Battery replacement system and method for battery replacement cabinet

Abstract

The invention discloses a battery replacement system and method for a battery replacement cabinet, relates to the technical field of battery management, and solves the technical problem that the battery replacement effect is poor due to the fact that the priority is relatively fixed by taking the electric quantity of a battery as a priority judgment standard in the prior art. A battery health score is generated based on battery cabinet data, and an exclusive charging strategy is generated; generating a battery dynamic priority based on the battery cabinet data and the battery health score; the batteries are replaced according to the sequence of the dynamic priorities of the batteries from high to low, the current state and the overall health state of the batteries are monitored in real time, and the battery replacement priority of the batteries is dynamically adjusted through the multi-source data of the batteries, so that the batteries in the battery replacement cabinet can be replaced in an optimal mode, and the effect of the battery replacement method is improved.

Description

Technical Field

[0001] The present application belongs to the field of battery management technology, and specifically relates to a battery replacement system and method for a battery swap cabinet. Background Art

[0002] As urban transportation becomes increasingly electrified, battery swap cabinets, as a new type of energy supply facility, are becoming a crucial component of the electric mobility ecosystem. "Battery swapping at a battery swap cabinet" means that when an electric vehicle, such as an electric bicycle or electric scooter, runs out of power, users can quickly swap batteries in a battery swap cabinet without waiting for a recharge. This allows for instant travel and efficient battery life. This model not only significantly improves travel efficiency but also alleviates the user pain points caused by insufficient charging stations and prolonged charging times.

[0003] In the existing technology, when replacing batteries, the battery power is often used as the priority judgment standard. The priority is relatively fixed, resulting in poor battery replacement effect; therefore, the battery replacement method of the battery swap cabinet still needs further improvement. Summary of the Invention

[0004] The present application aims to solve at least one of the technical problems existing in the prior art; to this end, the present application proposes a battery replacement system and method for a battery swap cabinet, which is used to solve the technical problem that the prior art often uses the battery power as the criterion for priority judgment, and the priority is relatively fixed, resulting in poor battery replacement effect.

[0005] To achieve the above objectives, the first aspect of the present application provides a battery replacement method for a battery swap cabinet, comprising: Get battery cabinet data; Generate a battery health score based on battery cabinet data and generate a dedicated charging strategy; the battery health score is a quantified result of the remaining battery life; Generate dynamic battery priority based on battery cabinet data and battery health score; the dynamic battery priority refers to the quantified result of the priority of several batteries when performing battery replacement; Replace batteries in descending order of battery dynamic priority.

[0006] Through the above steps, this application monitors the current status and overall health of each battery in real time, and dynamically adjusts its battery replacement priority based on multi-source data of the battery, thereby realizing the optimal replacement strategy for the batteries in the battery replacement cabinet, effectively improving the efficiency and effectiveness of the battery replacement method.

[0007] Furthermore, the generation of a battery health score based on battery cabinet data and a dedicated charging strategy includes: The battery cabinet data includes battery ID and battery data; Extract battery data including voltage, battery temperature and battery label; Generate a physical health score for the battery ID based on voltage and battery temperature; the physical health score is a quantified result of the current physical state of the battery; When the physical health score is less than the physical score threshold, the battery tag is set as a fault tag and a battery ID fault warning signal is generated; the physical score threshold is set based on experience; When the physical health score is greater than or equal to the physical score threshold, the battery tag is set as an available tag, and the capacity decay rate, charging time distribution, and temperature rise curve for N historical time points corresponding to the battery ID are obtained. Where N is a positive integer, and the specific value is set based on experience. The capacity decay rate, charging time distribution and temperature rise curve at the same historical time point are integrated through parameter weight coefficients to obtain health analysis data; Integrate the health analysis data of N historical time points into a health analysis sequence of the battery ID in the order of the historical time points; Inputting the battery ID and its corresponding health analysis sequence into the health prediction model to obtain the battery health score corresponding to the battery ID; the health prediction model is constructed using an LSTM model to evaluate the health score of the battery ID; Generate a dedicated charging strategy based on the battery health score and battery ID.

