Power storage battery health state monitoring method and system

Abstract

The invention relates to the technical field of power storage battery monitoring, in particular to a power storage battery health state monitoring method which comprises the following steps: acquiring multi-source state information of a power storage battery in real time, and acquiring ohm internal resistance, first charge transfer internal resistance, a comprehensive thermal impedance index and a first thermally induced vibration frequency correction value according to the multi-source state information; correcting a first charge transfer internal resistance based on the first thermally induced vibration frequency correction value; vectorizing the ohm internal resistance, the second charge transfer internal resistance, the first thermally induced vibration frequency correction value and the comprehensive thermal impedance index; dynamically predicting the health state of the battery, and outputting a prediction result; according to the method, through physical relevance design, electrochemical data, mechanical data and thermodynamic data are organically combined, the problem of data isolation in a traditional method is solved, and more accurate monitoring of the health state of the power storage battery is achieved.

Description

TECHNICAL FIELD

[0001] The present application relates to the technical field of power storage battery monitoring, in particular to a power storage battery health state monitoring method and system. BACKGROUND

[0002] The power storage battery refers to a storage battery providing power for electric vehicles, electric logistics vehicles, industrial and medical electric devices, portable electronic devices and the like.

[0003] The traditional power storage battery health state monitoring method mainly relies on voltage and current parameters to evaluate the battery health state. However, only according to these parameters, the complex changes in the battery, such as electrode interface aging, electrolyte drying and SEI film thickening, cannot be comprehensively and accurately reflected, and the battery health state cannot be accurately judged through simple voltage and current measurement. SUMMARY

[0004] The main purpose of the present application is to provide a power storage battery health state monitoring method and system, which aims to solve the technical problems in the background.

[0005] The present application provides a power storage battery health state monitoring method, which comprises the following steps:

[0006] Real-time acquisition of multi-source state information of the power storage battery, wherein the multi-source state information comprises voltage information, current information, temperature information, electrochemical impedance spectrum information and micro-vibration acceleration information;

[0007] Obtaining an ohmic internal resistance, a first charge transfer internal resistance and a comprehensive thermal impedance index according to the multi-source state information;

[0008] Obtaining a first thermal vibration frequency correction value according to the temperature information and the micro-vibration acceleration information;

[0009] Correcting the first charge transfer internal resistance based on the first thermal vibration frequency correction value to obtain a second charge transfer internal resistance;

[0010] Vectorizing the ohmic internal resistance, the second charge transfer internal resistance, the first thermal vibration frequency correction value and the comprehensive thermal impedance index respectively to obtain a multi-source feature vector;

[0011] Constructing a power storage battery health state comprehensive evaluation model based on a neural network, and inputting the multi-source feature vector into the power storage battery health state comprehensive evaluation model to dynamically predict the battery health state and output a prediction result;

[0012] Executing a battery charging and discharging instruction corresponding to the prediction result.

[0013] Preferably, the step of obtaining the first thermal vibration frequency correction value according to the temperature information and the micro-vibration acceleration information comprises:

[0014] extracting a main frequency and a sub-band energy proportion according to a frequency spectrum corresponding to the micro-vibration acceleration information;

[0015] obtaining a cell surface temperature of the power battery and a cell surface preset standard temperature according to the temperature information;

[0016] obtaining a thermal expansion coupling coefficient;

[0017] obtaining a temperature interference coefficient according to the cell surface temperature, the cell surface preset standard temperature and the thermal expansion coupling coefficient;

[0018] obtaining the first thermal vibration frequency correction value according to the temperature interference coefficient and the main frequency;

[0019] obtaining a micro-vibration sensitive coefficient according to the sub-band energy proportion;

[0020] judging whether the micro-vibration sensitive coefficient is greater than a preset value, and adjusting the first thermal vibration frequency correction value according to the micro-vibration sensitive coefficient if the micro-vibration sensitive coefficient is greater than the preset value.

[0021] Preferably, the step of correcting the first charge transfer resistance based on the first thermal vibration frequency correction value to obtain the second charge transfer resistance comprises:

[0022] obtaining a second thermal vibration frequency correction value of the battery in a primary use state;

[0023] calculating a dynamic change rate of the first thermal vibration frequency correction value and the second thermal vibration frequency correction value;

[0024] obtaining a battery packaging material stiffness correction coefficient;

[0025] obtaining a thermal vibration frequency dynamic correction coefficient according to the dynamic change rate and the battery packaging material stiffness correction coefficient;

[0026] adjusting the first charge transfer resistance based on the thermal vibration frequency dynamic correction coefficient to obtain the second charge transfer resistance.

[0027] Preferably, the step of obtaining the ohmic resistance, the first charge transfer resistance and the comprehensive thermal impedance index according to the multi-source state information comprises:

[0028] obtaining a battery thermal gradient value according to the multi-source state information;

[0029] extracting a high-frequency impedance imaginary part value according to the electrochemical impedance spectrum information;

[0030] calculating a comprehensive thermal impedance index according to the battery thermal gradient value and the high-frequency impedance imaginary number value.

[0031] Preferably, the step of obtaining the battery thermal gradient value according to the multi-source state information comprises:

[0032] obtaining the battery charge-discharge rate, the battery environmental convection coefficient, the battery column-level temperature, the battery surface temperature and the battery environmental temperature according to the multi-source state information;

[0033] obtaining an internal thermal resistance influence factor based on the battery charge-discharge rate;

[0034] obtaining an external heat dissipation influence factor based on the battery environmental convection coefficient;

[0035] obtaining the battery thermal gradient value according to the battery column-level temperature, the battery surface temperature, the battery environmental temperature, the internal thermal resistance influence factor and the external heat dissipation influence factor.

[0036] Preferably, the step of inputting the multi-source feature vector into the power storage battery health state comprehensive evaluation model to dynamically predict the battery health state and output the prediction result comprises:

[0037] obtaining the multi-source feature vector;

[0038] inputting the multi-source feature vector into a long short-term memory neural network, processing the multi-source feature vector based on the long short-term memory neural network, and outputting a battery health state prediction value;

[0039] obtaining a disturbance compensation factor at a target time, and adjusting the disturbance compensation factor at the target time based on a disturbance compensation factor at a previous time of the target time;

[0040] obtaining a battery health state evaluation value according to the disturbance compensation factor and the battery health state prediction value;

[0041] obtaining battery health state evaluation values at multiple different times, predicting the battery health state according to the change rates of the battery health state evaluation values at the multiple different times, and outputting a prediction result.

