Aluminum air battery discharge fault diagnosis method and system

Abstract

The invention relates to the technical field of battery discharge fault diagnosis, particularly discloses an aluminum air battery discharge fault diagnosis method and system, and realizes refined and intelligent fault diagnosis and isolation of an aluminum air battery block. The method comprises the following steps: acquiring the content of dissolved oxygen at a channel in a to-be-detected battery pack through optical fluorescence method dissolved oxygen sensors arranged in a cooling channel and an exhaust channel of the battery pack, executing discharge analysis and air circulation operation analysis, preliminarily judging an air circulation fault, and generating a circulation early warning signal; obtaining the battery parameters of the to-be-detected battery pack, determining the battery parameters of the battery packs in different blocks based on a preset division strategy, and carrying out the feature extraction, and obtaining the features of a target block; dynamic data of different blocks in the discharging process are obtained, and the deviation between the actual average dynamic data range and the expected dynamic data range is analyzed; and performing evaluation according to the deviation corresponding to the target block and the features of the target block, and screening fault batteries.

Description

Technical Field

[0001] This invention relates to the field of battery discharge fault diagnosis technology, and specifically to a method and system for diagnosing discharge faults in aluminum-air batteries. Background Technology

[0002] Aluminum-air batteries show broad application prospects in electric vehicles, drones, backup power supplies, and special equipment. Their working principle relies on the electrochemical reaction between the aluminum anode and oxygen from the air in an electrolyte, directly converting chemical energy into electrical energy. However, during actual discharge operation, the performance and reliability of aluminum-air batteries are constrained by various complex factors, making them prone to various discharge failures that severely impact their lifespan and system safety.

[0003] To identify specific battery discharge faults, most existing methods rely solely on monitoring a single external electrical parameter such as the battery's terminal voltage or discharge current. However, different types of faults may exhibit similar electrical performance degradation characteristics in their early stages, making it difficult to accurately distinguish and locate them based on changes in voltage or current alone, which can easily lead to misdiagnosis or missed diagnosis.

[0004] In particular, during the actual detection of aluminum-air battery discharge faults, there is a lack of fault type localization, especially the difficulty in accurately distinguishing between battery channel faults and battery block faults; and using only fixed and uniform thresholds to judge the state of all battery blocks ignores the performance differences and inconsistencies of different individual batteries, which can easily lead to false alarms. Summary of the Invention

[0005] The purpose of this invention is to provide a method and system for diagnosing discharge faults in aluminum-air batteries, solving the following technical problems:

[0006] How to provide a fault diagnosis method that can accurately locate and provide early warnings for aluminum-air battery packs in a multi-dimensional and layered manner, so as to overcome the difficulty in distinguishing the root causes of faults in existing technologies;

[0007] How to promptly address the defects in aluminum-air battery packs, such as missing status monitoring, masked local faults, and high false alarm rates due to poor adaptability of diagnostic models.

[0008] The objective of this invention can be achieved through the following technical solutions:

[0009] A method for diagnosing discharge faults in aluminum-air batteries, the method comprising:

[0010] S1. The dissolved oxygen content in the channels of the battery pack under test is collected by an optical fluorescence dissolved oxygen sensor installed in the cooling channel and exhaust channel of the battery pack, and it is determined whether the dissolved oxygen content is within the preset oxygen concentration range threshold.

[0011] If so, perform discharge analysis and proceed to step S3;

[0012] If not, perform an air circulation operation analysis to determine if there is an air circulation malfunction:

[0013] If so, a cyclical warning signal will be generated;

[0014] If not, then it is determined that there is a physical component failure, and proceed to step S2;

[0015] S2. Obtain the battery parameters of the battery pack to be tested, and extract features of the battery parameters of different blocks of the battery pack based on the preset division strategy to obtain the features of the target block.

[0016] S3. Acquire dynamic data of different blocks during the discharge process, including voltage and temperature; analyze the deviation between the actual average dynamic data range and the expected dynamic data range.

[0017] S4. Evaluate based on the deviation and characteristics of the target block to screen for faulty batteries.

[0018] Preferably, the air circulation operation analysis in step S1 includes:

[0019] Start the standby circulation pump, increase the power of the main circulation pump, check and indicate air circuit blockage faults, and determine the type of fault.