[0008] This application monitors the status of the battery in real time. During the battery replacement process, it does not consider abnormal batteries and screens out several batteries that can be replaced in real time. It also obtains the health score of each battery for the batteries that can be replaced through a pre-trained health prediction model, sets a reasonable charging strategy for each battery, and improves the battery life.

[0009] Furthermore, generating a physical health score of the battery ID based on the voltage and battery temperature includes: The constant current charging end voltage corresponding to the extracted voltage and discharge cut-off voltage ; By formula Calculate the voltage stability factor VWX; where, It is expressed as nominal voltage, and the specific value is set according to experience; Extract the real-time battery temperature corresponding to the battery temperature and the average value of the historical battery temperature within M hours; where M is a positive integer; Calculate the absolute value of the difference between the real-time battery temperature and the historical average battery temperature ; By formula Calculate the temperature anomaly coefficient WYX; where, Expressed as the temperature anomaly coefficient, ∈(0, 1), the specific value is set according to experience; and are temperature thresholds, and Greater than , the specific value is set according to experience; The physical health score of the battery ID is calculated by weighted summation of the voltage stability coefficient and the temperature anomaly coefficient; The calculation formula of the physical health score satisfies: ;in, and Expressed as weight coefficient, and ∈(0, 1), the specific value is set according to experience.

[0010] Furthermore, the parameter weight coefficient is generated by the following method, including: Obtain capacity decay rate, charging time distribution and temperature rise curve at historical time points; The linear correlation between capacity decay rate, charging time distribution and temperature rise curve and battery health score is quantified by Pearson correlation coefficient method. ; By formula Calculate the parameter weight coefficients corresponding to the capacity decay rate, charging time distribution, and temperature rise curve; where i and j represent the numbers corresponding to the capacity decay rate, charging time distribution, and temperature rise curve, and J represents the total number of numbers; p represents the correlation reinforcement factor, and P>0; in the calculation formula of the parameter weight coefficient, the higher the correlation between the capacity decay rate, charging time distribution, and temperature rise curve at a historical time point and the battery health score, the higher the corresponding parameter weight coefficient will be.

[0011] Furthermore, the health prediction model is constructed using an LSTM model, including: Get several historical battery IDs, historical health analysis sequences, and historical health scores; Divide several historical battery IDs, historical health analysis sequences, and historical health scores into training data, verification data, and test data; perform data preprocessing on the training data, verification data, and test data to obtain a training set, a verification set, and a test set; Select the LSTM model as the base model; Train the basic model using the training set, and adjust the learning rate and hyperparameters on the validation set to obtain the pre-trained model; By verifying the pre-trained model on the test set, we finally obtained a health prediction model whose input is the battery ID and health analysis sequence, and whose output is the battery health score corresponding to the battery ID.

[0012] Furthermore, the exclusive charging strategy for generating a battery ID based on the battery health score includes: Extract several battery tags as available tags, including battery IDs and their corresponding battery health scores; When the battery health score is greater than a health threshold of one, the battery ID corresponding to the battery health score is set to fast charge mode; When the battery health score is less than or equal to health threshold 1 and the battery health score is greater than or equal to health threshold 2, the battery ID corresponding to the battery health score is set to slow charge mode; When the battery health score is less than health threshold 2, a battery life shortage warning signal is generated, and the battery tag of the battery ID corresponding to the battery health score is set as a fault tag; wherein, health threshold 2 is greater than health threshold 1, and the specific value is set based on experience.

[0013] Furthermore, generating a dynamic battery priority based on battery cabinet data and battery health score includes: The battery cabinet data includes the battery ID and its corresponding battery data; Extracting several battery IDs and battery data of battery tags as available tags; Extracting the battery charging progress and battery standby time from the battery data; When the battery charging progress is less than the available threshold, the battery replacement tag corresponding to the battery ID of the battery charging progress is set to a non-replaceable tag; When the battery charging progress is greater than or equal to the available threshold, the battery swap tag corresponding to the battery ID of the battery charging progress is set as a swap tag; the available threshold is the limit at which the battery can be swapped, and the specific value is set based on experience; When there are replaceable tags among the battery replacement tags of several battery IDs, the battery dynamic priority of several battery IDs is generated based on the battery charging progress, battery standby time and battery health score; When there is no replaceable tag among the battery replacement tags of several battery IDs, a battery cabinet non-replaceable warning signal is generated.