[0042] Preferably, the step of adjusting the disturbance compensation factor at the target time based on the disturbance compensation factor at the previous time of the target time comprises:

[0043] calculating a first covariance according to the disturbance compensation factors at the previous two times of the target time;

[0044] calculating a second covariance according to the disturbance compensation factors at the target time and the previous time;

[0045] calculating an error covariance according to the first covariance and the second covariance;

[0046] repeating the steps of calculating the first covariance according to the interference compensation factors of the target time and the two previous times to calculating the error covariance according to the first covariance and the second covariance for a preset number of times to obtain a plurality of error covariances, and dynamically adjusting the interference compensation factor of the target time according to the error covariances.

[0047] The application further discloses a power storage battery health state monitoring system, characterized by comprising:

[0048] a data acquisition module configured to acquire multi-source state information of the power storage battery in real time, wherein the multi-source state information comprises voltage information, current information, temperature information, electrochemical impedance spectrum information and micro-vibration acceleration information;

[0049] an acquisition module configured to acquire an ohmic internal resistance, a first charge transfer internal resistance and a comprehensive thermal impedance index according to the multi-source state information;

[0050] a first thermal vibration frequency correction value is acquired according to the temperature information and the micro-vibration acceleration information;

[0051] a correction module configured to correct the first charge transfer internal resistance based on the first thermal vibration frequency correction value to obtain a second charge transfer internal resistance;

[0052] a multi-source fusion module configured to vectorize the ohmic internal resistance, the second charge transfer internal resistance, the first thermal vibration frequency correction value and the comprehensive thermal impedance index respectively to obtain a multi-source feature vector;

[0053] a calculation module configured to construct a power storage battery health state comprehensive evaluation model based on a neural network, and input the multi-source feature vector into the power storage battery health state comprehensive evaluation model to dynamically predict the battery health state and output a prediction result;

[0054] a control execution module configured to execute a battery charging and discharging instruction corresponding to the prediction result.

[0055] Preferably, the control execution module comprises:

[0056] a first-level response unit configured to limit a discharging current and activate a balanced charging;

[0057] a second-level response unit configured to cut off a connection between a device and the power storage battery and report a fault code;

[0058] a communication unit configured to transmit an emergency instruction.

[0059] Preferably, the control execution module further comprises:

[0060] A low-temperature heating unit is used to activate the heating film and adjust the charging parameters in a low-temperature environment.

[0061] The application has the advantages that: the application combines electrochemistry, mechanics and thermodynamics organically through physical correlation design, can reflect the battery aging caused by electrode interface aging, electrolyte drying and SEI film thickening, solves the problem of relying on voltage and current parameters to evaluate the battery health state in the traditional method, and realizes more accurate monitoring of the power storage battery health state. BRIEF DESCRIPTION OF DRAWINGS

[0062] Figure 1 It is a method flowchart of an embodiment of the application.

[0063] Figure 2 It is a system structure schematic diagram of an embodiment of the application.

[0064] The implementation, functional features and advantages of the application will be further described with reference to the embodiments and the accompanying drawings. DETAILED DESCRIPTION

[0065] It should be understood that the specific embodiments described herein are only used to explain the application, and are not used to limit the application.

[0066] As shown in the accompanying drawings, Figure 1 The application provides a power storage battery health state monitoring method, which comprises:

[0067] S1, collecting multi-source state information of the power storage battery in real time, wherein the multi-source state information comprises voltage information, current information, temperature information, electrochemical impedance spectrum information and micro-vibration acceleration information;

[0068] S2, obtaining ohmic internal resistance, first charge transfer internal resistance and comprehensive thermal impedance index according to the multi-source state information;

[0069] The ohmic internal resistance and the first charge transfer internal resistance are obtained by the existing BMS, the ohmic internal resistance represents the ion / electron conduction impedance of the electrolyte and the electrode material, and the charge transfer internal resistance reflects the electrode interface electrochemical reaction dynamics resistance;

[0070] S3, obtaining a first thermal vibration frequency correction value according to the temperature information and the micro-vibration acceleration information;

[0071] S4, correcting the first charge transfer internal resistance based on the first thermal vibration frequency correction value to obtain a second charge transfer internal resistance;

[0072] S5, vectorize the ohmic internal resistance, the second charge transfer internal resistance, the first thermal vibration frequency correction value and the comprehensive thermal impedance index respectively to obtain a multi-source feature vector;

[0073] S6, construct a dynamic state of health comprehensive evaluation model based on a neural network, and input the multi-source feature vector into the dynamic state of health comprehensive evaluation model to dynamically predict the state of health of the battery, and output a prediction result;

[0074] S7, execute a battery charging and discharging instruction corresponding to the prediction result.

[0075] As described in steps S1-S7 above, the application can more accurately assess the battery health status. During the use of the power storage battery, accurate monitoring of the battery health status is crucial for ensuring the safe operation of the equipment, optimizing the battery performance, and predicting the battery life. Traditional power storage battery health status monitoring mainly relies on voltage, current and other conventional parameters. Only relying on voltage and current cannot comprehensively and accurately reflect the complex changes inside the battery, such as electrode interface aging, electrolyte drying, and SEI film thickening. These problems cannot be detected through simple voltage and current measurement, which leads to inaccurate judgment of the battery health status and may affect the normal use of the battery and the safety of the equipment. The current common monitoring method fails to consider the correlation between the physical and chemical changes inside the battery and various parameters. The application collects multi-source information to more comprehensively obtain the state data of the battery in different aspects, providing a rich data basis for subsequent accurate assessment of the battery health status. The application overcomes the problem of insufficient information caused by relying only on voltage and current parameters. By obtaining the ohmic internal resistance, the first charge transfer internal resistance and the comprehensive thermal impedance index, the application can more deeply understand the physical and chemical changes inside the battery and reflect the health status of the battery from multiple angles. Compared with the traditional method, the application can more accurately grasp the internal condition of the battery. Furthermore, the application uses the first thermal vibration frequency correction value to evaluate the battery aging state, which can more accurately and comprehensively reflect the complex changes inside the battery. The method is a non-invasive means that can monitor the battery micro-vibration frequency change in real time and obtain the battery aging state information in a timely manner. Compared with the traditional method, the application can detect the signs of battery aging at an earlier stage. The method provides battery internal state information from the perspective of mechanical vibration. When electrical measurement is disturbed by noise, it can help determine the reliability of the measurement results and adjust the calculation value to improve the accuracy of charge transfer internal resistance measurement, so that the battery management system can more accurately control the battery. The application can capture long-term dependencies in time series data through long short-term memory neural networks, learn the evolution pattern of the battery state over time, and predict the battery health status at future time. According to the battery health status prediction result, the application controls the charging and discharging operation of the battery, realizes the reasonable control and management of the battery, avoids overcharging or undercharging during the charging process, ensures the battery works within a safe range during the discharging process, and improves the overall performance and safety of the battery system.

