Battery pack operation and maintenance monitoring method and system based on artificial intelligence

Through the operation and maintenance monitoring methods and systems of battery packs based on artificial intelligence, the problem of unbalanced battery pack voltage is solved, and the consistent management of battery pack performance is achieved, which extends the service life and reduces maintenance costs and accident risks.

CN118655476BActive Publication Date: 2024-11-19国家能源集团永州发电有限公司
View PDF 0 Cites 0 Cited by

Patent Information

Application Number
CN202410769204.5
Authority / Receiving Office
CN · China
Patent Type
Patents(China)
Current Assignee / Owner
Filing Date
2024-06-14
Publication Date
2024-11-19
Estimated Expiration
2044-06-14

AI Technical Summary

Technical Problem

During the charging and discharging process, voltage imbalances caused by differences in battery parameters and environment, resulting in some batteries being overcharged or overdischarged, shortening the service life of the battery pack and increasing maintenance costs and accident risks.

Method used

Adopt battery pack operation and maintenance monitoring methods and systems based on artificial intelligence, and collect battery pack data in real time through online monitoring equipment, analyze and process data to achieve active equalization and online sulfur removal, and use neural network models and machine learning algorithms to predict battery status and life , automated operation and maintenance management to maintain consistent battery pack performance.

Benefits of technology

It effectively extends the service life of the battery pack, reduces maintenance costs and accident risks, improves the safety and use efficiency of the battery pack, and realizes automated operation and maintenance management.

✦ Generated by Eureka AI based on patent content.

Smart Images

  • Figure CN118655476B_ABST
    Figure CN118655476B_ABST
Patent Text Reader

Abstract

The present invention discloses an artificial intelligence-based battery pack operation and maintenance monitoring method and system, which belongs to the field of battery monitoring. The method includes: configuring the online monitoring device through the parameter configuration interface of the host computer; the online monitoring device collects the operation data and ambient temperature of the battery pack in real time to determine whether to issue an alarm, whether to perform active equalization, activation operation and online desulfurization; the host computer analyzes and processes the operation data and ambient temperature to obtain visual information; according to different battery pack pressure difference alarm thresholds and the corresponding battery pack pressure difference historical alarm frequency, as well as the corresponding battery service life, the optimal battery pack pressure difference alarm threshold is obtained; based on the optimal battery pack pressure difference alarm threshold, the online monitoring device is configured. The present invention can greatly reduce the maintenance costs such as manpower and material resources, and improve the safety of battery use.
Need to check novelty before this filing date? Find Prior Art

Description

Technical Field

[0001] The present invention relates to the field of battery monitoring, and in particular to an artificial intelligence-based battery pack operation and maintenance monitoring method and system. Background Art

[0002] The battery is an essential backup power source for the power system and the last line of defense in the operation of the equipment. The operation of the existing battery pack has some technical defects. The battery pack of the DC system is generally composed of dozens to more than one hundred single batteries in series. When the battery pack is in the series state, the current of all batteries is consistent during charging and discharging. However, due to the differences in battery parameters, external environment and single self-discharge, the voltage of each single battery in the battery pack is actually not balanced. Some voltages are too high and some are too low, causing some single batteries in the battery pack to be overcharged or over-discharged. The overcharged battery evaporates water and increases internal resistance, resulting in a decrease in capacity. The undercharged battery coating ages and reduces active substances, which also causes a decrease in capacity. Once this process begins, the decrease in capacity will continue to accelerate with the increase of time, which will further deepen the inconsistency of battery parameters. It is this vicious cycle that greatly shortens the service life of the battery pack. Such long-term unbalanced charging will inevitably cause serious damage to some batteries. Once the battery fails, it will cause protection failure, switch refusal to operate, and channel interruption, with disastrous consequences. Summary of the invention

[0003] The present invention aims to solve at least one of the technical problems existing in the prior art. To this end, the present invention proposes a battery pack operation and maintenance monitoring method and system based on artificial intelligence, which can make the battery pack performance consistently operate in the best state, improve the backup time and operating life of the battery pack, timely detect and maintain the outdated batteries, greatly reduce the maintenance costs such as manpower and material resources, improve the safety of battery use, reduce the accident rate, and effectively save energy and reduce emissions.