[0014] Based on the preliminary screening of multiple batteries in the battery swap cabinet, this application further implements a second screening by real-time monitoring of the charging progress of these batteries, so as to accurately identify the target battery ID that can ultimately be used for battery swapping; by comprehensively analyzing and screening the battery status from two dimensions, it can maximize the selection of batteries that meet effective battery swapping conditions, significantly improving the overall efficiency and reliability of the battery swapping process.

[0015] Furthermore, the battery dynamic priority of several battery IDs is generated based on the battery charging progress, battery standby time and battery health score, including: The battery dynamic priority satisfies the following calculation formula: ;in, 、 and It is represented by the weight coefficient corresponding to the battery charging progress, battery standby time and battery health score of the k-th battery ID. 、 and ∈(0, 1); t represents the attenuation coefficient, t∈(0, 1); DCJ represents the battery charging progress, DDS represents the battery standby time, and DJP represents the battery health score; The weight coefficient is generated by the following methods, including: Get the battery replacement time, battery charging progress corresponding to the battery ID, battery standby time, and battery health score; Integrate the battery replacement time, battery charging progress corresponding to the battery ID, battery standby time, and battery health score into weighted analysis data; The weight analysis data is input into a weight generation model to obtain several weight coefficients corresponding to the battery ID; the weight generation model is constructed through an artificial intelligence model to evaluate the weight coefficients corresponding to the battery charging progress, battery standby time and battery health score of the battery ID.

[0016] Furthermore, the weight generation model is constructed through an artificial intelligence model, including: Obtaining several historical weight analysis data and their corresponding several historical weight coefficients; Dividing a number of historical weight analysis data and their corresponding historical weight coefficients into training data, verification data, and test data; performing data preprocessing on the training data, verification data, and test data to obtain a training set, a verification set, and a test set; Select an artificial intelligence model as the base model; Train the basic model using the training set, and adjust the learning rate and hyperparameters on the validation set to obtain the pre-trained model; By verifying the pre-trained model on the test set, we finally obtain a weight generation model whose input is weight analysis data and output is a number of weight coefficients.

[0017] When calculating the dynamic battery priority of several batteries, this application takes into account the three parameters of the battery charging progress, battery standby time and battery health score, and dynamically generates several weight coefficients through a pre-trained weight generation model, so that when performing several priority assessments, the optimal weight coefficient can be generated according to the battery replacement time point and the performance parameters of several batteries, thereby improving the accuracy of the battery dynamic priority and improving the effect of battery replacement.

[0018] Another aspect of the present invention provides a battery replacement system for a battery swap cabinet, comprising: a data acquisition module, a data analysis module, and an early warning module; the data acquisition module is connected to the data analysis module; the data analysis module is connected to the early warning module; The data acquisition module is used to acquire battery cabinet data through data acquisition equipment; The data analysis module generates a battery health score and a dedicated charging strategy based on the battery cabinet data; generates a battery dynamic priority based on the battery cabinet data and the battery health score; and replaces batteries in descending order of the battery dynamic priority. The early warning module makes corresponding prompts according to the early warning signal.

[0019] Compared with the prior art, the present invention has the following advantages: 1. This application generates a battery health score and an exclusive charging strategy based on battery cabinet data; generates a battery dynamic priority based on battery cabinet data and the battery health score; replaces batteries in descending order of the battery dynamic priority, monitors the current status and overall health status of several batteries in real time, and dynamically adjusts the battery replacement priority through multi-source data of the battery, so that the batteries in the battery replacement cabinet can be replaced in the best way, thereby improving the effect of the battery replacement method.

[0020] 2. When evaluating the battery health score, this application takes into account that the several parameters involved in the evaluation have different effects on the battery health score at different time points. Therefore, the corresponding several parameter weight coefficients are dynamically adjusted for each historical time point, so that when evaluating the battery health score, more accurate data can be used for evaluation, thereby improving the accuracy of the battery health score and providing accurate data support for subsequent exclusive charging strategies and battery replacement methods.

