Battery control method and device, electronic equipment, storage medium and program product

Abstract

The embodiment of the invention provides a battery control method and device, electronic equipment, a storage medium and a program product. The method comprises the following steps: collecting operation state data of a battery, wherein the operation state data comprises current data, temperature data and battery internal resistance data; and according to the current data and the temperature data, determining a battery control voltage, inputting the current data, the temperature data, the battery internal resistance data and the battery control voltage to a pre-trained neural network model, and outputting a state of charge and a state of health of the battery. Determining the power transmission efficiency of the battery according to the internal resistance data of the battery, and generating a power adjustment strategy according to the power transmission efficiency, the power adjustment strategy being used for adjusting the power consumption of the battery. According to the method, the monitoring precision of the state of charge and the state of health is improved, and the battery power consumption control effect is improved.

Description

TECHNICAL FIELD

[0001] The present application relates to the technical field of battery, and particularly relates to a battery control method and device, electronic equipment, storage medium and program product. BACKGROUND

[0002] Battery technology, as the core support of new energy industry, is widely used in many fields. The running stability of the battery system is directly related to the development quality and safety level of the related industry.

[0003] In the related art, the state of charge (SOC) is estimated based on current integration by coulomb counting, and the state of health (SOH) is determined in combination with the cycle number, and the power adjustment strategy is mostly adjusted based on the voltage threshold.

[0004] However, the SOC, SOH and power consumption control strategy obtained by the above method are not accurate enough. SUMMARY

[0005] The battery control method and device, electronic equipment, storage medium and program product provided by the embodiments of the present application can achieve the technical effect of improving the accuracy of the SOC, SOH and power consumption control strategy.

[0006] In a first aspect, the embodiments of the present application provide a battery control method, comprising:

[0007] Collecting running state data of the battery, the running state data comprising current data, temperature data and battery internal resistance data;

[0008] Determining a battery control voltage according to the current data and the temperature data;

[0009] Inputting the current data, the temperature data, the battery internal resistance data and the battery control voltage into a pre-trained neural network model to output the state of charge and the state of health of the battery;

[0010] Determining the power transmission efficiency of the battery according to the battery internal resistance data;

[0011] Generating a power adjustment strategy according to the power transmission efficiency, the power adjustment strategy being used to adjust the power consumption of the battery.

[0012] In a possible implementation, the battery control voltage is determined according to the current data and the temperature data, comprising:

[0013] Obtaining a prediction analysis factor and prediction accuracy data obtained based on the pre-trained neural network model;

[0014] Obtaining a preset penalty function;

[0015] The prediction analysis factor, the prediction accuracy data, the penalty function, the current data and the temperature data are input into a preset voltage control algorithm to determine the battery control voltage.

[0016] In a possible implementation,

[0017] According to the battery internal resistance data, the power transmission efficiency of the battery is determined, including:

[0018] The loop resistance of the battery, the preset initial efficiency and the current density integral data are obtained;

[0019] The loop resistance, the preset initial efficiency, the current density integral data and the battery internal resistance data are input into a preset efficiency algorithm to determine the power transmission efficiency of the battery.

[0020] In a possible implementation,

[0021] According to the power transmission efficiency, a power adjustment strategy is generated, including:

[0022] If the power transmission efficiency is greater than or equal to a preset efficiency threshold, a power adjustment strategy for maintaining the current loop resistance is generated;

[0023] If the power transmission efficiency is less than the preset efficiency threshold, a power adjustment strategy for adjusting the loop resistance until the power transmission efficiency is greater than or equal to the preset efficiency threshold is generated.

[0024] In a possible implementation, the method further includes:

[0025] According to the running state data, the state of charge and the state of health, a feature vector is generated;

[0026] The current cumulative cycle number of the battery is obtained;

[0027] According to the cumulative cycle number, a decay coefficient is determined;

[0028] The cumulative cycle number and the feature vector are input into a pre-trained neural network model to output a reference state value of the battery;

[0029] The feature vector, the cumulative cycle number, the decay coefficient and the reference state value are input into a preset fault prediction algorithm to obtain a fault feature value;

[0030] If the fault feature value is greater than a preset feature value threshold, a fault warning prompt information is output.

[0031] In a possible implementation, the method further includes:

[0032] According to the cumulative cycle number and the decay coefficient, a remaining life of the battery is determined.

[0033] In a second aspect, the embodiments of the present application provide a battery control device, comprising:

[0034] a collection module, configured to collect running state data of the battery, the running state data comprising current data, temperature data and battery internal resistance data;

[0035] a determination module, configured to determine a battery control voltage according to the current data and the temperature data;

[0036] a processing module, configured to input the current data, the temperature data, the battery internal resistance data and the battery control voltage into a pre-trained neural network model, and output a state of charge and a state of health of the battery;

[0037] The determination module is further configured to determine a power transmission efficiency of the battery according to the battery internal resistance data.

