CN116930764A - Fault diagnosis and risk prediction method for lithium battery energy storage system integrating power electronics - Google Patents

Abstract

The application provides a fault diagnosis and danger prediction method of a lithium battery energy storage system integrating power electronics, which comprises the following steps of S1, constructing a high-perceptibility battery; s2, performing micro-short circuit fault simulation of an actual battery unit, extracting fault expression characteristics, and determining a nondestructive diagnosis threshold; s3, performing an overcharge and overdischarge experiment on the high-perceptibility battery to cause artificial micro-short circuit, and performing fault diagnosis in real time by using a nondestructive fault diagnosis mode; s4, monitoring internal thermal quantity information and characteristic gas information of the high-perceptibility battery while causing the high-perceptibility battery to fail, and completing the selection of a temperature deviation threshold value and a gas concentration threshold value; and S5, monitoring a high-perceptibility battery fault evolution rule in the whole process, dividing a fault emergency program, and constructing a fault database. According to the application, the battery fault evolution process is monitored in real time according to the high-perceptibility battery, so that the subsequent conventional battery fault evolution process prediction is realized.

Description

Fault diagnosis and risk prediction method for lithium battery energy storage system integrating power electronics

Technical field

The present invention relates to the field of battery management technology, and specifically to fault diagnosis and risk prediction methods, media and terminals for lithium battery energy storage systems integrating power electronics.

Background technique

The evolution mechanism of battery failure is mainly divided into two categories. One is battery failure caused by external emergencies, and the other is failure caused by battery performance degradation that reduces battery reliability to a certain level. The evolution of faults caused by emergencies includes abnormal reactions of active components inside the battery caused by vibration, extrusion, puncture, external short circuit, high temperature, etc., which in turn leads to overheating and further triggers thermal runaway. This type of emergency failure should be considered in the design stage and its probability of occurrence should be minimized. Failures caused by reduced battery reliability due to battery performance degradation require fault management and control through fault diagnosis and hazard prediction methods.

During the operation of the battery energy storage system, the battery energy storage system will basically not encounter extreme working conditions such as puncture, extrusion, drop, seawater immersion, etc. Its main fault sources include two types. One type comes from power grid surge overvoltage, external short circuit, insulation damage, etc. Failures caused by additional stress shocks should be avoided or emergency protection should be implemented through reasonable design of the power converter and protection circuit; the second type of failure mainly comes from the destruction of the diaphragm caused by dendrite growth inside the battery and other reasons, forming a micro short circuit. The impact of micro-short circuit faults on battery safety is a gradual process. If the micro-short-circuit unit can be found and located in time, sufficient processing time can usually be left.

Search found:

The Chinese patent application with the publication number CN115219912A "A method and system for early fault diagnosis and safety advance warning of energy storage batteries" discloses an early fault diagnosis and safety advance warning method and system for energy storage batteries, including: obtaining storage battery The voltage signal of the energy battery; based on the voltage signal, multiple fault diagnosis methods are used to perform preliminary fault diagnosis results; the preliminary fault diagnosis results of the multiple fault diagnosis methods are mapped into a eigenvalue sequence; the eigenvalue sequence is subjected to a convex function Process and add bias; integrate the processed eigenvalue sequence with respect to time to obtain the early fault diagnosis results of the energy storage battery. It maps the fault diagnosis results into a sequence of eigenvalues, performs convex function processing and adds bias, and performs time integration. The calculation is complex and has no physical meaning.

The Chinese patent application with the publication number CN115327438A is "An Energy Storage Battery Pack Short-Circuit Fault Diagnosis Method, Battery Management Method and System". This invention discloses an energy storage battery pack short-circuit fault diagnosis method, which includes the following steps: Obtaining the requirements for fault detection Base parameters; standardize the maximum base eigenvalue vector, set the significance level, obtain the short-circuit fault detection threshold, and then obtain the minimum detectable fault estimate of the short-circuit fault, and determine the minimum detectable short-circuit resistance; based on the interleaved voltage quantity obtained in real time Measure the differential voltage timing vector of the topological energy storage battery pack, construct a standardized differential voltage measurement timing matrix, and calculate the corresponding real-time short-circuit fault detection index; when the short-circuit fault detection index exceeds the fault detection threshold, calculate the differential voltage channel contribution of the fault , determine the location of the short circuit fault. It uses the reference differential voltage mean vector and the reference differential voltage standard deviation diagonal matrix of the reference differential interleaved measurement voltage timing submatrix based on the interleaved voltage measurement topology. The calculation is complex and has no physical meaning.