[0021] 3. After the preliminary screening of several batteries in the battery cabinet, this application conducts a second screening by real-time monitoring of the charging progress of several batteries that have been preliminarily screened in the battery swap cabinet to obtain several battery IDs that can eventually be swapped. By analyzing and screening from two directions of the batteries, it is possible to select batteries that can be effectively swapped to the greatest extent, thereby improving the efficiency of battery swapping. BRIEF DESCRIPTION OF THE DRAWINGS

[0022] In order to more clearly illustrate the embodiments of the present application or the technical solutions in the prior art, the following briefly introduces the drawings required for use in the embodiments or the description of the prior art. Obviously, the drawings described below are only some embodiments of the present application. For ordinary technicians in this field, other drawings can be obtained based on these drawings without any creative work.

[0023] Figure 1 This is a flow chart of a battery replacement method for a battery swap cabinet in the present application; Figure 2 This is a schematic diagram of the principle of a battery replacement system for a battery swap cabinet in this application. DETAILED DESCRIPTION

[0024] The following will clearly and completely describe the technical solutions of this application in conjunction with the embodiments. Obviously, the embodiments described are only a part of the embodiments of this application, not all of them. Based on the embodiments in this application, all other embodiments obtained by ordinary technicians in this field without making creative efforts are within the scope of protection of this application.

[0025] See also Figure 1 The first embodiment of the present application provides a battery replacement method for a battery swap cabinet, comprising: Get battery cabinet data; Generate a battery health score and a dedicated charging strategy based on battery cabinet data; the battery health score quantifies the remaining battery life. Generate dynamic battery priority based on battery cabinet data and battery health scores; dynamic battery priority refers to the quantitative result of the priority of several batteries during battery replacement; Replace batteries in descending order of battery dynamic priority.

[0026] In this embodiment, the battery health score and dedicated charging strategy are generated based on the battery cabinet data, including: Battery cabinet data includes battery ID and battery data; Extracting the voltage, battery temperature, and battery tag of the battery data; the battery tag is a tag for evaluating the battery status; Generates a physical health score for the battery ID based on voltage and battery temperature. The physical health score quantifies the current physical state of the battery, for example, whether the current battery temperature and voltage are within normal ranges. When the physical health score is less than the physical score threshold, the battery tag is set as a fault tag and a battery ID fault warning signal is generated. The physical score threshold is set based on experience and is set to 0.6 in this embodiment. When the physical health score is greater than or equal to the physical score threshold, the battery tag is set as an available tag, and the capacity decay rate, charging time distribution, and temperature rise curve for N historical time points corresponding to the battery ID are obtained. Where N is a positive integer, and the specific value is set based on experience. In this embodiment, N is set to 30. The capacity decay rate, charging time distribution, and temperature rise curve at the same historical time point are combined using parameter weight coefficients to generate health analysis data. The parameter weight coefficients at each historical time point may vary, so the resulting health analysis data may also be different. Integrate the health analysis data of N historical time points into a health analysis sequence of the battery ID in the order of the historical time points; The battery ID and its corresponding health analysis sequence are input into the health prediction model to obtain the battery health score corresponding to the battery ID. The health prediction model is constructed using the LSTM model to evaluate the health score of the battery ID. Generate a dedicated charging strategy based on the battery health score and battery ID.

[0027] This embodiment automatically excludes abnormal batteries during the battery replacement process through real-time monitoring of the battery status and dynamically selects available batteries that meet the battery replacement conditions; then, a pre-trained health status prediction model is used to perform health scores on these batteries, and a reasonable personalized charging strategy is formulated for each battery based on the scoring results, thereby effectively extending the battery life while ensuring the battery replacement efficiency.

[0028] In this embodiment, the physical health score of the battery ID is generated based on the voltage and battery temperature, including: The constant current charging end voltage corresponding to the extracted voltage and discharge cut-off voltage ; By formula Calculate the voltage stability factor VWX; where, It is expressed as nominal voltage. The specific value is set according to experience. In this embodiment, Set to 3.7 volts; Extract the real-time battery temperature corresponding to the battery temperature and the average value of the historical battery temperature within M hours; where M is a positive integer, and in this embodiment, M is set to 24; Calculate the absolute value of the difference between the real-time battery temperature and the historical average battery temperature ; By formula Calculate the temperature anomaly coefficient WYX; where, Expressed as the temperature anomaly coefficient, ∈(0, 1); The specific value is set according to experience. In this embodiment, Set to ; and are temperature thresholds, and Greater than ; The specific value is set according to experience. In this embodiment, and Set to 15°C and 5°C respectively; The physical health score of the battery ID is calculated by weighted summation of the voltage stability coefficient and the temperature anomaly coefficient; The calculation formula for physical health score satisfies: ;in, and Expressed as weight coefficient, and ∈(0, 1), the specific value is set according to experience. In this embodiment, and Set to 0.6 and 0.4 respectively.