[0004] On the one hand, an embodiment of the present invention provides a battery pack operation and maintenance monitoring method based on artificial intelligence, comprising the following steps: S100, configuring an online monitoring device through a parameter configuration interface of a host computer, including configuring basic battery information, upper and lower limits of single cell voltage, upper and lower limits of internal resistance, upper temperature limit, upper charging current limit, upper and lower limits of battery pack voltage, battery pack pressure difference alarm threshold, SOC lower limit, SOH lower limit, measurement start current value, group voltage value mode, balancing control mode, balancing degree threshold, balancing time, internal resistance measurement mode, internal resistance measurement cycle and alarm mode; S200, the online monitoring device collects the operation data and ambient temperature of the battery pack in real time, determines whether to issue an alarm based on the configuration information, and determines whether to perform active balancing, activation operation and online desulfurization based on the operation data; the operation data includes: the voltage, internal resistance, capacity, electromotive force of the battery pack and the single cell, and the charge and discharge current of the battery pack; S300, in response to the host computer requesting an operation, the online monitoring device collects the real-time or historical operation data and ambient temperature through The communication port is sent to the host computer; the host computer analyzes and processes the operating data and ambient temperature, obtains visual information and displays it on the host computer visual interface; the visual information includes: real-time status information of each battery pack, real-time status information of a single cell, curves of the battery group voltage and single cell voltage during discharge, historical curves, real-time alarms and historical alarm information; S400, obtaining different battery pack pressure difference alarm thresholds and corresponding battery pack pressure difference historical alarm frequencies, as well as corresponding battery service life as training data, training to obtain a first neural network model with the battery pack pressure difference alarm threshold and the battery pack pressure difference alarm frequency as input and the battery service life as output; setting the maximum value of the alarm frequency and the minimum value of the battery service life as constraints, finding the maximum value of the objective function based on the first neural network model and the constraints, and obtaining the optimal battery pack pressure difference alarm threshold; configuring the online monitoring device based on the optimal battery pack pressure difference alarm threshold; wherein the objective function is s=T / W, T is the battery service life, and W is the battery pack pressure difference alarm frequency.

[0005] According to some embodiments of the present invention, the host computer analyzes and processes the operating data and the ambient temperature to obtain visualization information including: obtaining basic information of the single cells in each battery pack, including type, model, batch, charge and discharge times, and charge and discharge time, and grouping the single cells according to the basic information; based on multiple SOH prediction models, obtaining real-time status information of several single cells in different groups to predict the SOH of the single cells, and obtaining several SOH prediction values ​​corresponding to several single cells; according to the SOH prediction values ​​and SOH actual values ​​of several single cells, obtaining the prediction accuracy of each SOH prediction model corresponding to each group of single cells, and selecting the highest prediction value of each group of cells The SOH prediction model corresponding to the accuracy is used as the optimal prediction model for the grouped batteries; the SOH prediction is performed on the single cells of each battery group. When the performance consistency of the battery group is qualified, any single cell in the battery group is selected, and the optimal prediction model is obtained according to the grouping information, and the SOH prediction value of each single cell in the battery group is obtained based on the optimal prediction model and displayed on the visualization interface; when the performance consistency of the battery group is unqualified, the grouping information of each single cell in the battery group and the corresponding optimal prediction model are obtained, and the SOH prediction value of each single cell is obtained based on the corresponding optimal prediction model and displayed on the visualization interface; wherein the actual SOH value is obtained through charge and discharge test.

[0006] According to some embodiments of the present invention, the SOH prediction model includes several prediction models obtained based on machine learning algorithms, the machine learning algorithms including Gaussian process regression, support vector machine and random forest; the input of the SOH prediction model includes any one or more of the following: voltage, current, internal resistance and temperature of a single cell.

