Medium-and-long-term failure prediction and fault early warning method for storage battery pack

Abstract

The invention discloses a long-term failure prediction and fault early warning method in a storage battery pack, and relates to the technical field of data prediction, and the method comprises the following steps: S1, obtaining key characteristic parameters of the storage battery pack in real time; s2, constructing a health state evaluation model of the storage battery pack; s3, analyzing short-term dynamic change and long-term degradation trend of the battery; s4, a fault early warning model is constructed, and the model sends out early warning information when the battery is abnormal; s5, identifying and deducing a fault mode; and S6, making an optimization strategy. According to the long-term failure prediction and fault early warning method for the storage battery pack, the accuracy and timeliness of fault prediction of the storage battery are greatly improved through dynamic modeling of the health state. Compared with the prior art, the method has the advantages that the abnormal change of the battery can be quickly detected in a short time, the residual life of the battery is accurately predicted through long-term decline analysis, and the problems of misinformation and missing report in a traditional method are remarkably reduced.

Description

Technical Field

[0001] The present invention relates to the technical field of data prediction, and in particular to a method for predicting medium- and long-term failure of a battery pack and providing an early warning of a fault. Background Art

[0002] With the rapid development of energy storage technology, battery packs have been widely used in power systems, renewable energy storage, electric vehicles and other fields. However, battery packs will be affected by many factors during long-term use, causing their performance to gradually decline or even fail. In order to ensure the safe and reliable operation of battery packs, it is particularly important to timely predict their health status and provide fault warnings. At present, the health management of battery packs mainly relies on traditional monitoring methods based on single parameters, such as voltage, temperature and internal resistance. However, single parameter monitoring cannot fully reflect the complex operating characteristics of battery packs, especially when facing battery aging and potential faults, it cannot provide sufficient early warning capabilities.

[0003] Existing battery health management systems usually use some basic evaluation models, such as simple threshold determination methods based on battery voltage and temperature, or regression analysis based on historical data. These methods may be able to detect obvious signs of failure in the short term, but they often lack comprehensive evaluation and accurate prediction of the health status of the battery pack, especially when faced with nonlinear degradation and battery aging in complex working environments, and cannot respond efficiently and promptly. Summary of the invention

[0004] In view of the deficiencies in the prior art, the present invention provides a method for predicting medium- and long-term failures of a battery pack and for early warning of faults, so as to solve the problems raised in the above-mentioned background technology.

[0005] To achieve the above object, the present invention provides the following technical solutions: In a first aspect, an embodiment of the present invention provides a method for predicting medium- and long-term failure of a battery pack and early warning of a fault, comprising the following steps: S1. Obtain key characteristic parameters of the battery pack in real time; S2. Based on the acquired key characteristic parameters, a health status assessment model of the battery pack is constructed to monitor the health status of the battery pack in real time; S3. Analyze the short-term dynamic changes and long-term degradation trends of the battery based on the key characteristic parameters input by the health status assessment model and the health status value output; S4. Based on the key characteristic parameters obtained, a fault warning model is constructed, and the model issues a warning message when the battery is abnormal; S5. Identify and infer the failure mode based on the early warning information; S6. Develop optimization strategies based on the analysis results of short-term dynamic changes and long-term degradation trends.

[0006] To further optimize the technical solution, in step S1, the key characteristic parameters of the battery pack include current, internal resistance, temperature, number of charge and discharge cycles, and charging rate, which are used to characterize battery aging and performance degradation, and are acquired at high frequency through sensors.

[0007] To further optimize the technical solution, in step S2, in the health status assessment model of the battery pack: Assume that the battery's state of health SOH at time t is , and is subject to the internal resistance of the battery ,temperature , Charge and discharge times and charge and discharge rates The influence of factors; The battery pack health status assessment model is as follows: ; in, : Battery time The health status value at the time of the event is between 0 and 1, where 0 means complete failure and 1 means complete health. :time Internal resistance of the battery at all times; : Reference internal resistance value, which is the initial internal resistance value of the battery; :time The temperature of the battery at all times; : Reference temperature of the battery; : Battery temperature fluctuation range; :time The number of charge and discharge cycles of the battery at all times; : The maximum design cycle number of the battery; :time The intensity of charge and discharge current at the moment; : Maximum charge and discharge current of the battery; : is the weight coefficient, which is used to adjust the influence weight of each factor on the health status assessment.