[0029] The parameter weight coefficients in this embodiment are generated by the following methods, including: Obtain capacity decay rate, charging time distribution and temperature rise curve at historical time points; The linear correlation between capacity decay rate, charging time distribution and temperature rise curve and battery health score is quantified by Pearson correlation coefficient method. ; By formula Calculate the parameter weight coefficients corresponding to the capacity decay rate, charging time distribution, and temperature rise curve; wherein, i and j represent the numbers corresponding to the capacity decay rate, charging time distribution, and temperature rise curve, J represents the total number, and in this embodiment, J is 3; p represents the association reinforcement factor, P>0; the specific value is set based on experience, and in this embodiment, p is set to 1.5; in the calculation formula of the parameter weight coefficient, the higher the correlation between the capacity decay rate, charging time distribution, and temperature rise curve in the historical time point and the battery health score, the higher the corresponding parameter weight coefficient will be.

[0030] When evaluating the battery health score, this embodiment takes into account the different effects of the multiple parameters involved in the evaluation on the health score at different time periods. Therefore, the weight coefficients of the corresponding parameters are dynamically adjusted for each historical time point. In this way, when calculating the battery health score, the accuracy of the score can be significantly improved based on weighted data that is closer to the actual status, providing more accurate data support for the subsequent formulation of personalized charging strategies and the implementation of battery replacement plans.

[0031] The health prediction model in this embodiment is constructed using an LSTM model, including: Get several historical battery IDs, historical health analysis sequences, and historical health scores; Divide several historical battery IDs, historical health analysis sequences, and historical health scores into training data, validation data, and test data; perform data preprocessing on the training data, validation data, and test data to obtain training sets, validation sets, and test sets; the ratio of the training set, test set, and validation set is 7:2:1; Select the LSTM model as the base model; Train the basic model using the training set, and adjust the learning rate and hyperparameters on the validation set to obtain the pre-trained model; By verifying the pre-trained model on the test set, we finally obtained a health prediction model whose input is the battery ID and health analysis sequence, and whose output is the battery health score corresponding to the battery ID.

[0032] The dedicated charging strategy for generating a battery ID based on the battery health score in this embodiment includes: Extract several battery tags as available tags, including battery IDs and their corresponding battery health scores; When the battery health score is greater than the health threshold 1, the battery ID corresponding to the battery health score is set to fast charge mode; When the battery health score is less than or equal to health threshold 1 and the battery health score is greater than or equal to health threshold 2, the battery ID corresponding to the battery health score is set to slow charge mode; When the battery health score is less than health threshold 2, a battery life shortage warning signal is generated, and the battery tag of the battery ID corresponding to the battery health score is set as a fault tag; among them, health threshold 2 is greater than health threshold 1. The specific values ​​are set based on experience. In this embodiment, health threshold 2 and health threshold 1 are set to 0.8 and 0.3 respectively. In the set fast charging mode and slow charging mode, the corresponding charging parameters will be different.

[0033] In this embodiment, the dynamic battery priority is generated based on the battery cabinet data and the battery health score, including: Battery cabinet data includes battery ID and its corresponding battery data; Extracting several battery IDs and battery data of battery tags as available tags; Extract battery charging progress and battery standby time from battery data; When the battery charging progress is less than the available threshold, the battery replacement tag corresponding to the battery ID of the battery charging progress is set to a non-replaceable tag; When the battery charging progress is greater than or equal to the available threshold, the battery swap tag corresponding to the battery ID of the battery charging progress is set to a swap tag; the available threshold is the limit at which the battery can be swapped, and the specific value is set based on experience. In this embodiment, the available threshold is set to 50%; that is, when the battery charging progress of the battery ID is less than 50%, the battery cannot be swapped; When there are replaceable tags among the battery replacement tags of several battery IDs, the battery dynamic priority of several battery IDs is generated based on the battery charging progress, battery standby time and battery health score; When there is no replaceable tag among the battery replacement tags of several battery IDs, a battery cabinet non-replaceable warning signal is generated.