[0038] a generation module, configured to generate a power adjustment strategy according to the power transmission efficiency, the power adjustment strategy being used to adjust power consumption of the battery.

[0039] In a third aspect, the embodiments of the present application provide an electronic device, comprising a memory and a processor.

[0040] The memory stores computer execution instructions.

[0041] The processor executes the computer execution instructions stored in the memory, so that the processor executes the first aspect and / or various possible implementation manners of the first aspect.

[0042] In a fourth aspect, the embodiments of the present application provide a computer readable storage medium, the computer readable storage medium storing computer execution instructions, the computer execution instructions being executed by the processor to implement the first aspect and / or various possible implementation manners of the first aspect.

[0043] In a fifth aspect, the embodiments of the present application provide a computer program product, comprising a computer program, the computer program being executed by the processor to implement the first aspect and / or various possible implementation manners of the first aspect.

[0044] The battery control method, device, electronic device, storage medium, and program product provided in this application overcome the limitations of related technologies that rely solely on single parameters such as current and cycle number by collecting three core operating state data: current data, temperature data, and battery internal resistance data. Simultaneously, the battery control voltage is determined based on current and temperature data. Then, the current data, temperature data, battery internal resistance data, and battery control voltage are input into a pre-trained neural network model. The model's nonlinear mapping capability is used to accurately estimate the state of charge (SOC) and state of health, solving the problems of the Coulomb counting method being susceptible to environmental interference and having low estimation accuracy. Furthermore, this application combines battery internal resistance data to determine power transmission efficiency and generates a power adjustment strategy based on the power transmission efficiency. This replaces the coarse adjustment method based on voltage thresholds in related technologies, enabling a deep match between the power adjustment strategy and the actual operating state and energy efficiency level of the battery. This improves both the monitoring accuracy of SOC and state of health and the battery power consumption control effect, ultimately achieving a dual improvement in battery safety and energy efficiency. Attached Figure Description

[0045] The accompanying drawings, which are incorporated in and form part of this specification, illustrate embodiments consistent with this application and, together with the description, serve to explain the principles of this application.

[0046] Figure 1 This is a schematic diagram of an application scenario provided by an embodiment of this application;

[0047] Figure 2 A schematic flowchart illustrating a battery control method provided in an embodiment of this application;

[0048] Figure 3 A flowchart illustrating the method for fault early warning provided in this application embodiment;

[0049] Figure 4 This is a schematic diagram of the structure of a battery control device provided in an embodiment of this application;

[0050] Figure 5 This is a schematic diagram of the structure of an electronic device provided in an embodiment of this application.

[0051] The accompanying drawings illustrate specific embodiments of this application, which will be described in more detail below. These drawings and descriptions are not intended to limit the scope of the concept in any way, but rather to illustrate the concept of this application to those skilled in the art through reference to particular embodiments. Detailed Implementation

[0052] The exemplary embodiments will be described in detail below with reference to the drawings. In the following description, like reference numerals refer to like elements throughout the description. The following exemplary embodiments described below are not representative of all embodiments consistent with the present application. Rather, they are merely examples of apparatuses and methods consistent with some aspects of the present application as detailed in the appended claims.

[0053] Batteries are widely used in new energy storage fields, including electric vehicles, grid energy storage systems, distributed energy management systems, industrial backup power supplies, and smart grids. In the field of electric vehicles, the performance of batteries as the core power source directly determines the range, charging efficiency, and safety. In grid energy storage systems, batteries need to efficiently absorb fluctuating power from renewable energy sources such as photovoltaic and wind power, and release power during peak demand, while also needing to monitor battery health in real time to reduce large-scale failures. In industrial scenarios, energy storage batteries need to adapt to complex working conditions such as high temperature, low temperature, and high vibration environments, and meet the requirements of high reliability and long service life.

[0054] In related technologies, state of charge estimation mainly uses coulomb counting method, which calculates the change of electric quantity by integrating the charging and discharging current. Health state judgment is mostly only associated with the cumulative cycle number of the battery, and the degree of battery aging is roughly estimated according to the cycle number. The power adjustment strategy is relatively rough, and a preset fixed voltage threshold is generally used as the basis for adjustment. When the battery voltage reaches or exceeds the threshold, the corresponding power consumption adjustment action is triggered.

[0055] However, the above method leads to poor accuracy of SOC, SOH and power consumption control strategy, because the coulomb counting method does not take into account dynamic parameters such as temperature and internal resistance. Temperature changes the activity of the battery, and internal resistance increases energy loss, directly causing SOC estimation deviation. SOH judgment ignores the differential impact of environmental temperature on aging, and cannot reflect the real performance degradation. The power adjustment does not combine the real energy efficiency state of the battery, making the strategy disjointed with the actual working condition.