[0076] In one embodiment, the step S3 of obtaining the first thermal vibration frequency correction value according to the temperature information and the micro-vibration acceleration information comprises:

[0077] S31, extracting the main frequency and the secondary frequency band energy proportion according to the frequency spectrum corresponding to the micro-vibration acceleration information;

[0078] S32, obtaining the cell surface temperature of the power storage battery and the cell surface preset standard temperature according to the temperature information;

[0079] S33, obtaining a thermal expansion coupling coefficient;

[0080] S34, obtaining a temperature interference coefficient according to the cell surface temperature, the cell surface preset standard temperature and the thermal expansion coupling coefficient;

[0081] S35, obtaining a first thermal vibration frequency correction value according to the temperature interference coefficient and the main frequency, and the calculation formula is:

[0082]

[0083] In the formula, F v represents the first thermal vibration frequency correction value, F1 represents the main frequency, γ represents the micro-vibration sensitive coefficient, represents the temperature interference coefficient, η represents the thermal expansion coupling coefficient (used to quantify the stress change of the SEI film caused by thermal expansion, calibrated by measuring the stress of the SEI film at different temperatures. The higher the thermal expansion coupling coefficient, the greater the influence of temperature change on the thermal vibration frequency correction value, and vice versa. The smaller the influence), T1 represents the cell surface temperature, and T0 represents the cell surface preset standard temperature;

[0084] S36, obtaining a micro-vibration sensitive coefficient according to the sub-band energy proportion;

[0085] S37, judging whether the micro-vibration sensitive coefficient is greater than a preset value, if the micro-vibration sensitive coefficient is greater than the preset value, adjusting the first thermal vibration frequency correction value according to the micro-vibration sensitive coefficient.

[0086] As described in steps S31-S37 above, during the charging and discharging process of the battery, a series of chemical reactions and physical changes occur inside the battery, including the expansion and contraction of electrode materials, the diffusion of electrolyte, and these changes will cause the micro-vibration of the internal structure of the battery. When the battery ages, its internal structure and chemical composition will change, causing the SEI (Solid Electrolyte Interface membrane) film to thicken and the electrolyte to dry, which will affect the frequency and amplitude characteristics of the micro-vibration. In the prior art, if the battery is detected whether it is aging, the battery is usually disassembled or a complex electrochemical test method is used. Disassembling the battery can easily damage the battery, and the complex electrochemical test method increases the detection cost. Therefore, the embodiment proposes a non-intrusive detection method. First, a micro-vibration sensor is installed on the surface of the battery. The micro-vibration sensor is a MEMS sensor, which functions to collect the micro-vibration acceleration information of the battery. By setting the micro-vibration sensor, the micro-vibration acceleration changes inside the battery can be detected, so that the micro-vibration main frequency and the secondary frequency band energy ratio of the battery during operation can be monitored in real time. The present application analyzes the frequency spectrum of the micro-vibration acceleration information to extract the main frequency and the secondary frequency band energy ratio by fast Fourier transform. The main frequency is the frequency component with the most concentrated energy and the largest amplitude during the operation of the battery, which reflects the main frequency characteristics caused by the aging change of the internal structure and chemical composition of the power storage battery. Since the micro-vibration acceleration information changes with temperature, when the battery temperature rises, the main frequency of the micro-vibration is affected by the thermal softening effect and shifts to low frequency. The embodiment obtains a temperature interference coefficient and a first thermal vibration frequency correction value according to the temperature interference coefficient and the main frequency. By obtaining the first thermal vibration frequency correction value, the thermal shift and the true aging can be distinguished. In order to more comprehensively and accurately establish the relationship between the first thermal vibration frequency correction value and the battery health status, after obtaining the first thermal vibration frequency correction value, the present application also obtains a micro-vibration sensitivity coefficient according to the secondary frequency band energy ratio, and judges whether the micro-vibration sensitivity coefficient is greater than a preset value. If it is greater than the preset value, it means that the influence of the internal microscopic defects of the battery on the vibration is more significant. At this time, the first thermal vibration frequency correction value is adjusted according to the micro-vibration sensitivity coefficient, so that the adjusted first thermal vibration frequency correction value can be more accurately used for battery health status evaluation.

[0087] More preferably, the step S37 of obtaining the micro-vibration sensitivity coefficient according to the secondary frequency band energy ratio comprises:

[0088] S371, obtaining the secondary frequency band vibration energy and the total vibration energy;

[0089] S372, obtaining a micro-vibration sensitivity coefficient reference value;

[0090] S373, obtaining a micro-vibration sensitive coefficient according to the sub-band vibration energy, the total vibration energy and the micro-vibration sensitive coefficient reference value, and the calculation formula is:

[0091]

[0092] In the formula, γ represents the micro-vibration sensitive coefficient, γ0 represents the micro-vibration sensitive coefficient reference value, represents the sub-band energy ratio, E sub represents the sub-band vibration energy, E total represents the total vibration energy.

[0093] As described in steps S371-S373, the sub-band energy ratio shows the abnormal vibration energy ratio, including the response of harmonics, noise and secondary physical processes, such as high-frequency acoustic emission signals of micro-level SEI film rupture, low-frequency disturbance of environmental temperature fluctuation. If the sub-band energy ratio is high, it reflects that the abnormal disturbance of the secondary process is significant. Compared with the main frequency, it reflects the aging change of the internal structure and chemical composition of the power storage battery, and the sub-band energy ratio can better reflect the early micro-defects. For example, when the main frequency is stable, but the sub-band energy ratio is abnormally high, the micro-vibration sensitive coefficient is increased by combining the sub-band energy ratio, which can make the whole system more sensitive to the abnormal disturbance of the secondary physical process, and thus enhance the system robustness.

[0094] In one embodiment, the step S4 of correcting the first charge transfer resistance according to the first thermal vibration frequency correction value to obtain the second charge transfer resistance includes:

[0095] S41, obtaining a first thermal vibration frequency correction value;

[0096] S42, obtaining a second thermal vibration frequency correction value of the battery in a primary use state;

[0097] S43, calculating a dynamic change rate of the first thermal vibration frequency correction value and the second thermal vibration frequency correction value;

[0098] S44, obtaining a battery packaging material stiffness correction coefficient;

[0099] S45, obtaining a thermal vibration frequency dynamic correction coefficient according to the dynamic change rate and the battery packaging material stiffness correction coefficient;

[0100] S46, adjusting the first charge transfer resistance based on the thermal vibration frequency dynamic correction coefficient to obtain the second charge transfer resistance, and the calculation formula is:

[0101]

[0102] In the formula, Rcorrected2 represents a corrected second charge transfer resistance, R ct R1 represents a first charge transfer resistance, a1 represents a material stiffness correction factor, F v F1 represents a first thermal vibration frequency correction value, F v，new F2 represents a second thermal vibration frequency correction value, a represents a thermal vibration frequency dynamic correction factor.