[0007] According to some embodiments of the present invention, the host computer analyzes and processes the operating data and ambient temperature to obtain visualization information including: analyzing the received operating data based on the limit values ​​in the configuration information, and displaying the data that exceeds the limit values ​​in the operating data displayed on the visualization interface in a color different from the operating data that does not exceed the limit values; the limit values ​​include: upper and lower limits of single cell voltage, upper and lower limits of internal resistance, upper temperature limit, upper charging current limit, upper and lower limits of battery pack voltage, battery pack voltage difference alarm threshold, SOC lower limit and SOH lower limit.

[0008] According to some embodiments of the present invention, the method also includes: obtaining historical alarm information of eliminated aging batteries, training a neural network model based on the historical alarm information, and obtaining a second neural network model; the second neural network model takes the historical alarm information as input and the remaining service life as output; the historical alarm information includes the number and value of exceeding the upper and lower limits of voltage, the number and value of exceeding the upper and lower limits of internal resistance, the number and value of exceeding the upper temperature limit, the value of exceeding the lower limit of SOC, the value of exceeding the lower limit of SOH, and the time of issuing the alarm information; based on the second neural network model, the remaining life of the single cells of each battery pack is predicted, and the predicted values ​​obtained are displayed through a visual interface.

[0009] According to some embodiments of the present invention, the method also includes: when the internal resistance measurement mode is manual measurement, when the host computer issues a real-time internal resistance measurement request, locating the management personnel through the battery pack and single cell identification information of the real-time internal resistance measurement request; and sending the real-time internal resistance measurement request information to the mobile device of the management personnel.

[0010] The artificial intelligence-based battery pack operation and maintenance monitoring method of the embodiment of the present invention includes at least the following beneficial effects: the embodiment of the present invention realizes the network control of the internal resistance test, the unified network management of the online batteries, effectively improves the maintenance efficiency, timely discovers the hidden troubles, and reduces the maintenance of the battery; it can fully and dynamically sense the voltage, internal resistance, temperature changes and change trends of each battery, and on this basis, timely eliminate the sulfation of the battery found to be sulfided through the online desulfurization technology, slow down or reduce the internal resistance of the battery; for the battery that has not yet deteriorated, the active balancing technology is used to improve the consistency of each battery in the whole battery group, so as to slow down and inhibit the degradation rate of the battery. Through the online desulfurization and active balancing technology, the battery pack can be automatically operated and maintained, and finally the service life of the battery is extended; the embodiment of the present invention also analyzes the frequency and threshold of the historical alarm information of the battery pack voltage difference and the corresponding battery service life to obtain the optimal battery pack voltage difference alarm threshold, while ensuring the battery service life, reducing redundant alarm information and saving manpower.

[0011] On the other hand, an embodiment of the present invention provides an artificial intelligence-based battery pack operation and maintenance monitoring system, including: an acquisition layer, including a number of online monitoring devices for collecting battery pack information; a communication layer, used to connect the online monitoring devices and the main station layer through a switch and Ethernet; the main station layer, including a number of servers, for analyzing and managing battery pack information.

[0012] According to some embodiments of the present invention, the online monitoring device includes: a collection port, an internal resistance measurement switch, a communication port and an output port.

[0013] According to some embodiments of the present invention, the acquisition port includes: a battery voltage acquisition port, a current sensor Hall signal acquisition port, and an ambient temperature acquisition port.

[0014] The artificial intelligence-based battery pack operation and maintenance monitoring system of the embodiment of the present invention includes at least the following beneficial effects: the embodiment of the present invention realizes the network control of the internal resistance test, the unified network management of the online batteries, effectively improves the maintenance efficiency, timely discovers the hidden troubles, and reduces the maintenance of the batteries; it can fully and dynamically sense the voltage, internal resistance, temperature changes and change trends of each battery, and on this basis, timely eliminate the sulfation of the batteries found to be sulfided through the online desulfurization technology, slow down or reduce the internal resistance of the battery; for the batteries that have not yet deteriorated, the active balancing technology is used to improve the consistency of each battery in the whole battery group, so as to slow down and inhibit the degradation rate of the battery. Through the online desulfurization and active balancing technology, the battery pack can be automatically operated and maintained, and finally the service life of the battery is extended.