[0008] To further optimize the technical solution, in step S3, when analyzing the short-term dynamic changes and long-term degradation trends of the battery, the following steps are included: Short-term dynamic changes: Use time series analysis methods including sliding average and exponential smoothing to focus on capturing rapid changes in battery health status values ​​and promptly identify sudden abnormalities or rapid degradation trends of batteries; Long-term degradation trend: Fit the slow-changing trend of battery charge and discharge cycle times, battery maximum charge and discharge current, and internal resistance parameters, and use nonlinear regression and Bayesian inference methods to predict the battery life cycle; Analyze the change range of health status value in different time periods. Short-term fluctuation of health status value indicates potential failure, and long-term degradation trend provides a basis for formulating battery replacement plan.

[0009] To further optimize this technical solution, in the short-term dynamic changes: The sliding average helps smooth the data and remove sudden noise and interference by taking the average of the parameter values ​​over a period of time. The sliding average is used to reflect the short-term trend of the battery status and detect abnormal fluctuations. Exponential smoothing is used to more sensitively capture rapid changes in battery status by giving higher weights to recent observations; When the internal resistance of the battery at a certain moment is suddenly greater than the sliding average, a warning signal is promptly issued through short-term dynamic analysis to indicate possible failure or performance deterioration.

[0010] Further optimizing the technical solution, in the long-term degradation trend: The degradation of the battery shows a nonlinear trend. The nonlinear regression model is used to fit the experimental data to capture the nonlinear change law of the battery state. The Bayesian method models the battery degradation process by introducing a combination of prior information and observed data, and provides uncertainty estimates of the degradation, thereby obtaining prediction results about the battery health status in long-term degradation analysis.

[0011] Further optimizing the technical solution, in step S4, when constructing the fault warning model, the key characteristic parameter set of the battery is set to ,in Indicates time The moment operating parameters, the fault warning model outputs the fault risk factor , the fault warning model is as follows: ; in, is the state change rate term; is the abnormality index item; is the trend prediction item; is the weight coefficient used to adjust the impact of the three indicators on the final failure risk factor.

[0012] To further optimize this technical solution, in the fault warning model: State change rate term The calculation formula is as follows: ; Used to describe time The key characteristic parameters of the battery are as follows: The rate of change of is used to capture fast-changing abnormal behaviors; A set of key characteristic parameters of the battery before the model prediction moment; Abnormality index The calculation formula is as follows: ; in, : No. The reference value of a key characteristic parameter, that is, the mean value or nominal value during normal operation; : No. The standard deviation of the parameters is used to normalize the deviation amplitude; : number of parameters; Trend forecast items The calculation formula is as follows: ; in, : The length of the time window for analysis; : At time The rate of change of state.

[0013] To further optimize the technical solution, in step S5, when identifying and inferring the fault mode, first obtain a large amount of historical fault data, and perform training based on the historical fault data. A fault mode recognition system is established based on support vector machines, decision trees, and deep neural networks. The warning information is input into the fault mode recognition system. The working state of the battery is analyzed based on the abnormal key characteristic parameters in the warning information, and different fault modes are identified and inferred, including battery leakage, capacity decay, and short circuit.

[0014] Further optimizing the technical solution, in step S6, the optimization strategy includes adjusting the battery charging and discharging strategy, optimizing the temperature control system, and regularly performing status monitoring and calibration; And automatically recommend the best maintenance cycle and replacement plan to reduce the risk of failure.

[0015] In a second aspect, an embodiment of the present invention provides a computer device, including a memory and a processor, wherein the memory stores a computer program, wherein: when the computer program instructions are executed by the processor, the steps of a method for predicting medium- and long-term failures and early warning of faults of a battery pack as described in the first aspect of the present invention are implemented.