[0103] As described in steps S41-S46 above, the charge transfer resistance is a key parameter of the reaction electrode interface electrochemical reaction kinetics resistance, and its measurement accuracy directly affects the judgment of the reaction process inside the battery. With the charging and discharging cycles and environmental factors during the use of the battery, the charge transfer resistance will change, and it needs to be accurately corrected and monitored. In the traditional method, when measuring the charge transfer resistance, only the changes of voltage, current and other electrical parameters are considered, and the influence of the micro-vibration and temperature environment factors generated by the physical and chemical changes inside the battery on the resistance is ignored, which cannot fully capture the complex changes inside the battery, resulting in that the measurement result is difficult to accurately reflect the real resistance state of the battery, and further affecting the evaluation of the battery health state. There is a certain internal relationship between the micro-vibration acceleration information of the battery and the charge transfer resistance. With the increase of the aging degree of the battery, the micro-vibration acceleration information will change, and the change of the charge transfer resistance of the battery and the micro-vibration acceleration information shows a certain positive correlation, that is, the greater the resistance, the greater the change of the micro-vibration acceleration information. The micro-vibration acceleration information reflects the structure and mechanical state inside the battery, and the charge transfer process is closely related to the structure, surface state of the electrode material and the distribution of the electrolyte. When the battery ages and the active material on the electrode surface decreases, on the one hand, it will increase the charge transfer resistance, and on the other hand, it will also change the mechanical properties inside the battery, resulting in changes in the micro-vibration acceleration information. The present application can obtain the first thermal vibration frequency correction value reflecting the current thermal vibration state of the battery by collecting temperature information and micro-vibration acceleration information and analyzing the coupling relationship between the two. The first thermal vibration frequency correction value can reflect the physical and chemical changes inside the battery from the thermal-vibration angle, and provide a key parameter for more accurately correcting the charge transfer resistance in the future, which makes up for the shortcomings of the traditional method which only measures from the electrical angle. At the same time, the state of the battery when it is used for the first time can be used as a reference state, and the second thermal vibration frequency correction value obtained at this time represents the initial thermal-vibration characteristics of the battery. Comparing the first thermal vibration frequency correction value with the second thermal vibration frequency correction value during use can effectively reflect the changes of the internal structure and performance of the battery with the use time, and provide reference data for subsequent calculation of the dynamic change rate, so that the changes of the thermal vibration characteristics of the battery during use can be quantified, laying a foundation for accurately correcting the charge transfer resistance. According to the first thermal vibration frequency correction value and the second thermal vibration frequency correction value, the dynamic change rate can be calculated, which can reflect the change trend of the thermal vibration characteristics of the battery with time. By comparing the first thermal vibration frequency correction value with the second thermal vibration frequency correction value at different stages, the changes of the internal structure of the battery caused by chemical reactions, aging and other factors can be understood, and there is an inherent relationship between the changes of the charge transfer resistance and the changes of the internal structure of the battery. The dynamic change rate provides a key indicator reflecting the change trend of the battery internal changes for correcting the charge transfer resistance, making the correction process more in line with the actual use of the battery and improving the accuracy of the resistance correction. Since the stiffness of the battery packaging material will affect the stress distribution inside the battery,Further, the micro-vibration characteristics of the battery and the charge transfer resistance are affected, the rigidity of different packaging materials is different in the process of temperature change, charging and discharging cycle and the like, and the constraint and influence degree of the internal structure of the battery are different, the packaging material rigidity correction coefficient is introduced in the application, the influence of the packaging material on the internal parameters of the battery can be comprehensively considered, the correction of the charge transfer resistance is more comprehensive and accurate, the correction error caused by ignoring the packaging material factor is avoided, the dynamic change rate reflecting the change of the internal thermal vibration characteristics of the battery is combined with the packaging material rigidity correction coefficient, the influence of the physical and chemical changes in the battery and the packaging material on the battery can be comprehensively considered, the thermal vibration frequency dynamic correction coefficient is obtained, the thermal vibration frequency dynamic correction coefficient more comprehensively reflects the comprehensive effect of various factors on the thermal vibration characteristics and internal resistance of the battery, the thermal vibration frequency dynamic correction coefficient provides a comprehensive and more accurate correction factor for the final adjustment of the charge transfer resistance, the correction process is more scientific and reasonable, and the accuracy and reliability of the charge transfer resistance correction are improved, finally, according to the internal relationship between the thermal vibration characteristics of the battery and the charge transfer resistance, the first charge transfer resistance is adjusted through the thermal vibration frequency dynamic correction coefficient, compared with the traditional method, the charge transfer resistance is corrected from the comprehensive consideration of thermal-vibration characteristics and the influence of the packaging material in this step, the second charge transfer resistance can more accurately reflect the electrode interface electrochemical reaction kinetics resistance in the battery, and more reliable data support is provided for the battery health state evaluation, and the accuracy and reliability of the battery management system for judging the state of the battery are improved.

[0104] In one embodiment, the step of obtaining the ohmic resistance, the first charge transfer resistance and the comprehensive thermal impedance index according to the multi-source state information comprises:

[0105] S21, obtaining a battery thermal gradient value according to the multi-source state information;

[0106] S22, extracting a high-frequency impedance imaginary value according to the electrochemical impedance spectrum information; an integrated EIS measurement chip is used, an excitation source, a filter and a digital demodulation module are built-in, a 10kHz sine wave is generated, is superimposed to the battery charging and discharging current, the real part and the imaginary part are calculated by the internal DSP of the chip, and the high-frequency impedance imaginary value is collected;

[0107] S23, calculating a comprehensive thermal impedance index according to the battery thermal gradient value and the high-frequency impedance imaginary value, and the calculation formula is:

[0108] Z2=Z1(f1)*ΔT4;

[0109] In the formula, Z2 represents the comprehensive thermal impedance index, Z1(f1) represents the high-frequency impedance imaginary value at the frequency f1 (f1 can be 1Hz), and ΔT4 represents the battery thermal gradient value.