[0015] Additional aspects and advantages of the present invention will be given in part in the following description and in part will be obvious from the following description, or will be learned through practice of the present invention. BRIEF DESCRIPTION OF THE DRAWINGS

[0016] The above and / or additional aspects and advantages of the present invention will become apparent and easily understood from the description of the embodiments in conjunction with the following drawings, in which:

[0017] Figure 1 A schematic diagram of a method according to an embodiment of the present invention;

[0018] Figure 2 A schematic diagram of the system architecture of a system according to an embodiment of the present invention;

[0019] Figure 3 Visual information obtained by an embodiment of the present invention;

[0020] Figure 4 Visual information obtained by another embodiment of the present invention;

[0021] Figure 5 Visual information obtained by another embodiment of the present invention;

[0022] Figure 6 This is the visualization information obtained according to another embodiment of the present invention. DETAILED DESCRIPTION

[0023] Embodiments of the present invention are described in detail below, examples of which are shown in the accompanying drawings, wherein the same or similar reference numerals throughout represent the same or similar elements or elements having the same or similar functions. The embodiments described below with reference to the accompanying drawings are exemplary and are only used to explain the present invention, and cannot be understood as limiting the present invention.

[0024] In the description of the present invention, "several" means one or more, "more" means two or more, "greater than", "less than", "exceed" etc. are understood as not including the number itself, and "above", "below", "within" etc. are understood as including the number itself. If there is a description of "first" or "second", it is only used for the purpose of distinguishing the technical features, and cannot be understood as indicating or implying the relative importance or implicitly indicating the number of the indicated technical features or implicitly indicating the order of the indicated technical features.

[0025] The embodiment of the present invention is an online monitoring device that realizes online monitoring of battery internal resistance, single cell voltage, battery group voltage, current, temperature, etc., automatically adjusts the size of the charging current of each battery according to the capacity, so that each battery is in the same working state, prevents overcharging and undercharging of the battery, avoids long-term imbalance of the battery group and damage to the battery, so that the battery group is always kept in the best operating state: and through the intelligent analysis function, automatically analyzes the monitoring data to find out the failed or about to fail batteries, and promptly warns the user when the battery group single cell voltage, single cell internal resistance, group voltage, current, etc. exceed the threshold. Remote monitoring can be achieved through a variety of communication methods, and the scalability is strong.

[0026] Reference Figure 1 The embodiment of the present invention provides a battery pack operation and maintenance monitoring method based on artificial intelligence, comprising the following steps:

[0027] S100. Configure the online monitoring equipment through the parameter configuration interface of the host computer, including configuring the basic information of the battery, the upper and lower limits of the single cell voltage, the upper and lower limits of the internal resistance, the upper temperature limit, the upper charging current limit, the upper and lower limits of the battery pack voltage, the battery pack voltage difference alarm threshold, the SOC lower limit, the SOH lower limit, the measurement of the starting current value, the group voltage value mode, the balancing control mode, the balancing degree threshold, the balancing time, the internal resistance measurement mode, the internal resistance measurement cycle and the alarm method.

[0028] S200, the online monitoring equipment collects the operating data and ambient temperature of the battery pack in real time, determines whether to issue an alarm based on the configuration information, and determines whether to perform active balancing, activation operations and online desulfurization based on the operating data; the operating data of this embodiment includes: the voltage, internal resistance, capacity, electromotive force of the battery pack and single cells, and the charge and discharge current of the battery pack.

[0029] S300. In response to the operation requested by the host computer, the online monitoring device sends the collected real-time or historical operating data and ambient temperature to the host computer through the communication port; the host computer analyzes and processes the operating data and ambient temperature, obtains visual information and displays it on the visual interface of the host computer; the visual information includes: real-time status information of each battery group, real-time status information of each single cell, curves of the battery group voltage and single cell voltage during discharge, historical curves, real-time alarms and historical alarm information.