[0034] In this embodiment, the dynamic battery priority of several battery IDs is generated based on the battery charging progress, battery standby time, and battery health score, including: The dynamic battery priority satisfies the following calculation formula: ;in, 、 and It is represented by the weight coefficient corresponding to the battery charging progress, battery standby time and battery health score of the k-th battery ID. 、 and ∈(0, 1); t represents the attenuation coefficient, t∈(0, 1); the specific value is set based on experience. In this embodiment, t is set to 0.05 / hour; DCJ represents the battery charging progress, DDS represents the battery standby time, and DJP represents the battery health score; The weight coefficients are generated by the following methods, including: Obtain the battery swap time, battery charging progress corresponding to the battery ID, battery standby time, and battery health score; the battery swap time represents the battery swap scenario, and the battery swap focus varies in different scenarios. For example, during peak hours, the focus is on the user's battery swap experience, and it is necessary to ensure the supply of high-capacity batteries; during off-peak hours, the focus is on battery aging prevention and control, and it is necessary to prioritize cleaning of idle batteries. Integrate the battery replacement time, battery charging progress corresponding to the battery ID, battery standby time, and battery health score into weighted analysis data; The weight analysis data is input into the weight generation model to obtain several weight coefficients corresponding to the battery ID; the weight generation model is constructed through an artificial intelligence model to evaluate the weight coefficients corresponding to the battery charging progress, battery standby time and battery health score of the battery ID.

[0035] The weight generation model in this embodiment is constructed through an artificial intelligence model, including: Obtaining several historical weight analysis data and their corresponding several historical weight coefficients; Divide a number of historical weight analysis data and their corresponding historical weight coefficients into training data, validation data, and test data; perform data preprocessing on the training data, validation data, and test data to obtain a training set, a validation set, and a test set; the ratio of the training set, the test set, and the validation set is 7:2:1; Select an artificial intelligence model as the basic model; artificial intelligence models include BP models, etc. Train the basic model using the training set, and adjust the learning rate and hyperparameters on the validation set to obtain the pre-trained model; By verifying the pre-trained model on the test set, we finally obtain a weight generation model whose input is weight analysis data and output is a number of weight coefficients.

[0036] When calculating the dynamic priority of each battery, this embodiment comprehensively considers the three key parameters of the battery's charging progress, standby time, and health score, and dynamically generates the corresponding weight coefficient through a pre-trained weight generation model; it can adaptively adjust the weight of each parameter according to different battery replacement time points and changes in battery performance parameters, thereby achieving more accurate sorting in the priority evaluation process, improving the accuracy of battery dynamic priority evaluation, further optimizing the battery replacement strategy, and improving the efficiency and intelligence level of the battery replacement effect.

[0037] See also Figure 2 Another embodiment of the present application provides a battery replacement system for a battery swap cabinet, comprising: a data acquisition module, a data analysis module, and an early warning module; the data acquisition module is electrically and / or communicatively connected to the data analysis module; the data analysis module is electrically and / or communicatively connected to the early warning module; Data acquisition module: obtains battery cabinet data through data acquisition equipment; data acquisition equipment includes several sensors, etc. Data analysis module: Generates battery health scores and exclusive charging strategies based on battery cabinet data; generates dynamic battery priorities based on battery cabinet data and battery health scores; and replaces batteries in descending order of dynamic battery priority. Early warning module: Prompts accordingly based on early warning signals, including battery cabinet unreplaceable warning signal, battery ID failure warning signal, and battery life shortage warning signal.

[0038] Some of the data in the above formula are calculated by removing the dimensions and taking their numerical values. The formula is a formula that is closest to the actual situation obtained by software simulation of a large amount of collected data; the preset parameters and preset thresholds in the formula are set by technical personnel in this field according to actual conditions or obtained through simulation of a large amount of data.