[0110] As described in steps S21-S23, during the charge-discharge cycle of the battery, the structure evolution of the internal electrode material, the loss of the active material, the decomposition and drying of the electrolyte, these changes will cause the change of the ion transport path in the battery, the charge transfer process is blocked, and then the change of the battery impedance, at the same time, the chemical reaction in the battery is mostly exothermic reaction, the aggravation or abnormality of the reaction in the battery aging process will cause the increase of heat production, so that the temperature of the battery changes, and the change of temperature will in turn affect the chemical reaction rate and the ion diffusion coefficient in the battery, further affect the battery impedance, the temperature rise will accelerate the diffusion speed of the ions in the electrolyte, to a certain extent, reduce the battery impedance, but with the battery aging, the thickening of the passivation film on the electrode surface and other factors will make the impedance rise, forming a coupling relationship, with the deepening of the battery aging degree, the physical and chemical changes in the battery are intensified, leading to the coupling relationship between impedance and temperature deviating from the initial state, in the new battery, the impedance changes with the temperature relatively small and regular, but with the battery aging, the change range of the impedance at the same temperature change range increases, and the change trend becomes more complex, the high-frequency impedance imaginary part value is sensitive to the electrode-electrolyte interface state, the traditional battery parameter monitoring method often installs temperature sensors at a specific position, isolatedly measures the temperature, and does not link the temperature gradient change with the battery internal electrochemical process, due to the difference in heat production and heat dissipation conditions of different positions in the battery during the charge-discharge process, a temperature difference is generated, a thermal gradient is formed, through analysis and processing of the multi-source temperature related information, the thermal gradient can be quantified, which reflects the non-uniformity of the thermal distribution in the battery, and the non-uniformity is closely related to the physical and chemical processes in the battery, the battery thermal gradient value is obtained according to the multi-source state information, which provides a key thermal parameter for subsequent comprehensive evaluation of the battery thermal-electric characteristics, breaks the limitation of traditional isolated temperature measurement, makes the understanding of the battery thermal characteristics more in-depth and comprehensive, and helps to more accurately grasp the state change in the battery, since the high-frequency impedance imaginary part value is related to the ion diffusion and charge transfer processes in the battery, by extracting the value, the related information of the electrochemical kinetics characteristics in the battery can be obtained, and since the thermal process and the electrochemical reaction process in the battery interact with each other, the thermal gradient will change the ion diffusion rate and the electrode reaction activity, and then affect the electrochemical impedance, and the heat generated by the electrochemical process will also affect the temperature distribution, by multiplying the battery thermal gradient value reflecting the thermal characteristics and the high-frequency impedance imaginary part value reflecting the electrical characteristics, the thermal-electric coupling relationship can be quantified, and a comprehensive thermal impedance index is obtained, the comprehensive thermal impedance index comprehensively considers the interaction of the battery heat and electricity, the change of the impedance-temperature coupling parameter can reflect the SEI film thickening and the microscopic degradation phenomenon of lithium dendrite growth in real time, dynamically links the electrochemical degradation and the thermal gradient parameter, breaks the limitation of single physical field calculation, can effectively infer the aging state of the battery, and adopts the impedance-temperature coupling parameter to evaluate the battery aging state,The evaluation method can better adapt to different temperature conditions, avoid evaluation errors caused by changes in environmental temperature, and more comprehensively and accurately reflect the complex physical and chemical changes inside the battery compared to traditional methods of isolated measurement of electrical and thermal parameters, thereby providing more effective parameter indicators for battery state of health evaluation and performance optimization.

[0111] In one embodiment, the step of obtaining a battery thermal gradient value according to the multi-source state information comprises:

[0112] S211, obtaining a battery charge-discharge rate, a battery environment convection coefficient, a battery column-level temperature, a battery surface temperature, and a battery ambient temperature according to the multi-source state information;

[0113] S212, obtaining an internal thermal resistance influence factor based on the battery charge-discharge rate; as the battery charge-discharge rate increases, the internal thermal resistance will increase, and the importance of the internal thermal resistance influence factor for evaluating the battery state of health can be reduced, thereby reducing the influence of changes in the internal thermal resistance caused by non-aging conditions;

[0114] S213, obtaining an external heat dissipation influence factor based on the battery environment convection coefficient; as the device moving speed increases, the air flow rate increases, the battery environment convection coefficient increases, the heat dissipation efficiency increases, the external heat dissipation influence factor can be increased, the influence of the increase in the external heat dissipation efficiency caused by non-aging conditions is reduced, and the battery state of health monitoring is more accurate;

[0115] S214, obtaining a battery thermal gradient value according to the battery column-level temperature, the battery surface temperature, the battery ambient temperature, the internal thermal resistance influence factor, and the external heat dissipation influence factor, and the calculation formula is:

[0116] ΔT4=[α2*(T2-T1)+α3*(T1-T3)];

[0117] In the formula, ΔT4 represents the battery thermal gradient value, T2 represents the battery column-level temperature, T1 represents the battery surface temperature, T3 represents the battery ambient temperature, α2 represents the internal thermal resistance influence factor, and α3 represents the external heat dissipation influence factor.

[0118] As described in steps S211-S214, the conventional battery thermal state monitoring method mainly relies on single-point temperature measurement, only measuring the battery pole temperature, which has limitations. On the one hand, the change of the internal thermal resistance of the battery is ignored, such as the heat accumulation caused by the thickening of the solid electrolyte interface film (SEI film), which cannot fully reflect the real thermal conditions inside the battery. On the other hand, single-point temperature measurement is extremely sensitive to boundary conditions such as environmental temperature, and when the environmental temperature changes suddenly, it is easy to misjudge the battery thermal state, thereby affecting the accurate assessment of the battery health state. During the battery aging process, the SEI film will gradually thicken, which will increase the internal thermal resistance of the battery, causing heat to accumulate inside the battery, and the difference between the battery pole temperature and the battery surface temperature will increase. Since the battery charge-discharge rate determines the rate of chemical reaction inside the battery, which in turn affects the heat generation rate, the battery environmental convection coefficient reflects the heat exchange ability between the battery and the external environment, the battery pole temperature, the battery surface temperature and the battery environmental temperature respectively reflect the temperature conditions inside the battery, on the surface and in the environment, these parameters are interrelated and jointly affect the thermal state of the battery. Therefore, the present application obtains the battery charge-discharge rate, the battery environmental convection coefficient, the battery pole temperature, the battery surface temperature and the battery environmental temperature according to the multi-source state information, providing a rich and key data basis for subsequent accurate calculation of the battery thermal gradient value, changing the traditional limitation of only focusing on single temperature data, making the understanding of the battery thermal state more comprehensive and in-depth. Moreover, as the battery charge-discharge rate increases, the chemical reaction inside the battery intensifies, heat generation increases, and the internal thermal resistance also increases accordingly. When evaluating the battery health state, in order to distinguish between the internal thermal resistance changes caused by normal charge-discharge rate changes and the thermal resistance changes caused by battery aging and other reasons, by setting an internal thermal resistance influence factor and adjusting it according to the charge-discharge rate, the interference of internal thermal resistance changes caused by non-aging conditions on the battery health state evaluation can be reduced. The introduction of the internal thermal resistance influence factor enables more accurate separation of the thermal characteristic changes related to battery aging when evaluating the battery health state, improving the accuracy of battery health state judgment and avoiding misjudgment caused by charge-discharge rate fluctuations. Moreover, since the battery environmental convection coefficient is closely related to the external heat dissipation efficiency, such as the increase of device moving speed and air flow rate, the battery environmental convection coefficient increases, and the heat dissipation efficiency increases. When evaluating the battery health state, the influence of environmental factors on the heat dissipation efficiency needs to be excluded. By setting an external heat dissipation influence factor and adjusting it according to the environmental convection coefficient, the battery health state monitoring can be more accurate. The setting of the external heat dissipation influence factor effectively eliminates the interference of normal fluctuations of environmental factors on the battery thermal state evaluation, improves the accuracy and reliability of the battery health state monitoring, and makes the monitoring results more truly reflect the health status of the battery itself. By considering the internal, surface and environmental temperatures of the battery, as well as the influence of internal thermal resistance and external heat dissipation,The method can more accurately quantify the temperature difference inside the battery, i.e., the thermal gradient value, and fully considers various factors affecting the thermal state of the battery, more comprehensively and accurately reflects the temperature distribution inside the battery compared with the traditional method, and provides a more reliable basis for battery health state evaluation, thermal management strategy formulation, etc., and helps to improve the performance and service life of the battery.