[0030] S400, obtaining different battery pack pressure difference alarm thresholds and corresponding battery pack pressure difference historical alarm frequencies, and corresponding battery service life as training data, and training to obtain a first neural network model with the battery pack pressure difference alarm threshold and the battery pack pressure difference alarm frequency as input and the battery service life as output; setting the maximum value of the alarm frequency and the minimum value of the battery service life as constraints, and finding the maximum value of the objective function based on the first neural network model and the constraints to obtain the optimal battery pack pressure difference alarm threshold; configuring the online monitoring device based on the optimal battery pack pressure difference alarm threshold; wherein the objective function is s=T / W, T is the battery service life, and W is the battery pack pressure difference alarm frequency.

[0031] The embodiment of the present invention obtains the optimal battery voltage difference alarm threshold by analyzing the frequency and threshold of the battery voltage difference historical alarm information and the corresponding battery service life, thereby reducing redundant alarm information while ensuring the battery service life and saving manpower.

[0032] Reference Figure 3 , Figure 4 , Figure 5 and Figure 6 This embodiment monitors the single cell voltage and internal resistance temperature of the battery pack in real time, and has functions such as real-time data browsing, operation curve, alarm recording, statistics, etc., which are convenient for users to analyze and view. It can directly set the battery voltage high and low limit alarm, set the lagging battery equalization voltage high and low limit, manually perform equalization operation, automatic equalization operation, equalization strategy, equalization conditions, and has the battery voltage over-voltage and under-voltage alarm function.

[0033] In some embodiments, the host computer analyzes and processes the operating data and the ambient temperature to obtain the visualization information including: obtaining the basic information of the single cells in each battery pack, including the type, model, batch, charge and discharge times, and charge and discharge time, and grouping the single cells according to the basic information; based on multiple SOH prediction models, obtaining the real-time status information of several single cells in different groups to predict the SOH of the single cells, and obtaining several SOH prediction values ​​corresponding to several single cells; according to the SOH prediction values ​​and the actual SOH values ​​of several single cells, obtaining the prediction accuracy of each SOH prediction model corresponding to each group of single cells, and selecting the highest prediction accuracy of each group of cells for the prediction model; The corresponding SOH prediction model is used as the optimal prediction model for the grouped batteries; SOH prediction is performed on the single cells of each battery group. When the performance consistency of the battery group is qualified, any single cell in the battery group is selected, and the optimal prediction model is obtained according to the grouping information, and the SOH prediction value of each single cell in the battery group is obtained based on the optimal prediction model and displayed on the visualization interface; when the performance consistency of the battery group is unqualified, the grouping information of each single cell in the battery group and the corresponding optimal prediction model are obtained, and the SOH prediction value of each single cell is obtained based on the corresponding optimal prediction model and displayed on the visualization interface; wherein the actual SOH value is obtained through charge and discharge tests.

[0034] In some embodiments, the above-mentioned SOH prediction model includes several prediction models obtained based on machine learning algorithms, wherein the machine learning algorithms include Gaussian process regression, support vector machine and random forest; the input of the SOH prediction model includes any one or more of the following: voltage, current, internal resistance and temperature of the single cell.

[0035] In some embodiments, the host computer analyzes and processes the operating data and the ambient temperature to obtain visualization information including: analyzing the received operating data based on the limit values ​​in the configuration information, and displaying the data that exceeds the limit values ​​in the operating data displayed on the visualization interface in a color different from the operating data that does not exceed the limit values; wherein the limit values ​​include: upper and lower limits of single cell voltage, upper and lower limits of internal resistance, upper temperature limit, upper charging current limit, upper and lower limits of battery pack voltage, battery pack voltage difference alarm threshold, SOC lower limit and SOH lower limit.