[0016] In a third aspect, an embodiment of the present invention provides a computer-readable storage medium having a computer program stored thereon, wherein: when the computer program instructions are executed by a processor, the steps of a method for predicting medium- and long-term failures and early warning of faults of a battery pack as described in the first aspect of the present invention are implemented.

[0017] Compared with the prior art, the present invention provides a method for predicting medium- and long-term failure of a battery pack and early warning of a fault, which has the following beneficial effects: This method for predicting medium- and long-term failures and early warning of faults in a battery pack greatly improves the accuracy and timeliness of battery fault prediction through dynamic modeling of health status. Compared with the existing technology, this method can quickly detect abnormal changes in the battery in a short period of time, and accurately predict the remaining life of the battery through long-term decay analysis, significantly reducing the problems of false positives and false negatives in traditional methods. At the same time, based on a variety of key characteristic parameters of the battery, this method can provide a scientific basis for the operation and maintenance of the battery pack, help achieve more accurate health monitoring and fault warning, effectively reduce the operating risk of the system, and improve the safety and reliability of energy storage and electric transportation. BRIEF DESCRIPTION OF THE DRAWINGS

[0018] In order to more clearly illustrate the technical solutions of the embodiments of the present invention, the following briefly introduces the drawings required for use in the description of the embodiments. Obviously, the drawings described below are only some embodiments of the present invention. For ordinary technicians in this field, other drawings can be obtained based on these drawings without creative work.

[0019] Figure 1 A flowchart of a method for predicting medium- to long-term failure of a battery pack and early warning of a fault proposed by the present invention; Figure 2 The present invention provides a flowchart of a fault warning model in a method for predicting medium- to long-term failure of a battery pack and providing an early warning of a fault. DETAILED DESCRIPTION

[0020] In order to make the above-mentioned objects, features and advantages of the present invention more obvious and easy to understand, the specific implementation methods of the present invention are described in detail below in conjunction with the accompanying drawings.

[0021] In the following description, many specific details are set forth to facilitate a full understanding of the present invention, but the present invention may also be implemented in other ways different from those described herein, and those skilled in the art may make similar generalizations without violating the connotation of the present invention. Therefore, the present invention is not limited to the specific embodiments disclosed below.

[0022] Secondly, the term "one embodiment" or "embodiment" as used herein refers to a specific feature, structure, or characteristic that may be included in at least one implementation of the present invention. The term "in one embodiment" that appears in different places in this specification does not necessarily refer to the same embodiment, nor is it a separate or selective embodiment that is mutually exclusive with other embodiments.

[0023] Embodiment 1: Reference Figure 1-2 , which is the first embodiment of the present invention, provides a method for predicting medium- and long-term failure of a battery pack and early warning of a fault, comprising the following steps: S1. Obtain key characteristic parameters of the battery pack in real time.

[0024] In this embodiment, key characteristic parameters of the battery pack include current, internal resistance, temperature, number of charge and discharge cycles, and charging rate, which are used to characterize battery aging and performance degradation, and are acquired at high frequency through sensors.

[0025] In particular, changes in internal resistance are an early sign of battery aging, and internal resistance monitoring can identify potential anomalies before battery pack failures occur. In addition, as the battery operating environment changes (such as external factors such as temperature and humidity), these environmental variables also need to be recorded in real time and analyzed jointly with battery characteristic parameters.

[0026] S2. Based on the acquired key characteristic parameters, a health status assessment model for the battery pack is constructed to monitor the health status of the battery pack in real time.

[0027] In this embodiment, in the health status assessment model of the battery pack: Assume that the battery's state of health SOH at time t is , and is subject to the internal resistance of the battery ,temperature , Charge and discharge times and charge and discharge rates The influence of factors; The battery pack health status assessment model is as follows: ; in, : Battery time The health status value at the time of the event is between 0 and 1, where 0 means complete failure and 1 means complete health. :time Internal resistance of the battery at all times; : Reference internal resistance value, which is the initial internal resistance value of the battery; :time The temperature of the battery at all times; : Reference temperature of the battery; : Battery temperature fluctuation range; :time The number of charge and discharge cycles of the battery at all times; : The maximum design cycle number of the battery; :time The intensity of charge and discharge current at the moment; : Maximum charge and discharge current of the battery; : is the weight coefficient, which is used to adjust the influence weight of each factor on the health status assessment.