[0056] Therefore, in order to solve the problem of SOC estimation deviation due to the lack of temperature and internal resistance parameters in related technologies, the present application collects current, temperature, and battery internal resistance data to provide a basis for accurate estimation. In order to solve the defect that SOH judgment ignores the differential impact of the environment, the above multi-dimensional data and the battery control voltage determined based on current and temperature are cooperatively input into a neural network model, and the nonlinear mapping capability of the model is used to mine parameter correlations, achieving accurate output of SOC and SOH. In order to solve the problem that power adjustment relies on fixed voltage threshold and is disjointed with the actual working condition, the internal resistance data reflecting real-time energy efficiency are used to calculate power transmission efficiency, and a power adjustment strategy is generated accordingly, thereby improving the battery control performance.

[0057] For the convenience of understanding the method of the present application, an application scenario is exemplarily provided below, please refer to Figure 1 , Figure 1 An application scenario schematic diagram provided for the embodiment of the present application, which can include an electronic device 01 and a battery 02.

[0058] The electronic device 01 establishes a communication connection with the battery 02, and the electronic device 01 can receive the running state data of the battery 02 in real time and issue control instructions. The battery management module built-in the electronic device 01 will continuously collect the current data, temperature data and internal resistance data of the battery 02 according to the method of the present application. According to the current and temperature data, the battery control voltage under the current working condition is determined, and the above three types of data and the control voltage are input into the pre-trained neural network model to quickly output the state of charge and the health state of the battery 02. At the same time, the power transmission efficiency is determined based on the battery internal resistance data, and if the efficiency is lower than the preset threshold, a power adjustment strategy is generated to optimize the battery power consumption, so as to realize accurate control of the battery state and intelligent adjustment of the power consumption.

[0059] It can be understood that the above examples are only for illustration and do not limit the present application, and the specific application situation can be determined.

[0060] The technical solutions of the present application and how the technical solutions of the present application solve the above technical problems will be described in detail below with specific embodiments. The following specific embodiments can be combined with each other, and the same or similar concepts or processes can not be described in detail in some embodiments. The embodiments of the present application will be described below with reference to the drawings.

[0061] Please refer to Figure 2 , Figure 2 A flowchart of a battery control method provided for the embodiment of the present application, the execution subject of the method can be a battery control device. The battery control device can be realized by a computer program, or by a medium storing a related computer program, such as a U disk and / or an optical disk, or by an entity device integrated or installed with a related computer program, such as a chip or an electronic device. The electronic device can be a server, a server cluster, a smart terminal, etc., and the method can include the following steps:

[0062] S201, collecting running state data of the battery, the running state data including current data, temperature data and battery internal resistance data.

[0063] The running state data of the battery is collected by a distributed sensor array, which can be arranged in the following manner.

[0064] One temperature sensor is arranged for every two battery monomers, and one current sensor and one internal resistance module are arranged for the battery pack main loop. The temperature precision of the temperature sensor is about ±0.5℃, the temperature that can be detected by the temperature sensor is between-20℃ and 50℃, and the internal resistance that can be detected by the internal resistance monitoring module is between 0.1mΩ and 10Ω.

[0065] It can be understood that the above arrangement and the number and type of sensors used are only for illustration, for example, a voltage sensor can also be arranged for every one battery in the sensor array, and the voltage precision is about ±0.01V, etc., which is not limited in the present application.

[0066] The distributed sensor array collects current data , temperature data and battery internal resistance data and other original data at a preset frequency, for example, 10Hz.

[0067] After collecting the above original data, the Kalman filtering algorithm is used for filtering processing to remove noise and other interference information, and the noise elimination rate is ≥90%. The data after filtering processing is normalized to map the data to the 0-1 interval to reduce the influence of different dimensions. The data after the above processing process is ≤100ms, thereby obtaining the normalized current data , temperature data and battery internal resistance data .

[0068] S202, determine the battery control voltage according to the current data and the temperature data.

[0069] One possible implementation is to obtain the prediction analysis factor and the prediction accuracy data based on the pre-trained neural network model, and obtain the preset penalty function, and then input the prediction analysis factor, the prediction accuracy data, the penalty function, the current data and the temperature data into the preset voltage control algorithm to determine the battery control voltage.

[0070] In this embodiment, the pre-trained neural network model can be a fusion model obtained by learning through a long short-term memory (Long Short-Term Memory, LSTM) + convolutional neural network (Convolutional Neural Network, CNN).

[0071] Specifically, the preset voltage control algorithm can be as shown in the following formula (1):

[0072]

[0073] wherein, Battery control voltage; Different battery states, such as charging / discharging, and resting; The predictive analysis factor can be between 0.8 and 1.2. The pre-trained neural network model can be dynamically optimized according to the battery state to improve the predictive adaptability, with 1.05 being preferred. To improve prediction accuracy, a pre-trained neural network model can obtain prediction confidence based on the current battery state. ; The first penalty function is called "punishment," and its value can range from 0.1 to 0.5. This is the second penalty function, and its value range can also be from 0.1 to 0.5; The processed current data is monitored in real time; This is processed temperature data for real-time monitoring.

[0074] For example, suppose , , , , and All are 0.1. Substituting the above data into the above formula (1), we can get:

[0075] (1)

[0076] 3.85V is in line with the terminal voltage range of 2.5V~4.2V for lithium-ion batteries.