[0036] In some embodiments, the method of the embodiment of the present invention also includes: obtaining historical alarm information of eliminated aging batteries, training a neural network model based on the historical alarm information, and obtaining a second neural network model; the second neural network model takes the historical alarm information as input and the remaining service life as output; the historical alarm information includes the number and value of exceeding the upper and lower limits of voltage, the number and value of exceeding the upper and lower limits of internal resistance, the number and value of exceeding the upper temperature limit, the value of exceeding the SOC lower limit, the value of exceeding the SOH lower limit, and the time of issuing the alarm information; based on the second neural network model, the remaining life of the single cells of each battery pack is predicted, and the predicted values ​​obtained are displayed through a visual interface.

[0037] In some embodiments, the method of the embodiment of the present invention also includes: when the internal resistance measurement mode is manual measurement, when the host computer issues a real-time internal resistance measurement request, locating the manager through the battery pack and single cell identification information of the real-time internal resistance measurement request; and sending the real-time internal resistance measurement request information to the manager's mobile device.

[0038] The embodiment of the present invention realizes the network control of internal resistance test, the unified network management of online batteries, effectively improves the maintenance efficiency, timely discovers the hidden troubles, and reduces the maintenance of batteries; it can fully and dynamically sense the voltage, internal resistance, temperature changes and change trends of each battery, and on this basis, timely eliminate the sulfation of the battery found to be sulfided through online desulfurization technology, slow down or reduce the internal resistance of the battery; for the battery that has not yet deteriorated, the active balancing technology is used to improve the consistency of each battery in the whole battery group, so as to slow down and inhibit the degradation rate of the battery. Through online desulfurization and active balancing technology, the battery group can be automatically operated and maintained, and finally the service life of the battery is extended.

[0039] Reference Figure 2 Another aspect of the present invention provides a battery pack operation and maintenance monitoring system based on artificial intelligence, including:

[0040] The collection layer includes several online monitoring devices for collecting battery pack information.

[0041] The communication layer is used to connect the online monitoring equipment and the master station layer through switches and Ethernet.

[0042] The main station layer includes several servers, which are used to analyze and manage battery group information.

[0043] Among them, the online monitoring equipment includes: acquisition port, internal resistance measurement switch, communication port and output port. The acquisition port includes: battery voltage acquisition port, current sensor Hall signal acquisition port and ambient temperature acquisition port.

[0044] This embodiment realizes judging whether various parameters are normal during the operation of the battery by online monitoring of the values ​​of the voltage, current, internal resistance, capacity and ambient temperature of each battery. When the consistency of the battery group deteriorates, the battery online monitoring and management system will automatically start the online balancing function to balance the batteries and can perform online activation on batteries with poor performance, which can effectively prevent the battery group from being overcharged or undercharged, so that the battery group can operate in the best state with consistent performance.

[0045] The online monitoring equipment of this embodiment mainly realizes the following functions: 1. Online detection of the voltage of the battery pack and the single cell; 2. Online detection of the internal resistance of the battery pack and the single cell; 3. Online measurement of the capacity of the battery pack and the single cell; 4. Online measurement of the electromotive force of the battery pack and the single cell; 5. Detection of the charging and discharging current of the battery pack; 6. Capacity measurement of the ambient temperature of the battery pack; 7. Capacity of undervoltage alarm of the battery voltage; 8. Active battery pack balancing control, which can independently adjust the charge and discharge of the single cell; 9. Online activation of the single cell with poor performance; 10. Capacity monitoring and management of the relevant data of the battery pack discharge capacity verification; 11. Provide curve display of the whole group voltage and single cell voltage of the battery during the discharge process; 12. Capacity of saving relevant historical data and displaying data curve information; 13. Capable of communicating with the host computer through RS485 or network port.

[0046] Although specific embodiments are described herein, those of ordinary skill in the art will recognize that many other modifications or alternative embodiments are also within the scope of the present disclosure. For example, any of the functions and / or processing capabilities described in conjunction with a particular device or component may be performed by any other device or component. In addition, although various exemplary implementations and architectures have been described according to embodiments of the present disclosure, those of ordinary skill in the art will recognize that many other modifications to the exemplary implementations and architectures described herein are also within the scope of the present disclosure.