[0020] Preferably, the preset partitioning strategy is to divide the battery pack into blocks based on the physical structure of the battery pack, where the blocks are modules, battery clusters, or virtual units composed of a specific number of cells.

[0021] Preferably, the battery parameters include static parameters and internal resistance, and the target block characteristics include at least one of the following: voltage range of cells within the block, capacity standard deviation of cells within the block, and internal resistance consistency coefficient of the block.

[0022] Preferably, the specific process of analyzing the deviation in step S3 is as follows: calculate the rate of change of voltage and temperature relative to the initial value in different blocks during the constant current discharge or pulse discharge stage, and compare it with the expected rate of change threshold.

[0023] Preferably, the process of calculating the rate of change of voltage relative to the initial value in different blocks during the constant current discharge stage and comparing it with the expected rate of change threshold is as follows:

[0024] For any given block, its voltage change rate deviation value Calculated using the following formula:

[0025]

[0026] in, This represents the average voltage of the block at the start of constant current discharge. This is the first time the block has been in constant current discharge since the start of the process. Average voltage at time; Discharge time; The expected voltage change rate benchmark value for the same type of fault-free block under the same test conditions;

[0027] Judgment when When the voltage deviation exceeds the preset threshold, the voltage dynamic data of that block is determined to be abnormal.

[0028] The process of calculating the rate of temperature change relative to the initial value for different blocks during the constant current discharge stage and comparing it with the expected rate of change threshold is as follows:

[0029]

[0030] in, This represents the average temperature at the start of constant current discharge in this block. This is the first time the block has been in constant current discharge since the start of the process. Average temperature at any given time; Discharge time; This is the baseline value for the expected temperature rise rate of a fault-free block of the same model under the same test conditions;

[0031] Judgment when If the temperature deviation exceeds the preset threshold, the temperature dynamic data of that block is determined to be abnormal.

[0032] Preferably, the expected dynamic data range is obtained by statistical analysis of historical data of the same type of fault-free battery pack under the same ambient temperature and discharge rate, or by simulation based on the battery electrochemical model.

[0033] Preferably, step S4 includes:

[0034] The target block features are weighted and fused with the corresponding deviations to obtain a comprehensive health score;

[0035] The overall health score is compared with a preset fault threshold to determine the block fault; when the overall health score is lower than the preset fault threshold, the block or a specific cell in the block is determined to be a faulty battery.

[0036] Preferably, the process of obtaining the comprehensive health score is as follows:

[0037] The target block features and corresponding deviations of each block are used to construct a feature vector. ,in, Representing the The characteristics of the target block represent the first target block. One dynamic data deviation;

[0038] eigenvectors Compare the similarity with multiple preset typical fault mode benchmark vectors;

[0039] Typical fault mode reference vectors include: internal resistance increase mode reference vector. Consistency degradation mode baseline vector and the reference vector for the local overheating mode caused by insufficient air circulation. ;

[0040] A weighted fusion based on similarity comparison results yields a comprehensive health score. Calculated using the following formula:

[0041]

[0042] in, For vectors and The cosine similarity function, , , Here are the weighting coefficients for each failure mode, and ;

[0043] Set the first fault threshold. With the second fault threshold ,and < ;

[0044] Judgment if ≥ If so, the block is deemed healthy;

[0045] Judgment if > ≥ If so, the performance of the block is determined to be degraded, and a cyclical warning signal is generated;

[0046] Judgment if < If so, the block is determined to be faulty.

[0047] An aluminum-air battery discharge fault diagnosis system is provided to implement a method for diagnosing aluminum-air battery discharge faults. The system includes:

[0048] The data acquisition and primary diagnostic module is used to collect the dissolved oxygen content in the channels of the battery pack under test through optical fluorescence dissolved oxygen sensors installed in the cooling and exhaust channels, and to determine whether the dissolved oxygen content is within a preset oxygen concentration range threshold.

[0049] If so, the discharge analysis module is triggered;

[0050] If not, the air circulation control and analysis module will be triggered;

[0051] The air circulation control and analysis module, upon receiving a trigger signal from the environmental condition monitoring and fault preliminary judgment module, performs air circulation operation analysis to determine whether an air circulation fault exists.