[0077] Formula (1) improves the prediction accuracy by fusing the neural network model. With real-time current and temperature feedback, the battery control voltage is dynamically calculated. This battery control voltage can reduce overcharging / over-discharging of the battery and is output to the charge / discharge control module as a voltage setpoint to adapt to the voltage requirements under different operating conditions.

[0078] S203. Input the current data, temperature data, battery internal resistance data, and battery control voltage into the pre-trained neural network model, and output the battery's state of charge and health status.

[0079] The processed current data, temperature data, battery internal resistance data, and battery control voltage ( , , , The input is fed into a pre-trained neural network model.

[0080] The model calls upon a historical sample library, which can pre-store p=2000 sets of normal state data for lithium-ion batteries, including but not limited to different cycle numbers, temperatures, and loads. 、 、 and . Through feature extraction and dynamic optimization, SOC and SOH are output.

[0081] The model performs feature extraction, and the CNN module in the model performs spatial feature extraction on the input 、 、 and to obtain feature data such as the correlation pattern of current-temperature and the matching relationship of internal resistance-voltage.

[0082] The model performs dynamic optimization, and the LSTM module in the model, in combination with the time series data of the historical sample library, i.e., the time series of 2000 groups of normal state data, captures the trend data of the change of the battery state with the cycle number, such as the long-term rule that the increase of internal resistance leads to the decrease of SoH, and other dynamic optimization , based on the model prediction accuracy, for example = 0.97, then maintain = 1.05, and if <0.95, then adjust to 1.1.

[0083] Further, based on the above feature data and the optimized parameters, SOC, for example 85%, and SOH, for example 92%, are output.

[0084] Compared with the SOC obtained by the Coulomb counting method in the related art, for example 82%, the calculation error of the present application is ≤3%, and the SOC is more accurate. The , SOC, and SOH are output to the charge and discharge control module through the CAN bus (delay ≤50ms) to enable the charge and discharge control module to dynamically adjust the charge and discharge current and voltage threshold based on the accurate state parameters, thereby realizing safe and efficient charge and discharge control of the battery.

[0085] In S204, the power transmission efficiency of the battery is determined according to the battery internal resistance data.

[0086] One possible implementation is to obtain the loop resistance of the battery, the preset initial efficiency, and the current density integral data, input the loop resistance, the preset initial efficiency, the current density integral data, and the battery internal resistance data into the preset efficiency algorithm, and determine the power transmission efficiency of the battery.

[0087] Specifically, the preset efficiency algorithm can be as shown in the following formula (2):

[0088]

[0089] wherein, is the current power transmission efficiency, is the battery internal resistance, which is determined by battery material, structure, temperature and other factors; is the loop resistance, including battery internal resistance, connection resistance, line resistance, etc.; is the initial efficiency, which can be 0.8 to 0.9, indicating an ideal state, initial calibration value or system inherent reference efficiency; is the current density integral, the larger the integral term, the more uneven the current distribution, the greater the loss, and the lower the total efficiency.

[0090] S205, according to the power transmission efficiency, generating power adjustment strategy, power adjustment strategy is used to adjust the power consumption of the battery.

[0091] If the power transmission efficiency is greater than or equal to the preset efficiency threshold, a power adjustment strategy is generated to maintain the current loop resistance, and if the power transmission efficiency is less than the preset efficiency threshold, a power adjustment strategy is generated to adjust the loop resistance until the power transmission efficiency is greater than or equal to the preset efficiency threshold.

[0092] The power transmission efficiency of the battery directly reflects the effective degree of energy utilization, and the loop resistance Rp is a key factor affecting the efficiency. According to the calculation logic of energy transmission efficiency loss, the greater the loop resistance, the more energy loss in the form of heat and other forms during transmission, and the higher the invalid power consumption of the battery. When the power transmission efficiency is greater than or equal to the preset efficiency threshold, it means that the current loop resistance is in a reasonable range, and the energy loss is controlled at a low level, so a power adjustment strategy is generated to maintain the current loop resistance, thereby avoiding additional power consumption by maintaining the existing low loss state.

[0093] When the power transmission efficiency is less than the preset efficiency threshold, it means that the current loop resistance is too large, causing energy loss to exceed a reasonable range. At this time, the strategy of adjusting the loop resistance is generated to reduce Rp by optimizing the charging and discharging loop topology. As the loop resistance decreases, the efficiency loss (DPT) in the energy transmission process decreases, and the energy output by the battery can be more efficiently supplied to the load rather than being converted into invalid loss, thereby achieving the goal of reducing battery power consumption. In this process, the power adjustment strategy directly acts on the power transmission efficiency by precisely regulating the core variable of the loop resistance, ultimately achieving the purpose of controlling the battery power consumption.