[0028] In the model, Internal resistance Impact on SOH: As battery usage increases, the internal resistance of the battery will gradually increase, which means that the energy conduction efficiency inside the battery decreases, affecting the output capacity of the battery. The term reflects the impact of internal resistance changes on the health state. A larger internal resistance value means that the battery performance is degraded and the SOH value will be reduced.

[0029] temperature Impact on SOH: The temperature of the battery directly affects its chemical reaction rate and life. Too high a temperature will accelerate the battery's decay process. The term represents the deviation of the battery operating temperature from the reference temperature, reflecting the negative impact of temperature on battery health. The more the temperature deviates from the normal value, the lower the SOH value.

[0030] Charge and discharge times Impact on SOH: The number of charge and discharge cycles of a battery is one of the key factors that determine the battery life. Each charge and discharge will bring about a certain attenuation, causing the battery capacity to gradually decrease. The term represents the ratio of the battery's charge and discharge cycles to the maximum design cycle number, reflecting the frequency of battery use and its impact on its health status.

[0031] Charge and discharge current Impact on SOH: Excessive charge and discharge current will accelerate the aging process of the battery and shorten the battery life. The term represents the ratio of the charging current to the maximum charging current, reflecting the impact of the current on the health state.

[0032] In practical applications, the weight coefficient It needs to be adjusted according to the specific battery type, working environment and usage scenario. The weight coefficient can be automatically adjusted through machine learning algorithms, such as regression analysis or Bayesian optimization, so that the model can adapt to the health assessment needs of different types of batteries.

[0033] S3. Based on the key characteristic parameters input into the health status assessment model and the health status value output, analyze the short-term dynamic changes and long-term degradation trends of the battery.

[0034] In this embodiment, the analysis of the short-term dynamic changes and long-term degradation trends of the battery includes: Short-term dynamic changes: Use time series analysis methods including sliding average and exponential smoothing to focus on capturing rapid changes in battery health status values ​​and promptly identify sudden abnormalities or rapid degradation trends of batteries; Long-term degradation trend: Fit the slow-changing trend of battery charge and discharge cycle times, battery maximum charge and discharge current, and internal resistance parameters, and use nonlinear regression and Bayesian inference methods to predict the battery life cycle; Analyze the change range of health status value in different time periods. Short-term fluctuation of health status value indicates potential failure, and long-term degradation trend provides a basis for formulating battery replacement plan.

[0035] Furthermore, in the short-term dynamic changes: The sliding average helps smooth the data and remove sudden noise and interference by taking the average of the parameter values ​​over a period of time. The sliding average is used to reflect the short-term trend of the battery status and detect abnormal fluctuations. Exponential smoothing is used to more sensitively capture rapid changes in battery status by giving higher weights to recent observations; When the internal resistance of the battery at a certain moment is suddenly greater than the sliding average, a warning signal is promptly issued through short-term dynamic analysis to indicate possible failure or performance deterioration.

[0036] Furthermore, in the long-term degradation trend: The degradation of the battery shows a nonlinear trend. The nonlinear regression model is used to fit the experimental data to capture the nonlinear change law of the battery state. The Bayesian method models the battery degradation process by introducing a combination of prior information and observed data, and provides uncertainty estimates of the degradation, thereby obtaining prediction results about the battery health status in long-term degradation analysis.

[0037] If the changes in the internal resistance growth rate and capacity decay of the battery over a period of time conform to the decay law, the remaining service life of the battery (such as the remaining number of charge and discharge cycles) can be predicted. Through long-term decay analysis, the battery management system (BMS) can issue a battery replacement prompt when the battery capacity drops to a certain critical value, avoiding continued use when the battery performance decays to the point where it can no longer meet usage needs.