[0052] If so, a cyclical warning signal will be generated;

[0053] If not, it is determined that there is a physical component failure, and the feature extraction module is triggered;

[0054] The feature extraction module is used to obtain the battery parameters of the battery pack under test, and to extract features of the battery parameters of different blocks of the battery pack based on a preset partitioning strategy, thereby obtaining the features of the target block.

[0055] The discharge analysis module is used to acquire dynamic data of different blocks during the discharge process, including voltage and temperature; and to analyze the deviation between the actual average dynamic data range and the expected dynamic data range.

[0056] The evaluation module is used to evaluate and screen faulty batteries based on the deviation and characteristics of the target block.

[0057] The beneficial effects of this invention are:

[0058] Step S1, by determining the dissolved oxygen content, can immediately identify whether the problem lies in the battery's air circulation or electrochemical reaction, providing maintenance personnel with a clear direction for troubleshooting and greatly improving troubleshooting efficiency; it also enables precise location and effective isolation of the root cause of the fault, improving the accuracy of diagnosis.

[0059] Step S2 involves feature extraction and dynamic analysis by dividing the blocks (modules, battery clusters, or virtual units). An independent state analysis is established for each block. By calculating the consistency characteristics (such as voltage range, capacity standard deviation, and internal resistance consistency coefficient) and independent dynamic deviations within the block, the system can keenly detect signs of performance degradation in individual blocks, achieve refined local diagnosis, and prevent the expansion of local faults.

[0060] Steps S3 and S4 overcome the poor adaptability of the single threshold judgment method by adopting an intelligent evaluation method that compares the similarity of multiple feature vectors and multiple fault mode benchmark vectors. Dynamic data deviations are fused and evaluated, and a comprehensive health score is calculated through weighted fusion to accurately screen out faulty batteries. This process enables the diagnosis process to simulate expert experience and distinguish different types of fault modes (such as increased internal resistance, consistency degradation, and local overheating), significantly reducing the false alarm rate and false negative rate caused by individual battery differences and operating condition fluctuations, making the diagnostic conclusions more scientific and reliable.

[0061] Of course, any product implementing this invention does not necessarily need to achieve all the advantages described above at the same time. Attached Figure Description

[0062] To more clearly illustrate the technical solutions of the embodiments of the present invention, the accompanying drawings used in the description of the embodiments will be briefly introduced below. Obviously, the drawings described below are only some embodiments of the present invention. For those skilled in the art, other drawings can be obtained based on these drawings without creative effort.

[0063] Figure 1 This is a flowchart illustrating the steps of a method for diagnosing discharge faults in an aluminum-air battery according to the present invention.

[0064] Figure 2 This is a block diagram of an aluminum-air battery discharge fault diagnosis system according to the present invention. Detailed Implementation

[0065] The technical solutions of the embodiments of the present invention will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only some embodiments of the present invention, and not all embodiments. Based on the embodiments of the present invention, all other embodiments obtained by those skilled in the art without creative effort are within the scope of protection of the present invention.

[0066] Example 1

[0067] Please see Figure 1 As shown, this invention provides a method for diagnosing discharge faults in aluminum-air batteries. Through multi-dimensional data fusion and a hierarchical diagnostic strategy, it achieves accurate fault location in aluminum-air battery packs. The specific method steps include:

[0068] Firstly, in step S1, an optical fluorescence dissolved oxygen sensor, pre-installed in the battery pack cooling and exhaust channels, is used to collect the dissolved oxygen content in the flow channels in real time, enabling preliminary diagnosis of dissolved oxygen content and isolation of air faults. Then, the collected dissolved oxygen content is compared with a preset oxygen concentration range threshold for judgment.

[0069] If the dissolved oxygen content is within the preset threshold: this indicates that the air supply system is working properly and the battery performance problem may originate from the cell itself or the electrolyte. In this case, proceed directly to step S3 to perform discharge analysis.