[0094] Formula (2) can calculate the efficiency loss in the energy transmission process. When DPT < preset efficiency threshold, for example, 0.9, the loop resistance is automatically adjusted by optimizing the charging and discharging loop topology , output to the load balancing module until the efficiency meets the standard and the power consumption is reduced.

[0095] Exemplarily, assume , , the current density data is collected for 20 times, if the calculated current density integral data , which can reflect the uniformity of current distribution.

[0096] The above data is input into formula (2) to obtain:

[0097]

[0098] Satisfies When is greater than or equal to the preset efficiency threshold, the current loop parameter is maintained, that is, the current loop resistance is maintained. If is less than the preset efficiency threshold, adjust to 0.2Ω or less, and recalculate until the new is greater than or equal to the preset efficiency threshold.

[0099] In the present application, the data used can be encrypted and transmitted based on a preset encryption algorithm, which can be a symmetric encryption algorithm or an asymmetric encryption algorithm, etc.

[0100] The neural network model pre-trained in the present application is trained in the training stage by the way of "adversarial training" to improve the anti-interference ability of the model. That is, during training, malicious data with a small disturbance is intentionally input to let the model master the ability to identify and resist such attacks during the learning process, so that the error of the model output result can be controlled within 3% even when facing malicious input, avoiding the output of incorrect data due to model attack.

[0101] In the above embodiments of the present application, by collecting three types of core operating state data: current data, temperature data and battery internal resistance data, the single parameter limitation of related art relying only on current and cycle number is broken. At the same time, the battery control voltage is determined based on the current data and the temperature data, and then the current data, the temperature data, the battery internal resistance data and the battery control voltage are input into the pre-trained neural network model, and the nonlinear mapping ability of the model is used to realize accurate estimation of the state of charge and the state of health, solving the problem of low estimation accuracy of the coulomb counting method which is easily disturbed by environmental factors. On this basis, the present application determines the power transmission efficiency in combination with the battery internal resistance data, and generates a power adjustment strategy based on the power transmission efficiency, replacing the extensive adjustment mode based on voltage threshold in related art, so that the power adjustment strategy and the actual operating state of the battery, the energy efficiency level are deeply matched, which not only improves the monitoring accuracy of the state of charge and the state of health, but also improves the battery power consumption control effect, ultimately realizing the dual improvement of safety and energy efficiency of battery operation.

[0102] In the present application, fault early warning can also be performed, please refer to Figure 3 , Figure 3A flowchart of a method for fault early warning provided by an embodiment of the present application can include the following steps:

[0103] S301, generating a feature vector according to the running state data, the state of charge and the state of health.

[0104] The feature vector is generated by using a feature splicing method to integrate parameters in a fixed dimension order, thereby providing a standardized input for subsequent fault prediction and other processing.

[0105] S302, obtaining the current cumulative cycle number of the battery.

[0106] The battery management system is built-in with a cycle number counting module, which determines whether a complete charge-discharge cycle is completed by real-time monitoring of the feature inflection point of the battery charge-discharge curve.

[0107] After each complete cycle is detected, the system will update the cumulative cycle number in the non-volatile storage unit to reduce power loss data.

[0108] When this step is performed, the current cumulative cycle number of the battery recorded in the storage unit can be directly called.

[0109] S303, determining the attenuation coefficient according to the cumulative cycle number.

[0110] The cycle number-attenuation reference curve pre-stored in the battery management system is called, which is generated by accelerated aging experiments of the same type of battery. The current cumulative cycle number is substituted into the curve, and the theoretical capacity attenuation ratio under the corresponding cycle number is calculated by linear interpolation method. The theoretical attenuation ratio is converted into an attenuation coefficient, and a correction factor is introduced to fine-tune the battery if it has a special use scenario. Finally, the attenuation coefficient matching the current aging state of the battery is output.

[0111] S304, inputting the cumulative cycle number and the feature vector into a pre-trained neural network model to output the reference state value of the battery.

[0112] Suppose the cumulative cycle number exceeds the maximum range of the preset historical sample, for example, the maximum sample library is 5000 times, and the current is 5001 times. The pre-trained neural network model is then calculated by the attenuation coefficient a to output the reference state value of the battery.

[0113] The extension calculation can be shown in the following formula (3):

[0114]

[0115] wherein, is the reference state value of the battery, which can be the reference value under the running state data, the state of charge and the state of health; is the cumulative cycle number; is a decay coefficient; is a reference state value of the battery when the cumulative cycle number is 5000; represents an external calculation.

[0116] With the increase of , the value needs to be synchronized to the state range of normal aging, for example, the internal resistance value is 0.05 ohms at the 100th charge and discharge, and the value may be 0.08 ohms at the 1000th time, therefore, the reference state value needs to be determined according to the cumulative cycle number, so as to reduce the case of misjudging normal aging as failure.

[0117] S305, input the feature vector, the cumulative cycle number, the decay coefficient and the reference state value into the preset failure prediction algorithm to obtain a failure feature value.