[0038] S4. Based on the key characteristic parameters obtained, a fault warning model is constructed, and the model issues a warning message when the battery is abnormal.

[0039] In this embodiment, when constructing the fault warning model, the key characteristic parameter set of the battery is ,in Indicates time The moment operating parameters, the fault warning model outputs the fault risk factor , the fault warning model is as follows: ; in, is the state change rate term; is the abnormality index item; is the trend prediction item; is the weight coefficient used to adjust the impact of the three indicators on the final failure risk factor.

[0040] Furthermore, in the fault warning model: State change rate term The calculation formula is as follows: ; Used to describe time The key characteristic parameters of the battery are as follows: The rate of change of is used to capture fast-changing abnormal behaviors; A set of key characteristic parameters of the battery before the model prediction moment; The state change rate term represents the change rate vector of the parameter state, reflecting the severity of the operating state at the current moment.

[0041] Abnormality index The calculation formula is as follows: ; in, : No. The reference value of a key characteristic parameter, that is, the mean value or nominal value during normal operation; : No. The standard deviation of the parameters is used to normalize the deviation amplitude; : number of parameters.

[0042] The abnormality index item represents the deviation between the current state and the normal reference value (calculated based on the Z-score), which measures the overall abnormality of the system.

[0043] Trend forecast items The calculation formula is as follows: ; in, : The length of the time window for analysis; : At time The rate of change of state.

[0044] Indicates the past period of time The fluctuation trend of internal state changes is used to capture potential risks of rapid deterioration.

[0045] Automatic calibration via optimization algorithms such as genetic algorithms or Bayesian optimization.

[0046] When used, the model includes: Real-time monitoring and calculation: During operation, the battery parameter data (such as internal resistance, current, etc.) is collected through sensors to At the same time, the reference value of each parameter is calculated based on historical data. and standard deviation The model is constructed through the state change rate term Capture rapid abnormal fluctuations through abnormality indicators Assess the overall degree of deviation from the status.

[0047] Threshold warning: Set risk thresholds. When it is greater than the risk threshold, an early warning signal is triggered. In order to adapt to different environments and battery usage scenarios, the threshold can be dynamically adjusted according to the operating data. For example, in extreme temperature environments, the threshold tolerance is increased to avoid frequent false alarms.

[0048] Trend Assessment and Intervention: Trend Prediction Items of the Model Sensitive to the accumulation of historical fluctuations, so If the threshold is not exceeded, it can still be observed Identify potential failure trends in advance. When it is higher than the preset value, the system can take intervention measures in advance (such as reducing the charge and discharge rate or improving the heat dissipation conditions).

[0049] Multi-parameter adaptability: By adjusting the input parameter set The model can be adapted to different types of battery systems (such as single cells or battery packs) by flexible definition of the parameters. It can be optimized according to different application scenarios (such as electric vehicles or energy storage systems).

[0050] Self-learning and optimization: The fault warning model supports self-learning based on historical fault data. By continuously updating the reference value and weight coefficient, the system can improve the accuracy of the warning and reduce the false alarm rate.

[0051] In this embodiment, two models are actually applied, and the influence of temperature and internal resistance are considered. The application scenarios are as follows: System parameters: A lithium-ion battery pack with a capacity of 100Ah.

[0052] Initial State of Health (SOH): 100%.

[0053] Normal operating conditions: Temperature range: 15-45℃ Internal resistance range: Health status warning value: 70% Risk threshold of failure risk factor: 0.75.

[0054] Based on the model, operation data is recorded and analyzed: Initial state (time point ) temperature: Internal resistance: Health status: Failure risk factors: The battery pack operates stably in the initial state, all parameters are within the normal range, and there is no abnormal risk.

[0055] Running (time point , record abnormal changes): temperature: (Beyond normal range) Internal resistance: (Deviation from normal range) Health status: calculated according to the health status assessment model, 78% Fault risk factor: calculated according to the fault warning model, 0.88 The temperature and internal resistance during operation increase abnormally, the health status approaches the warning value, and the failure risk factor exceeds the risk threshold.