[0119] In one embodiment, the step of inputting the multi-source feature vector into the power storage battery health state comprehensive evaluation model to dynamically predict the battery health state and output a prediction result comprises:

[0120] S61, obtain the multi-source feature vector, and the formula is:

[0121]

[0122] In the formula, represents a transposition operation, converting a row vector into a column vector, X t represents the multi-source feature vector at time t, R Ω respectively represent the ohmic internal resistance, R ct represents the second charge transfer internal resistance, Z2 represents the comprehensive thermal impedance index, F v represents the thermal vibration frequency correction value;

[0123] S62, input the multi-source feature vector into a long short-term memory neural network, process the multi-source feature vector based on the long short-term memory neural network, and output a battery health state prediction value. The long short-term memory neural network learns the internal relationship between the multi-source feature vector and the battery health state prediction value to realize output of the battery health state prediction value, and the formula is:

[0124] Y t = LSTM(Y t-1 , X t ) + w t ;

[0125] In the formula, Y t represents the battery health state prediction value at time t, which at least includes SOH t , battery capacity, LSTM represents a long short-term memory neural network, Y t-1 represents the battery health state prediction value at time t-1, X t represents the multi-source feature vector at time t, and w trepresents the process noise at time t; the process noise specifically includes complex chemical reactions inside the battery that are difficult to predict, variable environmental factors, and limitations of the model, the battery state of health prediction value obtained according to the battery historical operation data is compared with the corresponding experimentally determined true value, and based on the variance of the error of each prediction, w t ;

[0126] S63, obtain an interference compensation factor (including environmental interference factors such as temperature, humidity, etc.) at the target time, and adjust the interference compensation factor based on the interference compensation factor at the last time of the target time;

[0127] S64, obtain an evaluation noise coefficient (including measurement device error and data transmission loss factors, the value is determined by experiment);

[0128] S65, obtain a battery state of health evaluation value according to the interference compensation factor and the battery state of health prediction value, and the formula is:

[0129] C t =h(Y t )+μ t ;

[0130] In the formula, C t represents the battery state of health evaluation value at time t, h(Y t ) represents a calculation function of the battery state of health prediction value combined with the interference compensation factor, and μ t represents the evaluation noise coefficient;

[0131] Taking SOH t (Y t =SOH t ) as the battery state of health prediction value and temperature (σ t =1-β*|T4-T5|) as the interference compensation factor, the obtaining step of h(Y t ) includes:

[0132] S651, obtain a battery average temperature (an average of the battery column level temperature, the battery surface temperature, and the battery environment temperature) and a battery standard average temperature (an average of the battery column level temperature, the battery surface temperature, and the battery environment temperature under standard conditions) according to the obtained temperature information;

[0133] S652, obtain a temperature influence empirical coefficient (when the temperature changes, the battery state of health prediction value changes accordingly, and the specific value is determined according to multiple battery experiment data fitting);

[0134] S653, obtain a battery state of health prediction value according to the battery average temperature, the battery standard average temperature, and the temperature influence empirical coefficient, and the expanded formula is:

[0135] h(Y t )=SOH t *(1-β*|T5-T6|);

[0136] In the formula, SOH t T5 represents the predicted battery health status at time t, T6 represents the battery average temperature, and β represents the empirical coefficient of temperature influence.

[0137] S66. Obtain battery health status assessment values ​​at multiple different times, and predict battery health status based on the rate of change of the battery health status assessment values ​​at multiple different times; specifically, collect battery health status assessment values ​​at preset time intervals, calculate adjacent battery health status assessment values ​​sequentially, plot the rate of change over time with time as the horizontal axis and the rate of change as the vertical axis, calculate the mean and variance of the rate of change, and judge the trend of the rate of change based on the rate of change curve and statistics. If the mean is negative and the variance is small, it indicates that the battery health status assessment value is decreasing relatively stably; if the variance is large, it indicates that the battery health status assessment value fluctuates drastically. Based on the linear relationship between the battery health status assessment value and time, a linear regression model is used to predict the battery health status.

[0138] As described in steps S61-S66 above, during the charging and discharging process of the battery, its internal state presents a complex dynamic change with time, which has long-term dependence and nonlinear characteristics. The traditional method usually judges the health state according to the parameters such as voltage, current, temperature, etc. of the battery. This way ignores the complex physical and chemical processes inside the battery and cannot adapt to changes in different use environments, charging and discharging modes, etc. In complex working conditions, such as frequent start-stop and extreme temperature, the prediction result has large deviation, cannot timely and accurately predict potential faults of the battery, and affects the normal operation and safety of the equipment. Since the battery is in operation, these different types of parameters reflect the internal physical and chemical changes from multiple aspects. Therefore, the present application integrates them into a multi-source feature vector, which can more comprehensively represent the battery state and provide rich and accurate input for the subsequent model. Compared with the traditional single or small amount of parameter input, the multi-source feature vector contains more comprehensive information, laying a foundation for accurate prediction of the battery health state, so that the model can capture more factors affecting the battery health. In order to capture the long-term dependence in the time series data and learn the evolution pattern of the battery state over time, the present application inputs the multi-source feature vector into a long short-term memory neural network to output the battery health state prediction value. At the same time, since the battery is disturbed by various external and internal factors during the actual aging process, these factors will affect the battery health state evaluation. Therefore, the present application quantifies the influence of interference by establishing an interference compensation factor and adjusting it according to historical information, which can more accurately reflect the change of interference. The introduction and adjustment of the interference compensation factor enable the model to effectively cope with interference factors in actual operation, improve the accuracy and reliability of the battery health state evaluation, and avoid misjudgment caused by interference. After quantifying the interference factors, the prediction value is combined to consider the influence of actual interference on the battery health state, and the prediction value is corrected to obtain an evaluation value that is more in line with the actual situation. Moreover, since the battery health state changes with time, the present application can understand the trend of the battery health state change by analyzing the change rate of the battery health state evaluation value at different times, and predict the future health state in combination with the current evaluation value. From the time dimension, the change of the battery health state can be analyzed to discover the deterioration trend of the battery health state in advance.