[0118] Specifically, the preset failure prediction algorithm can be as shown in the following formula (4):

[0119]

[0120] wherein, is a failure feature vector value; is a proportional coefficient, which can be 0.5 to 1.5; M is the sample number; is a feature vector of real-time observation value, which changes with the increase of , the battery aging will cause the value to change, for example, there is a trend of increasing internal resistance and decreasing capacity.

[0121] The modulus of is determined as the failure feature value.

[0122] In this application, different chemical system batteries such as lithium ion and solid state can be supported, and by parameterizing the , in the formula, the related failure feature value can be obtained without modifying the hardware. Thus, the adaptability is strong, and the adaptation cost can be reduced by about 30%.

[0123] S306, if the failure feature value is greater than a preset feature value threshold, output a failure warning prompt information.

[0124] In this embodiment, the failure warning threshold, i.e. the feature value threshold, is determined in advance by training a large number of historical failure samples, which is assumed to be 0.1.

[0125] According to the fault feature value obtained according to the above steps, it is judged whether the value is greater than 0.1, if less than or equal to 0.1, it is determined that the battery is in a normal operation state, and the current operation strategy is maintained. If greater than 0.1, it is determined that the battery has a fault risk, such as abnormal increase of internal resistance, precursor of thermal runaway, etc., and the communication module triggers a warning push, and outputs a fault warning prompt information through local or cloud, to remind the user.

[0126] Optionally, the communication module can be integrated with various interfaces and protocols, including but not limited to CAN interface (connected to energy storage system, delay ≤ 50 ms), standard protocol (connected to smart grid), and Message Queuing Telemetry Transport (MQTT) protocol, for connecting to the cloud, etc.

[0127] In the above embodiments of the present application, by fusing the battery operation state data, state of charge, state of health to construct a feature vector, and introducing the cumulative cycle number and the attenuation coefficient determined based on the number, the reference state value output by the pre-trained neural network model is combined to form a multi-dimensional, full-life-cycle battery state data input system, which effectively makes up for the defect that a single parameter cannot fully reflect the battery state. The core parameters are cooperatively input into the preset fault prediction algorithm, which can deeply mine the correlation between parameters and abnormal evolution law, so that the precision of the output fault feature value is significantly improved, and then the fault warning is realized through the comparison between the fault feature value and the threshold value. The passive mode of post-response in fault monitoring is changed to the active protection of pre-judgment, so as to avoid the safety risk caused by battery failure in advance.

[0128] In the present application, the remaining life of the battery can also be determined according to the cumulative cycle number and the attenuation coefficient.

[0129] Remaining Useful Life (RUL) refers to the remaining cycle number required for the battery to decay from the current state to 85% of the rated value.

[0130] Attenuation coefficient The current aging speed of the battery, The closer to 0.85, the faster the aging.

[0131] The cumulative cycle number and the attenuation coefficient are input into the following formula (5) to output the remaining life of the battery:

[0132]

[0133] Wherein, represents the current cumulative cycle number.

[0134] Exemplarily, assuming = 0.92, =500, input the above formula (5), get:

[0135]

[0136] In the above embodiment of the application, the remaining life of the battery is determined by accumulating the cycle number and the attenuation coefficient, which can make up for the defects of traditional single parameter estimation. The cumulative cycle number provides a macro basis, and the attenuation coefficient dynamically corrects the theoretical life loss, so as to reduce the misjudgment of aging rate.

[0137] In the application, by simulating the working condition of-20℃~50℃, it can be concluded that the application and related technologies are compared:

[0138] After 1000 cycles, the capacity retention rate of the battery of the application is 82%, and that of the traditional BMS is 65%. The energy conversion efficiency of the battery of the application is 98.7%, and that of the related technology is 92.3%. The thermal runaway early warning of the application is 5.2min in advance, and the related technology has no early warning capability.

[0139] Please refer to Figure 4 , Figure 4 A structural schematic diagram of a battery control device provided by an embodiment of the application is shown in Figure 4 , which comprises:

[0140] The acquisition module 401 is configured to acquire the running state data of the battery, and the running state data comprises current data, temperature data and battery internal resistance data.

[0141] The determination module 402 is configured to determine the battery control voltage according to the current data and the temperature data.

[0142] The processing module 403 is configured to input the current data, the temperature data, the battery internal resistance data and the battery control voltage into a pre-trained neural network model, and output the state of charge and the health state of the battery.

[0143] The determination module 402 is further configured to determine the power transmission efficiency of the battery according to the battery internal resistance data.

[0144] The generation module 404 is configured to generate a power adjustment strategy according to the power transmission efficiency, and the power adjustment strategy is used to adjust the power consumption of the battery.

[0145] In a possible implementation manner, the determination module 402 is specifically configured to:

[0146] Obtain the prediction analysis factor and the prediction accuracy data based on the pre-trained neural network model.

[0147] Obtain a preset penalty function.

[0148] The prediction analysis factor, the prediction accuracy data, the penalty function, the current data and the temperature data are input into a preset voltage control algorithm to determine a battery control voltage.