[0056] Triggering an early warning signal indicates abnormal growth in temperature and internal resistance, a sharp drop in health status, and a high risk of failure.

[0057] Suggestions: Short-term measures: Immediately reduce the battery operating load and check whether the cooling system is abnormal and causes the temperature to rise.

[0058] Long-term measures: Perform in-depth maintenance on the battery pack and replace aging cells when necessary to avoid system failure.

[0059] S5. Based on the early warning information, identify and infer the failure mode.

[0060] In this embodiment, when identifying and inferring fault modes, a large amount of historical fault data is first obtained, and training is performed based on the historical fault data. A fault mode recognition system is established based on support vector machines, decision trees, and deep neural networks. The warning information is input into the fault mode recognition system, and the working status of the battery is analyzed based on the abnormal key characteristic parameters in the warning information. Different fault modes are identified and inferred, including battery leakage, capacity decay, and short circuit.

[0061] Fault mode identification not only helps improve the accuracy of fault prediction, but also helps operation and maintenance personnel take targeted intervention measures to extend the service life of the battery.

[0062] S6. Develop optimization strategies based on the analysis results of short-term dynamic changes and long-term degradation trends.

[0063] In this embodiment, the optimization strategy includes adjusting the battery charging and discharging strategy, optimizing the temperature control system, and regularly performing status monitoring and calibration; And automatically recommend the best maintenance cycle and replacement plan to reduce the risk of failure.

[0064] Different maintenance and management strategies can be tailored based on the use scenarios of battery packs (such as electric vehicles, energy storage power stations, etc.). For example, for battery packs under high load conditions, the heat dissipation design can be strengthened to reduce the impact of temperature on battery degradation; while for batteries under low load conditions, the charging strategy can be optimized to slow down the degradation rate. Through precise health management and optimization measures, the efficiency and life of the battery can be greatly improved.

[0065] Embodiment 2: This embodiment also provides a computer device, which is suitable for a method for predicting medium- and long-term failures and early warning of faults of a battery pack, and includes a memory and a processor; the memory is used to store computer-executable instructions, and the processor is used to execute computer-executable instructions to implement a method for predicting medium- and long-term failures and early warning of faults of a battery pack as proposed in the above embodiment.

[0066] This embodiment also provides a storage medium on which a computer program is stored. When the program is executed by a processor, a method for predicting medium- and long-term failure of a battery pack and providing an early warning of a fault is implemented as proposed in the above embodiment.

[0067] The computer device may be a terminal, and the computer device includes a processor, a memory, a communication interface, a display screen and an input device connected through a system bus. Among them, the processor of the computer device is used to provide computing and control capabilities. The memory of the computer device includes a non-volatile storage medium and an internal memory. The non-volatile storage medium stores an operating system and a computer program. The internal memory provides an environment for the operation of the operating system and the computer program in the non-volatile storage medium. The communication interface of the computer device is used to communicate with an external terminal in a wired or wireless manner, and the wireless manner can be achieved through WIFI, an operator network, NFC (near field communication) or other technologies. The display screen of the computer device may be a liquid crystal display screen or an electronic ink display screen, and the input device of the computer device may be a touch layer covered on the display screen, or a key, trackball or touchpad provided on the housing of the computer device, or an external keyboard, touchpad or mouse, etc.

[0068] If the function is implemented in the form of a software functional unit and sold or used as an independent product, it can be stored in a computer-readable storage medium. Based on this understanding, the technical solution of the present invention, or the part that contributes to the prior art, or the part of the technical solution, can be embodied in the form of a software product. The computer software product is stored in a storage medium, including several instructions for a computer device (which can be a personal computer, server, or network device, etc.) to perform all or part of the steps of the methods of each embodiment of the present invention. The aforementioned storage medium includes: U disk, mobile hard disk, read-only memory (ROM, Read-Only Memory), random access memory (RAM, Random Access Memory), disk or optical disk, etc. Various media that can store program codes.