[0139] In one embodiment, the step of adjusting the interference compensation factor of the target time based on the interference compensation factor of the previous time of the target time comprises:

[0140] S631, calculating a first covariance according to the interference compensation factors of the previous two times of the target time;

[0141] S632, calculating a second covariance according to the interference compensation factors of the target time and the previous time;

[0142] S633, calculating an error covariance according to the first covariance and the second covariance;

[0143] S634, repeat the steps of calculating the first covariance according to the interference compensation factor of the target time and the two previous times, and calculating the error covariance according to the first covariance and the second covariance, according to the preset number of times, to obtain a plurality of error covariances, and dynamically adjust the interference compensation factor of the target time according to the error covariance. The dynamic adjustment method is: when the error covariance is greater, the interference compensation factor of the target time is closer to the interference compensation factor of the previous time; when the error covariance is smaller, the interference compensation factor of the target time is closer to itself.

[0144] As described above in steps S631-S634, in the actual battery monitoring process, the environment may be constantly changing, and environmental interference factors will affect the results of battery monitoring. The first covariance can measure the overall error of the two variables of the interference compensation factor of the target time and the two previous times, and reflect the change relationship and fluctuation degree between the two times. The second covariance can measure the change relationship and fluctuation degree between the two variables of the interference compensation factor of the target time and the previous time. In order to avoid the abnormal impact of environmental factor mutation on the results of battery monitoring, the error covariance combines the change information of the interference compensation factor at different time intervals in the previous two steps, can quantify the error degree of the change of the interference compensation factor at different times, reflects the uncertainty and fluctuation range of the change of the interference compensation factor, and provides a basis for judging the change stability of the interference compensation factor, and further provides a key reference for reasonably adjusting the interference compensation factor of the target time, so that the adjusted interference compensation factor is more in line with the actual interference. Through repeated calculation, the change information of the interference compensation factor on multiple time segments can be obtained, and the overall change trend and fluctuation rule of the interference compensation factor can be comprehensively and accurately grasped by comprehensively considering multiple error covariances. According to the adjustment principle that "the greater the error covariance, the closer the interference compensation factor of the target time to the interference compensation factor of the previous time; the smaller the error covariance, the closer the interference compensation factor of the target time to itself", because the greater error covariance means that the change of the interference compensation factor is unstable, and it is more reasonable to adjust the value of the previous time which is relatively stable; when the error covariance is small, the current interference compensation factor is relatively stable, and the value can be maintained. This dynamic adjustment method can adaptively adjust according to the actual change of the interference compensation factor, fully considers the complexity and dynamics of the interference factor, and can more timely and accurately correct the interference compensation factor compared with the traditional fixed or simple linear adjustment method, thereby effectively improving the accuracy and reliability of the state of health evaluation of the power storage battery.

[0145] As shown in Figure 2 , the present application also discloses a power storage battery state of health monitoring system, comprising:

[0146] The data acquisition module 1 acquires multi-source state information of the power storage battery in real time, wherein the multi-source state information includes voltage information, current information, temperature information, electrochemical impedance spectrum information and micro-vibration acceleration information;

[0147] The acquisition module 2 is configured to acquire an ohmic internal resistance, a first charge transfer internal resistance and a comprehensive thermal impedance index according to the multi-source state information;

[0148] The acquisition module 2 is configured to acquire a first thermal vibration frequency correction value according to the temperature information and the micro-vibration acceleration information;

[0149] The correction module 3 corrects the first charge transfer internal resistance based on the first thermal vibration frequency correction value to obtain a second charge transfer internal resistance;

[0150] The multi-source fusion module 4 is configured to vectorize the ohmic internal resistance, the second charge transfer internal resistance, the first thermal vibration frequency correction value and the comprehensive thermal impedance index respectively to obtain a multi-source feature vector;

[0151] The calculation module 5 is configured to construct a power storage battery health state comprehensive evaluation model based on a neural network, and input the multi-source feature vector into the power storage battery health state comprehensive evaluation model to dynamically predict the battery health state and output a prediction result;

[0152] The control execution module 6 is configured to execute a battery charging and discharging instruction corresponding to the prediction result.

[0153] The control execution module includes:

[0154] The first level response unit (SOH≤80%): limit the discharge current to 0.5C and activate the equalization charging;

[0155] The second level response unit (SOH≤70%): cut off the main relay and report a fault code;

[0156] The communication unit: supports 5G and CAN bus dual channels, and preferentially transmits emergency instructions through the CAN bus;

[0157] The control execution module further includes:

[0158] The low-temperature heating unit is configured to activate the heating film and adjust the charging parameters in a low-temperature environment.

[0159] If the battery temperature is below 5°, the heating film is activated.

[0160] It is to be understood that the terminology "including", "comprising", or any other variation thereof, is intended to cover a non-exclusive inclusion such that process, method, article, or apparatus that comprises a list of elements does not include only those elements but can also include other elements not expressly listed or inherent to such process, method, article, or apparatus. An element proceeded by "comprises a... " does not, without more constraints, exclude the presence of additional identical elements in the process, method, article, or apparatus that comprises the element.

[0161] The preferred embodiments of the present application have been described above with the specific language and / or illustrative examples, but it should be understood that the patent protection is not limited to the specific embodiments and / or examples described. Any alterations and / or further modifications in the described embodiments and / or examples, made by any one of ordinary skill in the art, having the benefit of this disclosure, are to be considered within the scope of the patent protection.