[0149] In a possible implementation, the determining module 402 is specifically configured to:

[0150] The loop resistance, the preset initial efficiency and the current density integral data of the battery are obtained.

[0151] The loop resistance, the preset initial efficiency, the current density integral data and the internal resistance data of the battery are input into a preset efficiency algorithm to determine a power transmission efficiency of the battery.

[0152] In a possible implementation, the generating module 404 is specifically configured to:

[0153] If the power transmission efficiency is greater than or equal to the preset efficiency threshold, a power adjustment strategy for maintaining the current loop resistance is generated.

[0154] If the power transmission efficiency is less than the preset efficiency threshold, a power adjustment strategy for adjusting the loop resistance until the power transmission efficiency is greater than or equal to the preset efficiency threshold is generated.

[0155] In a possible implementation, the processing module 403 is further configured to:

[0156] A feature vector is generated according to the running state data, the state of charge and the state of health.

[0157] The current cumulative cycle number of the battery is obtained.

[0158] An attenuation coefficient is determined according to the cumulative cycle number.

[0159] The cumulative cycle number and the feature vector are input into a pre-trained neural network model to output a reference state value of the battery.

[0160] The feature vector, the cumulative cycle number, the attenuation coefficient and the reference state value are input into a preset fault prediction algorithm to obtain a fault feature value.

[0161] If the fault feature value is greater than a preset feature value threshold, a fault warning prompt information is output.

[0162] In a possible implementation, the processing module 403 is further configured to:

[0163] The remaining life of the battery is determined according to the cumulative cycle number and the attenuation coefficient.

[0164] The battery control device provided in the embodiment can execute the method provided in the method embodiment, and has similar implementation principles and technical effects, which will not be described here in detail.

[0165] Figure 5 A structural schematic diagram of an electronic device is provided in an embodiment of the present application. As shown in the figure, the electronic device provided in the embodiment includes at least one processor 501 and a memory 502. Optionally, the device further includes a communication component 503. The processor 501, the memory 502 and the communication component 503 are connected through a bus 504. Figure 5

[0166] In the implementation process, the at least one processor 501 executes the computer execution instructions stored in the memory 502, so that the at least one processor 501 executes the above-mentioned method.

[0167] The specific implementation process of the processor 501 can refer to the above-mentioned method embodiments, which have similar implementation principles and technical effects, and will not be described here in detail.

[0168] In the above-mentioned embodiments, it should be understood that the processor can be a central processing unit (CPU), and can also be other general-purpose processors, digital signal processors (DSP), application specific integrated circuits (ASIC) and the like. The general-purpose processor can be a microprocessor or the processor can also be any conventional processor and the like. The steps of the method disclosed in the application can be directly embodied as the execution of the hardware processor, or executed by the combination of hardware and software modules in the processor.

[0169] The memory can include a random access memory (RAM), and can also include a non-volatile memory (NVM), for example, at least one disk memory.

[0170] The bus can be an industry standard architecture (ISA) bus, a peripheral component (PCI) bus or an extended industry standard architecture (EISA) bus and the like. The bus can be divided into an address bus, a data bus, a control bus and the like. For the convenience of representation, the bus in the drawings of the present application does not limit to only one bus or one type of bus.

[0171] The present application also provides a computer program product, including a computer program, which is executed by the processor to realize the above-mentioned method.​

[0172] The application further provides a computer readable storage medium, wherein computer execution instructions are stored in the computer readable storage medium, and when a processor executes the computer execution instructions, the method described above is realized.

[0173] The readable storage medium described above can be realized by any type of volatile or non-volatile storage device or a combination thereof, such as static random access memory, electrically erasable programmable read-only memory, erasable programmable read-only memory, programmable read-only memory, read-only memory, magnetic storage, flash memory, magnetic disk or optical disk. The readable storage medium can be any available medium that can be accessed by a general-purpose or special-purpose computer.

[0174] An exemplary readable storage medium is coupled to the processor, so that the processor can read information from the readable storage medium and write information to the readable storage medium. Of course, the readable storage medium can also be an integral part of the processor. The processor and the readable storage medium can be located in an application specific integrated circuit (ASIC). Of course, the processor and the readable storage medium can also exist as discrete components in the device.

[0175] The division of units is only a logical function division, and when actually implemented, there can be another division mode, for example, a plurality of units or components can be combined or integrated into another system, or some features can be ignored or not executed. Another point is that the coupling or direct coupling or communication connection between the units or components shown or discussed can be indirect coupling or communication connection through some interfaces, devices or units, which can be electrical, mechanical or other forms.

[0176] The units described as separate components can or can not be physically separated, and the components shown as units can or can not be physical units, that is, they can be located in one place, or they can be distributed on a plurality of network units. According to actual needs, some or all of the units can be selected to achieve the purpose of the embodiment.

[0177] In addition, the functional units in each embodiment of the application can be integrated into one processing unit, or each unit can be physically present separately, or two or more units can be integrated into one unit.