[0070] If the dissolved oxygen content is lower than the preset threshold: This indicates an abnormality in the air supply, and an air circulation operation analysis is initiated. The air circulation operation analysis in this step includes: starting the backup circulation pump, increasing the power of the main circulation pump, checking and alerting to air path blockage faults, and determining the fault type.

[0071] If the dissolved oxygen content returns to the normal threshold after the above operations are performed, it is determined that there is an air circulation failure (such as a decrease in the efficiency of the main circulation pump), and a circulation warning signal is generated to prompt maintenance personnel to check the power components such as the circulation pump.

[0072] If the dissolved oxygen content does not improve significantly after the operation, a physical component fault is identified (such as a blocked air intake pipe, a failed filter, or a sensor malfunction), and the process proceeds to step S2. In step S1, when the voltage is abnormal, the dissolved oxygen content analysis immediately clarifies whether the problem lies in the battery's air circulation or an electrochemical reaction, providing maintenance personnel with a clear direction for troubleshooting and greatly improving troubleshooting efficiency. This achieves precise location and effective isolation of the root cause of the fault, improving diagnostic accuracy. Traditional methods infer the air system status through indirect parameters such as voltage and current, which are lagging and inaccurate. By detecting potential problems such as insufficient air circulation power or physical blockage of the air passage before the battery's electrical performance shows significant degradation, early warning is achieved, avoiding serious accidents such as sudden battery power failure due to oxygen supply interruption, and significantly improving the safety and reliability of system operation.

[0073] Secondly, in step S2, when it is determined that there is a physical component failure or that a detailed cell analysis is required, the system obtains the static battery parameters and internal resistance of the battery pack under test, including but not limited to: the static open-circuit voltage of each cell, the historical nominal capacity, and the internal resistance measured by the AC impedance method or the DC pulse method.

[0074] The system divides the battery pack into blocks based on a preset partitioning strategy. In this embodiment, a block can be a module or battery cluster based on physical structure, or a virtual unit composed of a specific number of cells based on management needs. For a large battery pack composed of multiple cells, early faults of individual cells or modules are easily masked by the overall data. By dividing the system into blocks (modules, battery clusters, or virtual units) for feature extraction and dynamic analysis, and establishing an independent state analysis for each block, the system can accurately detect performance degradation signs in individual blocks by calculating the consistency characteristics (such as voltage range, capacity standard deviation, and internal resistance consistency coefficient) and independent dynamic deviations within the block. This enables refined local diagnosis and prevents the expansion of local faults.

[0075] For each partitioned block, feature extraction is performed to obtain the target block features. These features are used to quantify the consistency state within the block, including:

[0076] Voltage range of cells within a block: reflects the synchronization of cells over a short period of time;

[0077] Standard deviation of cell capacity within a block: reflects the consistency of cell degradation over its lifespan;

[0078] Block internal resistance consistency coefficient The formula for calculating it is (maximum internal resistance - minimum internal resistance) / average internal resistance.

[0079] Third, in step S3, during the constant current discharge or pulse discharge stage, real-time dynamic data, namely voltage and temperature, are collected for each block; the deviation between the actual dynamic data and the expected dynamic data range for each block is analyzed. The expected dynamic data range is obtained through statistical analysis of a large amount of historical data from fault-free battery packs of the same model under the same ambient temperature and discharge rate, or it can be obtained through battery electrochemical model simulation. The specific process for analyzing the deviation in step S3 is as follows: the rate of change of voltage and temperature relative to the initial value for different blocks during the constant current discharge or pulse discharge stage is calculated and compared with the expected rate of change threshold. Specifically, the analysis process for voltage deviation is as follows:

[0080] The process of calculating the rate of change of voltage relative to the initial value in different blocks during the constant current discharge stage and comparing it with the expected rate of change threshold is as follows:

[0081] For any given block, its voltage change rate deviation value Calculated using the following formula:

[0082]

[0083] in, This represents the average voltage of the block at the start of constant current discharge. This is the first time the block has been in constant current discharge since the start of the process. Average voltage at time; Discharge time; The expected voltage change rate benchmark value for the same type of fault-free block under the same test conditions;

[0084] Judgment when When the voltage deviation exceeds the preset threshold, the voltage dynamic data of that block is determined to be abnormal.