[0069] The logic and / or steps represented in the flowchart or otherwise described herein, for example, can be considered as an ordered list of executable instructions for implementing logical functions, and can be embodied in any computer-readable medium for use by an instruction execution system, device or apparatus (such as a computer-based system, a system including a processor, or other system that can fetch instructions from an instruction execution system, device or apparatus and execute instructions), or in conjunction with such instruction execution systems, devices or apparatuses. For the purposes of this specification, "computer-readable medium" can be any device that can contain, store, communicate, propagate or transmit a program for use by an instruction execution system, device or apparatus, or in conjunction with such instruction execution systems, devices or apparatuses.

[0070] More specific examples of computer-readable media (a non-exhaustive list) include the following: an electrical connection with one or more wires (electronic devices), a portable computer disk case (magnetic devices), a random access memory (RAM), a read-only memory (ROM), an erasable and programmable read-only memory (EPROM or flash memory), an optical fiber device, and a portable compact disk read-only memory (CDROM). In addition, the computer-readable medium may even be a paper or other suitable medium on which the program is printed, since the program may be obtained electronically, for example, by optically scanning the paper or other medium, followed by editing, deciphering or, if necessary, processing in another suitable manner, and then stored in a computer memory.

[0071] It should be understood that the various parts of the present invention can be implemented by hardware, software, firmware or a combination thereof. In the above embodiments, multiple steps or methods can be implemented by software or firmware stored in a memory and executed by a suitable instruction execution system. For example, if implemented by hardware, as in another embodiment, it can be implemented by any one of the following technologies known in the art or their combination: a discrete logic circuit with a logic gate circuit for implementing a logic function for a data signal, a dedicated integrated circuit with a suitable combination of logic gate circuits, a programmable gate array (PGA), a field programmable gate array (FPGA), etc.

[0072] It should be noted that the above embodiments are only used to illustrate the technical solutions of the present invention rather than to limit it. Although the present invention has been described in detail with reference to the preferred embodiments, those skilled in the art should understand that the technical solutions of the present invention may be modified or replaced by equivalents without departing from the spirit and scope of the technical solutions of the present invention, which should all be included in the scope of the claims of the present invention.

Claims

1. A method for predicting medium- and long-term failure of a battery pack and early warning of faults, characterized in that: The following steps are involved: S1. Obtain key characteristic parameters of the battery pack in real time; S2. Based on the acquired key characteristic parameters, a health status assessment model of the battery pack is constructed to monitor the health status of the battery pack in real time; S3. Analyze the short-term dynamic changes and long-term degradation trends of the battery based on the key characteristic parameters input by the health status assessment model and the health status value output; S4. Based on the key characteristic parameters obtained, a fault warning model is constructed, and the model issues a warning message when the battery is abnormal; S5. Identify and infer the failure mode based on the early warning information; S6. Develop optimization strategies based on the analysis results of short-term dynamic changes and long-term degradation trends.

2. A method for predicting medium- and long-term failure of a battery pack and early warning of a fault according to claim 1, characterized in that: In step S1, key characteristic parameters of the battery pack include current, internal resistance, temperature, number of charge and discharge cycles, and charging rate, which are used to characterize battery aging and performance degradation, and are acquired at high frequency through sensors.

3. The method for predicting medium- and long-term failure of a battery pack and early warning of a fault according to claim 1, characterized in that: In step S2, in the health status assessment model of the battery pack: Assume that the battery's state of health SOH at time t is , and is subject to the internal resistance of the battery ,temperature , Charge and discharge times and charge and discharge rates The influence of factors; The battery pack health status assessment model is as follows: ； in, : Battery time The health status value at the time of the event is between 0 and 1, where 0 means complete failure and 1 means complete health. :time Internal resistance of the battery at all times; : Reference internal resistance value, which is the initial internal resistance value of the battery; :time The temperature of the battery at all times; : Reference temperature of the battery; : Battery temperature fluctuation range; :time The number of charge and discharge cycles of the battery at all times; : The maximum design cycle number of the battery; :time The intensity of charge and discharge current at the moment; : Maximum charge and discharge current of the battery; : is the weight coefficient, which is used to adjust the influence weight of each factor on health status assessment.