Claims

1. A method for monitoring the health status of a power battery, characterized in that, include: Real-time acquisition of multi-source state information of power batteries, including voltage information, current information, temperature information, electrochemical impedance spectroscopy information and micro-vibration acceleration information; The ohmic internal resistance, the first charge transfer internal resistance, and the comprehensive thermal impedance index are obtained based on the multi-source state information. The first thermally induced vibration frequency correction value is obtained based on the temperature information and micro-vibration acceleration information; The first charge transfer internal resistance is corrected based on the first thermally induced vibration frequency correction value to obtain the second charge transfer internal resistance. The ohmic internal resistance, the second charge transfer internal resistance, the first thermally induced vibration frequency correction value, and the comprehensive thermal impedance index are vectorized respectively to obtain a multi-source feature vector. A comprehensive evaluation model for the health status of a power battery based on a neural network is constructed, and the multi-source feature vector is input into the comprehensive evaluation model for the health status of the power battery to dynamically predict the battery health status and output the prediction results. Execute the battery charge / discharge command corresponding to the predicted result.

2. The method for monitoring the health status of a power battery according to claim 1, characterized in that, The step of obtaining the first thermally induced vibration frequency correction value based on the temperature information and micro-vibration acceleration information includes: The energy proportions of the main frequency and secondary frequency bands are extracted based on the spectrum corresponding to the micro-vibration acceleration information. The surface temperature of the battery cell and the preset standard temperature of the battery cell surface are obtained based on the temperature information. Obtain the thermal expansion coupling coefficient; The temperature interference coefficient is obtained based on the cell surface temperature, the preset standard temperature of the cell surface, and the thermal expansion coupling coefficient. The first thermally induced vibration frequency correction value is obtained based on the temperature interference coefficient and the dominant frequency; The micro-vibration sensitivity coefficient is obtained based on the energy ratio of the sub-frequency band. Determine whether the micro-vibration sensitivity coefficient is greater than a preset value. If the micro-vibration sensitivity coefficient is greater than the preset value, adjust the first thermally induced vibration frequency correction value according to the micro-vibration sensitivity coefficient.

3. The method for monitoring the health status of a power battery according to claim 1, characterized in that, The step of correcting the first charge transfer internal resistance based on the first thermally induced vibration frequency correction value to obtain the second charge transfer internal resistance includes: Obtain the second thermally induced vibration frequency correction value of the battery under initial use conditions; Calculate the dynamic rate of change between the first thermally induced vibration frequency correction value and the second thermally induced vibration frequency correction value; Obtain the stiffness correction coefficient for battery packaging materials; The dynamic correction coefficient for thermally induced vibration frequency is obtained based on the dynamic rate of change and the stiffness correction coefficient of the battery packaging material. The first charge transfer resistance is adjusted based on the dynamic correction coefficient of the thermally induced vibration frequency to obtain the second charge transfer resistance.

4. The method for monitoring the health status of a power battery according to claim 1, characterized in that, The steps of obtaining the ohmic internal resistance, the first charge transfer internal resistance, and the comprehensive thermal impedance index based on the multi-source state information include: The battery thermal gradient value is obtained based on the multi-source state information; Extract the high-frequency impedance imaginary part value based on the electrochemical impedance spectroscopy information; The comprehensive thermal impedance index is calculated based on the battery thermal gradient value and the imaginary part of the high-frequency impedance.

5. The method for monitoring the health status of a power battery according to claim 4, characterized in that, The step of obtaining the battery thermal gradient value based on the multi-source state information includes: Based on the multi-source state information, the battery charge / discharge rate, battery environmental convection coefficient, battery column temperature, battery surface temperature, and battery ambient temperature are obtained. The internal thermal resistance influence factor is obtained based on the battery charge / discharge rate. External heat dissipation influencing factors are obtained based on the battery environment convection coefficient; The battery thermal gradient value is obtained based on the battery column temperature, the battery surface temperature, the battery ambient temperature, the internal thermal resistance influence factor, and the external heat dissipation influence factor.

6. The method for monitoring the health status of a power battery according to claim 1, characterized in that, The step of inputting the multi-source feature vector into the comprehensive health status assessment model of the power battery to dynamically predict the battery health status and output the prediction result includes: Obtain the multi-source feature vector; The multi-source feature vectors are input into a long short-term memory neural network, and the multi-source feature vectors are processed based on the long short-term memory neural network to output a predicted value of battery health status. Obtain the interference compensation factor at the target time, and adjust the interference compensation factor at the target time based on the interference compensation factor at the previous time. The battery health status assessment value is obtained based on the interference compensation factor and the predicted battery health status value. Obtain battery health status assessment values ​​at multiple different times, predict battery health status based on the rate of change of the battery health status assessment values ​​at multiple different times, and output the prediction results.

7. The method for monitoring the health status of a power battery according to claim 6, characterized in that, The step of adjusting the interference compensation factor at the target time based on the interference compensation factor of the previous time step includes: The first covariance is calculated based on the interference compensation factors of the two time points preceding the target time. The second covariance is calculated based on the interference compensation factor between the target time and the previous time. Calculate the error covariance based on the first covariance and the second covariance; Repeat the steps of calculating the first covariance based on the interference compensation factor of the two moments before the target time to calculating the error covariance based on the first covariance and the second covariance a preset number of times to obtain multiple error covariances, and dynamically adjust the interference compensation factor of the target time based on the error covariances.

8. A power battery health status monitoring system, characterized in that, include: The data acquisition module collects multi-source state information of the power battery in real time, including voltage information, current information, temperature information, electrochemical impedance spectroscopy information, and micro-vibration acceleration information. The acquisition module is used to acquire the ohmic internal resistance, the first charge transfer internal resistance, and the comprehensive thermal impedance index based on the multi-source state information. Used to obtain a first thermally induced vibration frequency correction value based on the temperature information and micro-vibration acceleration information; The correction module corrects the first charge transfer internal resistance based on the first thermally induced vibration frequency correction value to obtain the second charge transfer internal resistance; The multi-source fusion module is used to vectorize the ohmic internal resistance, the second charge transfer internal resistance, the first thermally induced vibration frequency correction value and the comprehensive thermal impedance index respectively to obtain a multi-source feature vector. The calculation module is used to construct a comprehensive evaluation model of the health status of a power battery based on a neural network, and input the multi-source feature vector into the comprehensive evaluation model of the health status of the power battery to dynamically predict the battery health status and output the prediction results. The control execution module is used to execute the battery charging and discharging commands corresponding to the prediction results.

9. A power battery health status monitoring system according to claim 8, characterized in that, The control execution module includes: The first-stage response unit is used to limit the discharge current and activate equalization charging. The second-level response unit is used to disconnect the equipment from the power battery and report the fault code. The communication unit is used to transmit emergency instructions.

10. A power battery health status monitoring system according to claim 8, characterized in that, The control execution module further includes: The low-temperature heating unit is used to activate the heating film and adjust the charging parameters in low-temperature environments.

Images