The Chinese patent application with publication number CN107422266B is "A fault diagnosis method and device for a large-capacity battery energy storage system". This invention relates to a fault diagnosis method and device for a large-capacity battery energy storage system, including obtaining the information of the battery energy storage system. Data to be diagnosed; input the data to be diagnosed as a test sample into a pre-built BP neural network model for fault diagnosis, and output the fault diagnosis results. This application uses a neural network, which has high time complexity and space complexity and has no physical meaning.

The Chinese patent application with publication number CN115524613A "Method, device, system and storage medium for internal short circuit fault diagnosis of energy storage batteries" discloses a method, device, system and storage medium for internal short circuit fault diagnosis of energy storage batteries. The method includes: obtaining a first electrical characteristic of an energy storage battery, predicting whether a suspected internal short-circuit fault occurs in the energy storage battery based on the first electrical characteristic, and calculating a set of current heat generation and internal material temperature values of the energy storage battery. , and determine the heat accumulation position of the energy storage battery based on the current heat generation and the internal material temperature value set; determine whether an internal short-circuit fault occurs in the energy storage battery based on the suspected internal short circuit fault and the heat accumulation location. Short circuit fault. It does not discuss how to obtain the fault threshold.

Contents of the invention

In view of the deficiencies in the prior art, the purpose of the present invention is to provide a fault diagnosis and risk prediction method for a lithium battery energy storage system that integrates power electronics.

According to one aspect of the present invention, a fault diagnosis and risk prediction method for a lithium battery energy storage system integrating power electronics is provided, including:

Constructing a high-perceivability battery;

Carry out micro-short-circuit fault simulation of actual battery units, extract fault performance characteristics, and determine non-destructive diagnosis thresholds;

Conduct overcharge and over-discharge experiments on the high-sensitivity battery to cause artificial micro-short circuits, and use non-destructive fault diagnosis methods combined with the non-destructive diagnostic threshold to perform fault diagnosis in real time;

While triggering the high-perceivability battery failure, monitor the internal thermal quantity information and characteristic gas information of the high-perceivability battery, and complete the selection of the temperature deviation threshold and the gas concentration threshold;

Based on the non-destructive diagnosis threshold, the temperature deviation threshold and the gas concentration threshold, the highly perceptible battery fault evolution rules are monitored throughout the process, fault emergency procedures are divided, and a fault database is constructed.

Preferably, the structure of the high-perceivability battery includes:

Select energy storage system battery cell and module specifications;

Make high-perceivability electric cores and complete packaging, embed temperature sensors, gas sensors and pressure sensors in the electric cores, and set up transmission media to realize internal information transmission of the electric cores;

Perform cycle testing on the packaged high-sensitivity battery cells and the selected energy storage system battery cells to detect battery parameters, where the battery parameters include capacity, internal resistance, open circuit voltage and thermal resistance;

Optimize and iterate the high-perceivability battery cell to make it consistent with the battery parameter curve of the selected energy storage system battery cell;

According to the module specifications of the energy storage system battery, the high-sensitivity cells are assembled and modularly packaged; temperature sensors, gas sensors and pressure sensors are pre-embedded in the module package, and a transmission medium is provided to implement the module Internal information transfer.

Preferably, the method of simulating micro-short circuit faults of actual battery cells, extracting fault performance characteristics, and determining non-destructive diagnosis thresholds includes:

Select conventional battery modules with the same specifications as the battery cells and modules of the energy storage system;

A micro-short-circuit battery module is formed by adding multiple short-circuit resistors with different resistance values to the cell tabs in the conventional battery module;

Control the power interface converter of the energy storage unit to inject a voltage ripple of specific frequency and waveform into the micro-short-circuit battery module to conduct online measurement of battery cell and module level impedance, compare the difference between the measured impedance and the impedance before fault simulation, and complete the impedance information Compare the deviations to obtain the impedance difference, obtain the relationship between the impedance difference and the short-circuit resistor resistance, and determine the micro-short-circuit fault deviation impedance threshold through the relationship;

Control the power interface converter of the energy storage unit to make the micro-short-circuit battery module run in the same working condition as the actual energy storage system, and compare the data reported by the converter and battery management system with the voltage, current and temperature data under normal working conditions to select Arrange the first several quantities from large to small in variation amplitude value as characteristic quantities, obtain the relationship between the characteristic quantities and the resistance of the short-circuit resistor, and determine the voltage and current deviation threshold;

Utilize the electrothermal coupling model of the energy storage unit to construct a digital mirror model of the energy storage system to ensure that the mirror model and the actual physical model operate under the same temperature and current conditions, and monitor the state deviation between the output of the mirror model and the actual physical model. , obtain the relationship between the voltage residual and the short-circuit resistor resistance, and complete the voltage residual threshold selection.