[0178] If the functions are implemented in the form of software function units and sold or used as independent products, they can be stored in a computer readable storage medium. Based on this understanding, the technical solutions of the present application or the parts of the technical solutions that essentially contribute to the prior art can be embodied in the form of a software product. The computer software product is stored in a storage medium and includes a number of instructions for causing a computer device (which can be a personal computer, a server, or a network device, etc.) to execute all or part of the steps of the embodiments of the method of the present application. The aforementioned storage medium includes: a U disk, a mobile hard disk, a read-only memory (ROM), a random access memory (RAM), a magnetic disk or an optical disk, and various media that can store program codes.

[0179] It can be understood by those skilled in the art that all or part of the steps of the above-mentioned method embodiments can be completed by program instruction related hardware. The aforementioned program can be stored in a computer readable storage medium. When the program is executed, the steps of the above-mentioned method embodiments are executed; and the aforementioned storage medium includes: ROM, RAM, magnetic disk or optical disk, and various media that can store program codes.

[0180] Finally, it should be noted that: those skilled in the art will easily derive other embodiments of the present application after considering the specification and practicing the application disclosed herein. The present application is intended to cover any variations, uses, or adaptations of the present application that follow the general principles of the present application and include common knowledge or conventional technical means in the art that are not disclosed in the present application, and is not limited to the precise structure described above and shown in the drawings, and various modifications and changes can be made without departing from the scope thereof. The scope of the present application is only limited by the appended claims.

Claims

1. A method for controlling a battery, characterized in that, include: Collect battery operating status data, including current data, temperature data, and battery internal resistance data; The battery control voltage is determined based on the current data and the temperature data. The current data, temperature data, battery internal resistance data, and battery control voltage are input into a pre-trained neural network model, which outputs the battery's state of charge and health status. The power transmission efficiency of the battery is determined based on the battery internal resistance data. Based on the power transmission efficiency, a power adjustment strategy is generated, which is used to adjust the power consumption of the battery.

2. The method according to claim 1, characterized in that, The step of determining the battery control voltage based on the current data and the temperature data includes: Obtain the prediction analysis factors and prediction accuracy data based on the pre-trained neural network model; Get the preset penalty function; The predictive analysis factor, the predictive accuracy data, the penalty function, the current data, and the temperature data are input into a preset voltage control algorithm to determine the battery control voltage.

3. The method according to claim 1, characterized in that, Determining the power transfer efficiency of the battery based on the battery's internal resistance data includes: Obtain the battery's loop resistance, preset initial efficiency, and current density integral data; The circuit resistance, the preset initial efficiency, the current density integral data, and the battery internal resistance data are input into a preset efficiency algorithm to determine the power transmission efficiency of the battery.

4. The method according to claim 1, characterized in that, The generation of a power adjustment strategy based on the power transmission efficiency includes: If the power transmission efficiency is greater than or equal to a preset efficiency threshold, a power adjustment strategy to maintain the current loop resistance is generated. If the power transmission efficiency is less than the preset efficiency threshold, a power adjustment strategy is generated to adjust the loop resistance until the power transmission efficiency is greater than or equal to the preset efficiency threshold.

5. The method according to claim 1, characterized in that, Also includes: Based on the operating status data, the state of charge, and the health status, a feature vector is generated; Obtain the current cumulative cycle count of the battery; The attenuation coefficient is determined based on the cumulative number of cycles. The cumulative number of cycles and the feature vector are input into the pre-trained neural network model, and the baseline state value of the battery is output. The feature vector, the cumulative number of cycles, the attenuation coefficient, and the baseline state value are input into a preset fault prediction algorithm to obtain fault feature values. If the fault characteristic value is greater than the preset characteristic value threshold, a fault warning message will be output.

6. The method according to claim 5, characterized in that, Also includes: The remaining lifespan of the battery is determined based on the cumulative number of cycles and the degradation coefficient.

7. A battery control device, characterized in that, include: The acquisition module is used to acquire battery operating status data, including current data, temperature data, and battery internal resistance data. The determining module is used to determine the battery control voltage based on the current data and the temperature data; The processing module is used to input the current data, the temperature data, the battery internal resistance data, and the battery control voltage into a pre-trained neural network model, and output the state of charge and health status of the battery. The determining module is further configured to determine the power transmission efficiency of the battery based on the battery internal resistance data. A generation module is used to generate a power adjustment strategy based on the power transmission efficiency, the power adjustment strategy being used to adjust the power consumption of the battery.

8. An electronic device, characterized in that, include: Memory, processor; The memory stores computer-executed instructions; The processor executes computer execution instructions stored in the memory, causing the processor to perform the method as described in any one of claims 1-6.

9. A computer-readable storage medium, characterized in that, The computer-readable storage medium stores computer-executable instructions, which, when executed by a processor, are used to implement the method as described in any one of claims 1-6.

10. A computer program product, characterized in that, Includes a computer program that, when executed by a processor, implements the method described in any one of claims 1-6.

Images