[0047] It will be appreciated by those skilled in the art that all or some of the steps and systems in the disclosed method above may be implemented as software, firmware, hardware and appropriate combinations thereof. Some physical components or all physical components may be implemented as software executed by a processor, such as a central processing unit, a digital signal processor or a microprocessor, or may be implemented as hardware, or may be implemented as an integrated circuit, such as an application specific integrated circuit. Such software may be distributed on a computer-readable medium, which may include a computer storage medium (or a non-transitory medium) and a communication medium (or a temporary medium). As known to those skilled in the art, the term computer storage medium includes volatile and non-volatile, removable and non-removable media implemented in any method or technology for storing information (such as computer-readable instructions, data structures, program modules or other data). Computer storage media include, but are not limited to, RAM, ROM, EEPROM, flash memory or other memory technologies, CD-ROM, digital versatile disks (DVD) or other optical disk storage, magnetic cassettes, magnetic tapes, disk storage or other magnetic storage devices, or any other medium that may be used to store desired information and may be accessed by a computer. Furthermore, it is well known to those skilled in the art that communication media typically embodies computer readable instructions, data structures, program modules, or other data in a modulated data signal such as a carrier wave or other transport mechanism, and may include any information delivery media.

[0048] The above is a specific description of the preferred implementation of the present application, but the present application is not limited to the above-mentioned implementation mode. Technical personnel familiar with the field can also make various equivalent deformations or substitutions without violating the spirit of the present application. These equivalent deformations or substitutions are all included in the scope defined by the claims of the present application.

Claims

1. A battery pack operation and maintenance monitoring method based on artificial intelligence, characterized in that: The following steps are involved: S100, configuring the online monitoring equipment through the parameter configuration interface of the host computer, including configuring the basic information of the battery, the upper and lower limits of the single cell voltage, the upper and lower limits of the internal resistance, the upper temperature limit, the upper charging current limit, the upper and lower limits of the battery pack voltage, the battery pack voltage difference alarm threshold, the SOC lower limit, the SOH lower limit, the measured starting current value, the group voltage value mode, the balancing control mode, the balancing degree threshold, the balancing time, the internal resistance measurement mode, the internal resistance measurement cycle and the alarm mode; S200, the online monitoring device collects the operating data and ambient temperature of the battery pack in real time, determines whether to issue an alarm based on the configuration information, and determines whether to perform active balancing, activation operation and online desulfurization based on the operating data; the operating data includes: the voltage, internal resistance, capacity, electromotive force of the battery pack and the single cell, and the charge and discharge current of the battery pack; S300, in response to the host computer's request for operation, the online monitoring device sends the collected real-time or historical operating data and ambient temperature to the host computer through the communication port; the host computer analyzes and processes the operating data and ambient temperature, obtains visual information and displays it on the host computer's visual interface; the visual information includes: real-time status information of each battery pack, real-time status information of each single cell, curves of the battery pack voltage and single cell voltage during discharge, historical curves, real-time alarms and historical alarm information; S400, obtaining different battery pack pressure difference alarm thresholds and corresponding battery pack pressure difference historical alarm frequencies, and corresponding battery service life as training data, and training to obtain a first neural network model with the battery pack pressure difference alarm threshold and the battery pack pressure difference alarm frequency as input and the battery service life as output; setting the maximum value of the alarm frequency and the minimum value of the battery service life as constraints, and finding the maximum value of the objective function based on the first neural network model and the constraints to obtain the optimal battery pack pressure difference alarm threshold; configuring the online monitoring device based on the optimal battery pack pressure difference alarm threshold; wherein the objective function is s=T / W, T is the battery service life, and W is the battery pack pressure difference alarm frequency.

