from the first automobile includes: Setting multiple value ranges of information belonging to the fuzzy information type; sending the value range to the first car respectively; Instruct the first vehicle to collect a precise value of the information belonging to the fuzzy information type, and determine the value range to which the precise value of the information belonging to the fuzzy information type belongs as the fuzzy value.

4. The method for identifying the battery status of an electric vehicle according to claim 3, characterized in that: The collecting of the first battery usage information and the first battery status information of at least one first vehicle to the cloud also includes: For different first automobiles, setting different value ranges; Clustering the precise values ​​of the information belonging to the precise information type collected from each of the first automobiles to obtain at least one cluster; For any of the clusters, the intersection of the fuzzy values ​​corresponding to the first vehicles corresponding to the cluster is obtained to obtain an intersection range, and the fuzzy values ​​before the intersection are replaced with the intersection range as a component of the first battery usage information.

5. The method for identifying the battery status of an electric vehicle according to claim 1, characterized in that: The step of using the battery status recognition model to recognize second battery status information corresponding to the second vehicle includes: Deploy the battery status recognition model in the cloud; Collecting the second battery usage information from the second car and transmitting it to the cloud; Inputting the second battery usage information into the battery status identification model for processing in the cloud; The second battery status information output by the battery status recognition model is obtained.

6. The method for identifying the battery status of an electric vehicle according to claim 1, characterized in that: The step of using the battery status recognition model to recognize second battery status information corresponding to the second vehicle includes: According to the battery status recognition model, obtaining a vehicle-mounted recognition model; Deploy the vehicle-mounted recognition model on the second vehicle terminal; collecting second battery usage information from the second vehicle; Inputting the second battery usage information into the vehicle-mounted identification model for processing at the second vehicle end; The second battery status information output by the vehicle-mounted identification model is obtained.

7. The method for identifying the battery status of an electric vehicle according to claim 6, characterized in that: The step of acquiring a vehicle-mounted identification model according to the battery status identification model includes: Distilling the battery status recognition model in the cloud to obtain the vehicle-mounted recognition model; The vehicle-mounted recognition model is sent to the second vehicle terminal.

8. The method for identifying the battery status of an electric vehicle according to claim 7, characterized in that: The step of acquiring a vehicle identification model according to the battery status identification model further includes: When the second car belongs to any of the first cars, obtaining a fuzziness degree of the fuzzy value collected from the second car; According to the fuzziness level, the deployable scale of the vehicle-mounted recognition model is determined; the deployable scale is negatively correlated with the fuzziness level.

9. A computer device, characterized in that: It comprises a memory and a processor, the memory is used to store at least one program, and the processor is used to load at least one program to execute the battery status identification method of an electric vehicle as described in any one of claims 1-8.

10. A computer-readable storage medium storing a program executable by a processor, characterized in that: The program executable by the processor is used to execute the battery status identification method of the electric vehicle as described in any one of claims 1 to 8 when executed by the processor.

Images