4. The method for predicting medium- and long-term failure of a battery pack and early warning of a fault according to claim 1, characterized in that: In step S3, analyzing the short-term dynamic changes and long-term degradation trends of the battery includes: Short-term dynamic changes: Use time series analysis methods including sliding average and exponential smoothing to focus on capturing rapid changes in battery health status values ​​and promptly identify sudden abnormalities or rapid degradation trends of batteries; Long-term degradation trend: Fit the slow-changing trend of battery charge and discharge cycle times, battery maximum charge and discharge current, and internal resistance parameters, and use nonlinear regression and Bayesian inference methods to predict the battery life cycle; Analyze the change range of health status value in different time periods. Short-term fluctuation of health status value indicates potential failure, and long-term degradation trend provides a basis for formulating battery replacement plan.

5. A method for predicting medium- and long-term failure of a battery pack and early warning of faults according to claim 4, characterized in that: In the short-term dynamic changes: The sliding average helps smooth the data and remove sudden noise and interference by taking the average of the parameter values ​​over a period of time. The sliding average is used to reflect the short-term trend of the battery status and detect abnormal fluctuations. Exponential smoothing is used to more sensitively capture rapid changes in battery status by giving higher weights to recent observations; When the internal resistance of the battery at a certain moment is suddenly greater than the sliding average, a warning signal is promptly issued through short-term dynamic analysis to indicate possible failure or performance deterioration.

6. A method for predicting medium- to long-term failure and early warning of faults of a battery pack according to claim 4, characterized in that: Among the long-term deterioration trends: The degradation of the battery shows a nonlinear trend. The nonlinear regression model is used to fit the experimental data to capture the nonlinear change law of the battery state. The Bayesian method models the battery degradation process by introducing a combination of prior information and observed data, and provides uncertainty estimates of the degradation, thereby obtaining prediction results about the battery health status in long-term degradation analysis.

7. The method for predicting medium- to long-term failure and early warning of faults of a battery pack according to claim 1, characterized in that: In step S4, when constructing the fault warning model, the key characteristic parameter set of the battery is set as ,in Indicates time The moment operating parameters, the fault warning model outputs the fault risk factor , the fault warning model is as follows: ； in, is the state change rate term; is the abnormality index item; is the trend prediction item; is the weight coefficient used to adjust the impact of the three indicators on the final failure risk factor.

8. A method for predicting medium- to long-term failure and early warning of faults of a battery pack according to claim 7, characterized in that: In the fault warning model: State change rate term The calculation formula is as follows: ； Used to describe time The key characteristic parameters of the battery are as follows: The rate of change of is used to capture fast-changing abnormal behaviors; A set of key characteristic parameters of the battery before the model prediction moment; Abnormality index The calculation formula is as follows: ； in, : No. The reference value of a key characteristic parameter, that is, the mean value or nominal value during normal operation; : No. The standard deviation of the parameters is used to normalize the deviation amplitude; : number of parameters; Trend forecast items The calculation formula is as follows: ； in, : The length of the time window for analysis; : At time The rate of change of state.

9. The method for predicting medium- to long-term failure and early warning of faults of a battery pack according to claim 1, characterized in that: In step S5, when identifying and inferring the fault mode, first obtain a large amount of historical fault data, and perform training based on the historical fault data. A fault mode recognition system is established based on support vector machines, decision trees, and deep neural networks. The warning information is input into the fault mode recognition system. The working state of the battery is analyzed based on the abnormal key characteristic parameters in the warning information, and different fault modes are identified and inferred, including battery leakage, capacity decay, and short circuit.

10. The method for predicting medium- to long-term failure and early warning of faults of a battery pack according to claim 1, characterized in that: In step S6, the optimization strategy includes adjusting the battery charging and discharging strategy, optimizing the temperature control system, and regularly performing status monitoring and calibration; And automatically recommend the best maintenance cycle and replacement plan to reduce the risk of failure.

Images