[0085] The analysis process for temperature deviation is as follows:

[0086] For any given block, the process of calculating the rate of temperature change relative to the initial value during the constant current discharge phase and comparing it with the expected rate of change threshold is as follows:

[0087]

[0088] in, This represents the average temperature at the start of constant current discharge in this block. This is the first time the block has been in constant current discharge since the start of the process. Average temperature at any given time; Discharge time; This is the baseline value for the expected temperature rise rate of a fault-free block of the same model under the same test conditions;

[0089] Judgment when If the temperature deviation exceeds the preset threshold, the temperature dynamic data of that block is determined to be abnormal.

[0090] Step S3 above overcomes the poor adaptability of the single threshold judgment method by adopting an intelligent evaluation method that compares the similarity of multiple feature vectors with multiple fault mode benchmark vectors. This model not only considers real-time dynamic data (voltage, temperature deviation) but also integrates static / quasi-static features (internal resistance consistency, etc.) that characterize the long-term health status of the battery. By calculating a comprehensive health score through weighted fusion, the diagnostic process can simulate expert experience and distinguish different types of fault modes (such as increased internal resistance, consistency degradation, and local overheating), significantly reducing the false alarm rate and false negative rate caused by individual battery differences and operating condition fluctuations, making the diagnostic conclusions more scientific and reliable.

[0091] Fourth, step S4 fuses and evaluates the target block features extracted in step S2 with the dynamic data deviation calculated in step S3 to accurately screen out faulty batteries; the target block features and corresponding deviations are weighted and fused to obtain a comprehensive health score; specifically, the process of obtaining the comprehensive health score is as follows:

[0092] Constructing feature vectors: The target block features and corresponding deviations of each block are used to construct a feature vector. ,in, Representing the The characteristics of the target block represent the first target block. One dynamic data deviation;

[0093] Calculate the comprehensive health score: [The sentence is incomplete and requires more context to translate accurately.] The similarity is compared with several preset typical fault mode reference vectors; the typical fault mode reference vectors include: internal resistance increase mode reference vector. Consistency degradation mode baseline vector and the reference vector for the local overheating mode caused by insufficient air circulation. These baseline vectors are derived from a large amount of fault data, for example:

[0094] Reference vector for increasing internal resistance mode Characteristic: Voltage deviation Significantly high, while the block resistance consistency coefficient is also high. Larger than average, among which Indicates the first Characteristics of each target block (internal resistance consistency coefficient);

[0095] Consistency degradation mode baseline vector The characteristics are significantly higher block voltage range and capacity standard deviation;

[0096] Insufficient air circulation leads to localized overheating mode reference vector Characteristic: Temperature deviation Significantly higher, while voltage deviation It may not be obvious in the early stages;

[0097] A weighted fusion based on similarity comparison results yields a comprehensive health score. Calculated using the following formula:

[0098]

[0099] in, For vectors and The cosine similarity function, , , Here are the weighting coefficients for each failure mode, and ;

[0100] The overall health score is compared with a preset fault threshold to determine block faults; when the overall health score is lower than the preset fault threshold, the block or a specific cell within the block is determined to be a faulty battery. Specifically:

[0101] Set the first fault threshold. With the second fault threshold ,and < ;

[0102] Judgment if ≥ If so, the block is deemed healthy;

[0103] Judgment if > ≥ If so, the performance of the block is determined to be degraded, and a cyclical warning signal is generated;

[0104] Judgment if < If so, the block is determined to be faulty.

[0105] In one implementation, the preset partitioning strategy is to divide the battery pack into blocks based on its physical structure. The blocks are modules, battery clusters, or virtual units composed of a specific number of cells.

[0106] In one implementation, the expected dynamic data range is obtained by statistical analysis of historical data of fault-free battery packs of the same model under the same ambient temperature and discharge rate, or by simulation based on a battery electrochemical model.

[0107] Example 2

[0108] Please refer to the following: A discharge fault diagnosis system for aluminum-air batteries. Figure 2 As shown, a method for diagnosing discharge faults in aluminum-air batteries is implemented. The system includes:

[0109] The data acquisition and primary diagnostic module is used to collect the dissolved oxygen content in the channels of the battery pack under test through optical fluorescence dissolved oxygen sensors installed in the cooling and exhaust channels, and to determine whether the dissolved oxygen content is within a preset oxygen concentration range threshold.