[0039] The working principle of this application is: by obtaining battery cabinet data; generating a battery health score and an exclusive charging strategy based on the battery cabinet data; generating a battery dynamic priority based on the battery cabinet data and the battery health score; replacing batteries in order of high to low battery dynamic priority, monitoring the current status and overall health status of several batteries in real time, and dynamically adjusting the battery replacement priority through multi-source data of the battery, so that the batteries in the battery replacement cabinet can be replaced in the best way, improving the effect of the battery replacement method, and avoiding the problem that the existing technology often uses the battery power as the priority judgment standard, the priority is relatively fixed, and the battery replacement effect is poor.

[0040] The above embodiments are only used to illustrate the technical method of the present application and are not intended to limit it. Although the present application has been described in detail with reference to the preferred embodiments, those skilled in the art should understand that the technical method of the present application can be modified or replaced by equivalents without departing from the spirit and scope of the technical method of the present application.

Claims

1. A battery replacement method for a battery swap cabinet, characterized in that: include: Get battery cabinet data; Generate battery health scores and exclusive charging strategies based on battery cabinet data; The battery health score is a quantified result of the remaining battery life. Generate dynamic battery priority based on battery cabinet data and battery health score; the dynamic battery priority refers to the quantified result of the priority of several batteries when performing battery replacement; Replace batteries in descending order of their dynamic priority.

2. A battery replacement method for a battery swap cabinet according to claim 1, characterized in that: Generating a battery health score and a dedicated charging strategy based on battery cabinet data includes: The battery cabinet data includes battery ID and battery data; Extract battery data including voltage, battery temperature and battery label; Generates a physical health score for the battery ID based on voltage and battery temperature; When the physical health score is less than the physical score threshold, the battery tag is set as a fault tag and a battery ID fault warning signal is generated; When the physical health score is greater than or equal to the physical score threshold, the battery tag is set to an available tag, and the capacity decay rate, charging time distribution, and temperature rise curve for N historical time points corresponding to the battery ID are obtained; where N is a positive integer; The capacity decay rate, charging time distribution and temperature rise curve at the same historical time point are integrated through parameter weight coefficients to obtain health analysis data; Integrate the health analysis data of N historical time points into a health analysis sequence of the battery ID in the order of the historical time points; Inputting the battery ID and its corresponding health analysis sequence into the health prediction model to obtain the battery health score corresponding to the battery ID; the health prediction model is constructed using an LSTM model to evaluate the health score of the battery ID; Generate a dedicated charging strategy based on the battery health score and battery ID.

3. A battery replacement method for a battery swap cabinet according to claim 2, characterized in that: The physical health score of the battery ID is generated based on the voltage and battery temperature, including: The constant current charging end voltage corresponding to the extracted voltage and discharge cut-off voltage ; By formula Calculate the voltage stability factor VWX; where, Expressed as nominal voltage; Extract the real-time battery temperature corresponding to the battery temperature and the average value of the historical battery temperature within M hours; where M is a positive integer; Calculate the absolute value of the difference between the real-time battery temperature and the historical average battery temperature ; By formula Calculate the temperature anomaly coefficient WYX; where, Expressed as the temperature anomaly coefficient, ∈(0,1); and are temperature thresholds, and Greater than ; The physical health score of the battery ID is calculated by weighted summation of the voltage stability coefficient and the temperature anomaly coefficient.

4. A battery replacement method for a battery swap cabinet according to claim 2, characterized in that: The parameter weight coefficient is generated by the following methods, including: Obtain capacity decay rate, charging time distribution and temperature rise curve at historical time points; The linear correlation between capacity decay rate, charging time distribution and temperature rise curve and battery health score is quantified by Pearson correlation coefficient method. ; By formula Calculate the parameter weight coefficients corresponding to the capacity decay rate, charging time distribution, and temperature rise curve; where i and j represent the numbers corresponding to the capacity decay rate, charging time distribution, and temperature rise curve, and J represents the total number; and p represents the association enhancement factor, P>

0.

5. A battery replacement method for a battery swap cabinet according to claim 2, characterized in that: The health prediction model is constructed using an LSTM model, including: Get several historical battery IDs, historical health analysis sequences, and historical health scores; Divide several historical battery IDs, historical health analysis sequences, and historical health scores into training data, verification data, and test data; perform data preprocessing on the training data, verification data, and test data to obtain a training set, a verification set, and a test set; Select the LSTM model as the base model; Train the basic model using the training set, and adjust the learning rate and hyperparameters on the validation set to obtain the pre-trained model; By verifying the pre-trained model on the test set, we finally obtained a health prediction model whose input is the battery ID and health analysis sequence, and whose output is the battery health score corresponding to the battery ID.