2. The method for battery pack operation and maintenance monitoring based on artificial intelligence according to claim 1, characterized in that: The host computer analyzes and processes the operating data and the ambient temperature to obtain visual information including: Obtaining basic information of the single cells in each battery pack, including type, model, batch, charge and discharge times, and charge and discharge time, and grouping the single cells according to the basic information; Based on multiple SOH prediction models, real-time status information of several monomer batteries in different groups is obtained to predict the SOH of the monomer batteries, and several SOH prediction values ​​corresponding to the several monomer batteries are obtained; According to the SOH predicted values ​​and actual SOH values ​​of several single cells, the prediction accuracy of each SOH prediction model corresponding to each group of single cells is obtained, and the SOH prediction model corresponding to the highest prediction accuracy of each group of cells is selected as the best prediction model for the grouped cells; Perform SOH prediction on the single cells of each battery pack. When the performance consistency of the battery pack is qualified, select any single cell in the battery pack, obtain the best prediction model according to the grouping information, and obtain the SOH prediction value of each single cell of the battery pack based on the best prediction model and display it on the visual interface; When the performance consistency of the battery pack is unqualified, the grouping information of each single cell in the battery pack and the corresponding optimal prediction model are obtained, and the SOH prediction value of each single cell is obtained based on the corresponding optimal prediction model and displayed on the visual interface; The actual SOH value is obtained through charge and discharge testing.

3. The method for battery pack operation and maintenance monitoring based on artificial intelligence according to claim 2 is characterized in that: The SOH prediction model includes several prediction models obtained based on machine learning algorithms, including Gaussian process regression, support vector machine and random forest; the input of the SOH prediction model includes any one or more of the following: voltage, current, internal resistance and temperature of a single cell.

4. The method for battery pack operation and maintenance monitoring based on artificial intelligence according to claim 1, characterized in that: The host computer analyzes and processes the operating data and the ambient temperature to obtain visual information including: The received operating data is analyzed based on the limit values ​​in the configuration information, and the data exceeding the limit values ​​in the operating data displayed on the visualization interface is displayed in a color different from the operating data within the limit values; the limit values ​​include: upper and lower limits of single cell voltage, upper and lower limits of internal resistance, upper temperature limit, upper charging current limit, upper and lower limits of battery pack voltage, battery pack voltage difference alarm threshold, SOC lower limit and SOH lower limit.

5. The method for battery pack operation and maintenance monitoring based on artificial intelligence according to claim 1, characterized in that: The method further comprises: Obtain historical alarm information of the eliminated aging battery, and train the neural network model based on the historical alarm information to obtain a second neural network model; the second neural network model takes the historical alarm information as input and the remaining service life as output; the historical alarm information includes the number and value of exceeding the upper and lower limits of voltage, the number and value of exceeding the upper and lower limits of internal resistance, the number and value of exceeding the upper temperature limit, the value of exceeding the lower limit of SOC, the value of exceeding the lower limit of SOH, and the time of issuing the alarm information; The remaining life of the single cells of each battery pack is predicted based on the second neural network model, and the obtained predicted values ​​are displayed through a visual interface.

6. The method for battery pack operation and maintenance monitoring based on artificial intelligence according to claim 1, characterized in that: The method further comprises: When the internal resistance measurement mode is manual measurement, when the host computer issues a real-time internal resistance measurement request, the manager is located through the battery pack and single cell identification information of the real-time internal resistance measurement request; and the real-time internal resistance measurement request information is sent to the mobile device of the manager.

7. An artificial intelligence-based battery pack operation and maintenance monitoring system, used to implement the method described in any one of claims 1 to 6, characterized in that: include: The collection layer includes several online monitoring devices for collecting battery pack information; The communication layer is used to connect the online monitoring equipment and the master station layer through switches and Ethernet; The main station layer includes several servers, which are used to analyze and manage battery group information.

8. The artificial intelligence-based battery pack operation and maintenance monitoring system according to claim 7 is characterized in that: The online monitoring device comprises: a collection port, an internal resistance measurement switch, a communication port and an output port.

9. The artificial intelligence-based battery pack operation and maintenance monitoring system according to claim 8, characterized in that: The acquisition ports include: a battery voltage acquisition port, a current sensor Hall signal acquisition port and an ambient temperature acquisition port.

10. The artificial intelligence-based battery pack operation and maintenance monitoring system according to claim 8, characterized in that: The communication port is an RS485 interface.