[0110] If so, the discharge analysis module is triggered;

[0111] If not, the air circulation control and analysis module will be triggered;

[0112] The air circulation control and analysis module, upon receiving a trigger signal from the environmental condition monitoring and fault preliminary judgment module, performs air circulation operation analysis to determine whether an air circulation fault exists.

[0113] If so, a cyclical warning signal will be generated;

[0114] If not, it is determined that there is a physical component failure, and the feature extraction module is triggered;

[0115] The feature extraction module is used to obtain the battery parameters of the battery pack under test, and to extract features of the battery parameters of different blocks of the battery pack based on a preset partitioning strategy, thereby obtaining the features of the target block.

[0116] The discharge analysis module is used to acquire dynamic data of different blocks during the discharge process, including voltage and temperature; and to analyze the deviation between the actual average dynamic data range and the expected dynamic data range.

[0117] The evaluation module is used to evaluate and screen faulty batteries based on the deviation and characteristics of the target block.

[0118] In summary, the above-described Examples 1 and 2 comprehensively cover the main fault types of aluminum-air batteries through the above steps. Moreover, through automated judgment and control, the reliance on human experience is reduced, so that even without in-depth electrochemical expertise, precise maintenance can be performed based on the clear warning signals or fault location information output by the system designed in this method. The operation is convenient and practical.

[0119] The various embodiments in this specification are described in a progressive manner. Similar or identical parts between embodiments can be referred to mutually. Each embodiment focuses on describing the differences from other embodiments. In particular, the embodiments of apparatus, devices, and non-volatile computer storage media are basically similar to the method embodiments, and therefore described more simply; relevant parts can be referred to the descriptions of the method embodiments.

[0120] The above content is merely an example and illustration of the concept of the present invention. Those skilled in the art can make various modifications or additions to the specific embodiments described or use similar methods to replace them, as long as they do not deviate from the concept of the invention or exceed the scope defined in this application, they should all fall within the protection scope of the present invention.

Claims

1. A method for diagnosing discharge faults in aluminum-air batteries, characterized in that, The method includes: S1. The dissolved oxygen content in the channels of the battery pack under test is collected by an optical fluorescence dissolved oxygen sensor installed in the cooling channel and exhaust channel of the battery pack, and it is determined whether the dissolved oxygen content is within the preset oxygen concentration range threshold. If so, perform discharge analysis and proceed to step S3; If not, perform an air circulation operation analysis to determine if there is an air circulation malfunction: If so, a cyclical warning signal will be generated; If not, then it is determined that there is a physical component failure, and proceed to step S2; S2. Obtain the battery parameters of the battery pack to be tested, and extract features of the battery parameters of different blocks of the battery pack based on the preset division strategy to obtain the features of the target block. S3. Acquire dynamic data of different blocks during the discharge process, including voltage and temperature; analyze the deviation between the actual average dynamic data range and the expected dynamic data range. S4. Evaluate based on the deviation and characteristics of the target block to screen for faulty batteries.

2. The method for diagnosing discharge faults in an aluminum-air battery according to claim 1, characterized in that, The air circulation operation analysis in step S1 includes: Start the standby circulation pump, increase the power of the main circulation pump, check and indicate air circuit blockage faults, and determine the type of fault.

3. The method for diagnosing discharge faults in an aluminum-air battery according to claim 1, characterized in that, The preset partitioning strategy is to divide the battery pack into blocks based on its physical structure. The blocks are modules, battery clusters, or virtual units composed of a specific number of cells.

4. The method for diagnosing discharge faults in an aluminum-air battery according to claim 3, characterized in that, The battery parameters include static parameters and internal resistance, and the target block characteristics include at least one of the following: voltage range of cells within the block, capacity standard deviation of cells within the block, and internal resistance consistency coefficient of the block.

5. The method for diagnosing discharge faults in an aluminum-air battery according to claim 1, characterized in that, The specific process of analyzing the deviation in step S3 is as follows: calculate the rate of change of voltage and temperature relative to the initial value in different blocks during the constant current discharge or pulse discharge stage, and compare it with the expected rate of change threshold.