6. A battery replacement method for a battery swap cabinet according to claim 2, characterized in that: The exclusive charging strategy for generating a battery ID based on the battery health score includes: Extract several battery tags as available tags, including battery IDs and their corresponding battery health scores; When the battery health score is greater than a health threshold of one, the battery ID corresponding to the battery health score is set to fast charge mode; When the battery health score is less than or equal to health threshold 1 and the battery health score is greater than or equal to health threshold 2, the battery ID corresponding to the battery health score is set to slow charge mode; When the battery health score is less than health threshold 2, a battery life shortage warning signal is generated, and the battery tag of the battery ID corresponding to the battery health score is set as a fault tag; wherein health threshold 2 is greater than health threshold 1.

7. A battery replacement method for a battery swap cabinet according to claim 1, characterized in that: Generating dynamic battery priorities based on battery cabinet data and battery health scores includes: The battery cabinet data includes the battery ID and its corresponding battery data; Extracting several battery IDs and battery data of battery tags as available tags; Extracting the battery charging progress and battery standby time from the battery data; When the battery charging progress is less than the available threshold, the battery replacement tag corresponding to the battery ID of the battery charging progress is set to a non-replaceable tag; When the battery charging progress is greater than or equal to the available threshold, the battery replacement tag corresponding to the battery ID of the battery charging progress is set to a replaceable tag; When there are replaceable tags among the battery replacement tags of several battery IDs, the battery dynamic priority of several battery IDs is generated based on the battery charging progress, battery standby time and battery health score; When there is no replaceable tag among the battery replacement tags of several battery IDs, a battery cabinet non-replaceable warning signal is generated.

8. A battery replacement method for a battery swap cabinet according to claim 7, characterized in that: The battery dynamic priority of several battery IDs is generated based on the battery charging progress, battery standby time and battery health score, including: The battery dynamic priority satisfies the following calculation formula: ;in, 、 and It is represented by the weight coefficient corresponding to the battery charging progress, battery standby time and battery health score of the k-th battery ID. 、 and ∈(0, 1); t represents the attenuation coefficient, t∈(0, 1); DCJ represents the battery charging progress, DDS represents the battery standby time, and DJP represents the battery health score; The weight coefficient is generated by the following methods, including: Get the battery replacement time, battery charging progress corresponding to the battery ID, battery standby time, and battery health score; Integrate the battery replacement time, battery charging progress corresponding to the battery ID, battery standby time, and battery health score into weighted analysis data; The weight analysis data is input into a weight generation model to obtain several weight coefficients corresponding to the battery ID; the weight generation model is constructed through an artificial intelligence model to evaluate the weight coefficients corresponding to the battery charging progress, battery standby time and battery health score of the battery ID.

9. A battery replacement method for a battery swap cabinet according to claim 8, characterized in that: The weight generation model is constructed through an artificial intelligence model, including: Obtaining several historical weight analysis data and their corresponding several historical weight coefficients; Dividing a number of historical weight analysis data and their corresponding historical weight coefficients into training data, verification data, and test data; performing data preprocessing on the training data, verification data, and test data to obtain a training set, a verification set, and a test set; Select an artificial intelligence model as the base model; Train the basic model using the training set, and adjust the learning rate and hyperparameters on the validation set to obtain the pre-trained model; By verifying the pre-trained model on the test set, we finally obtain a weight generation model whose input is weight analysis data and output is a number of weight coefficients.

10. A battery replacement system for a battery swap cabinet, applied to a battery replacement method for a battery swap cabinet according to any one of claims 1 to 9, characterized in that: include: Interconnected data acquisition module and data analysis module; The data acquisition module is used to acquire battery cabinet data through data acquisition equipment; The data analysis module generates a battery health score and a dedicated charging strategy based on the battery cabinet data; Generate dynamic battery priority based on battery cabinet data and battery health scores; replace batteries in descending order of priority.

Images