6. The method for diagnosing discharge faults in an aluminum-air battery according to claim 5, characterized in that, The process of calculating the rate of change of voltage relative to the initial value in different blocks during the constant current discharge stage and comparing it with the expected rate of change threshold is as follows: For any given block, its voltage change rate deviation value Calculated using the following formula: ； in, This represents the average voltage of the block at the start of constant current discharge. This is the first time the block has been in constant current discharge since the start of the process. Average voltage at time; Discharge time; The expected voltage change rate benchmark value for the same type of fault-free block under the same test conditions; Judgment when When the voltage deviation exceeds the preset threshold, the voltage dynamic data of that block is determined to be abnormal. The process of calculating the rate of temperature change of different blocks relative to the initial value during the constant current discharge stage and comparing it with the expected rate of change threshold includes: ； in, This represents the average temperature at the start of constant current discharge in this block. This is the first time the block has been in constant current discharge since the start of the process. Average temperature at any given time; Discharge time; This is the baseline value for the expected temperature rise rate of a fault-free block of the same model under the same test conditions; Judgment when If the temperature deviation exceeds the preset threshold, the temperature dynamic data of that block is determined to be abnormal.

7. The method for diagnosing discharge faults in an aluminum-air battery according to claim 6, characterized in that, The expected dynamic data range is obtained by statistical analysis of historical data of the same type of fault-free battery pack under the same ambient temperature and discharge rate, or by simulation based on the battery electrochemical model.

8. The method for diagnosing discharge faults in an aluminum-air battery according to claim 1, characterized in that, The S4 step includes: The target block features are weighted and fused with the corresponding deviations to obtain a comprehensive health score. The comprehensive health score is compared with a preset fault threshold to determine the block fault.

9. A method for diagnosing discharge faults in an aluminum-air battery according to claim 8, characterized in that, The process of obtaining the comprehensive health score is as follows: The target block features of each block and the corresponding deviation are used to construct a feature vector. ,in, Representing the The characteristics of the target block represent the first target block. One dynamic data deviation; The feature vector Compare the similarity with multiple preset typical fault mode benchmark vectors; The typical fault mode reference vector includes: internal resistance increase mode reference vector. Consistency degradation mode baseline vector and the reference vector for the local overheating mode caused by insufficient air circulation. ; The comprehensive health score is based on a weighted fusion of similarity comparison results. Calculated using the following formula: ； in, For vectors and The cosine similarity function, , , Here are the weighting coefficients for each failure mode, and ; Set the first fault threshold. With the second fault threshold ,and < ; Judgment if ≥ If so, the block is deemed healthy; Judgment if > ≥ If so, the performance of the block is determined to be degraded, and a cyclical warning signal is generated; Judgment if < If so, the block is determined to be faulty.

10. A discharge fault diagnosis system for aluminum-air batteries, characterized in that, The system, used to implement the aluminum-air battery discharge fault diagnosis method according to any one of claims 1-9, comprises: The data acquisition and primary diagnostic module is used to collect the dissolved oxygen content in the channels of the battery pack under test through optical fluorescence dissolved oxygen sensors installed in the cooling and exhaust channels, and to determine whether the dissolved oxygen content is within a preset oxygen concentration range threshold. If so, the discharge analysis module is triggered; If not, the air circulation control and analysis module will be triggered; The air circulation control and analysis module, upon receiving a trigger signal from the environmental condition monitoring and fault preliminary judgment module, performs air circulation operation analysis to determine whether an air circulation fault exists. If so, a cyclical warning signal will be generated; If not, it is determined that there is a physical component failure, and the feature extraction module is triggered; The feature extraction module is used to obtain the battery parameters of the battery pack under test, and to extract features of the battery parameters of different blocks of the battery pack based on a preset partitioning strategy, thereby obtaining the features of the target block. The discharge analysis module is used to acquire dynamic data of different blocks during the discharge process, including voltage and temperature; and to analyze the deviation between the actual average dynamic data range and the expected dynamic data range. The evaluation module is used to evaluate and screen faulty batteries based on the deviation and characteristics of the target block.

Images