Preferably, the control energy storage unit power conversion interface injects a voltage ripple of specific frequency and waveform into the micro-short-circuit battery module to conduct online measurement of battery cells and module-level impedance, including:

According to the required waveform and frequency of the excitation voltage, the duty cycle disturbance signal of the corresponding waveform and frequency is injected into the PWM control signal of the power interface converter of the energy storage unit, and the micro-short-circuit battery module is injected with a specific frequency and waveform. voltage ripple;

Sampling battery cell and module voltage and current signals;

After Fourier operation is performed on the battery voltage and current sampling signals, the amplitude and phase of the voltage and current sampling signals at different frequencies are obtained;

Calculate the battery impedance according to the battery voltage and current, that is: assuming that the Fourier algorithm is used to extract the battery cell voltage and current signal at the frequency lω _1. The amplitudes are c _{l_U} and c _{l_I} respectively, and the phases are respectively and/> The impedance information of the battery unit at frequency lω ₁ can be calculated:/>

Preferably, the overcharge and over-discharge experiments on the high-perceivability battery cause artificial micro-short circuits, and the non-destructive fault diagnosis method is combined with the non-destructive diagnosis threshold to perform fault diagnosis in real time, including:

Conduct overcharge and over-discharge experiments on the high-perceivability battery, use multiple fault diagnosis methods to conduct real-time monitoring of micro-short circuits, and continuously record the changes in each fault characteristic quantity during the overcharge and over-discharge experiments;

When it is determined based on the non-destructive diagnosis threshold that a micro-short circuit has occurred in the high-sensitivity battery, the overcharge and over-discharge experiment will be stopped, and the high-sensitivity battery will be charged and discharged until its state of charge is in the range of 40% to 60%. .

Preferably, while triggering a high-perceivable battery failure, the internal thermal quantity information and characteristic gas information of the high-perceivable battery are monitored, and the selection of the temperature deviation threshold and the gas concentration threshold is completed, including:

After recording non-destructive fault diagnosis, monitor the internal and external thermal information of the highly perceptible battery, and record the temperature deviation at this time as the fault threshold;

After recording non-destructive fault diagnosis, monitor the concentration changes of characteristic gases inside and outside the highly perceptible battery, and record the concentration at this time as the fault threshold.

Preferably, the entire process monitors the evolution rules of highly perceptible battery faults, divides fault emergency procedures, and builds a fault database, including:

By running the highly perceptible battery after the fault within the normal SOC range, the changes in fault characteristic quantities of various methods are continuously recorded; at the same time, a control test is set up to run the high perceptible battery after the fault under derated working conditions. Continuously record the changes in fault characteristics of various methods;

Compare the fault evolution rules under normal operation and derated operation, study the impact of operation at different powers on the evolution of battery module micro-short circuits, and divide the fault evolution stages and fault emergency levels based on the research results to construct a fault database.

According to a second aspect of the present invention, a lithium battery energy storage system fault diagnosis and risk prediction system integrating power electronics is provided, including:

Sensing module, which constructs a highly perceptible battery;

Non-destructive threshold module, which simulates micro-short circuit faults of actual battery cells, extracts fault performance characteristics, and determines non-destructive diagnosis thresholds;

A micro-destructive test module, which conducts overcharge and over-discharge experiments on the high-perceivability battery to cause artificial micro-short circuits, and uses non-destructive fault diagnosis methods combined with the non-destructive diagnostic threshold to perform fault diagnosis in real time;

The minimal loss threshold module, while triggering the high-perceivability battery failure, monitors the internal thermal quantity information and characteristic gas information of the high-perceivability battery, and completes the selection of the temperature deviation threshold and the gas concentration threshold;

Database module, which monitors the evolution rules of highly perceptible battery faults throughout the process, divides fault emergency procedures, and builds a fault database.

According to a third aspect of the present invention, a terminal is provided, including a memory, a processor, and a computer program stored in the memory and executable on the processor. When the processor executes the program, it can be used to execute the A fault diagnosis and risk prediction method for a lithium battery energy storage system integrating power electronics, or running the fault diagnosis and risk prediction system for a lithium battery energy storage system integrating power electronics.

According to a fourth aspect of the present invention, there is provided a computer-readable storage medium on which a computer program is stored. When executed by a processor, the program can be used to perform fault diagnosis and analysis of the lithium battery energy storage system integrating power electronics. Hazard prediction method, or, running the fault diagnosis and hazard prediction system of the lithium battery energy storage system integrating power electronics.

Compared with the existing technology, it has at least one of the following beneficial effects:

1. The battery reliability is reduced due to battery performance degradation, which in turn causes a longer failure process. The fault evolution law is not closely connected with the information on the key characteristics of external faults. The fault diagnosis and risk prediction method and system of the lithium battery energy storage system integrating power electronics in the embodiment of the present invention monitors the battery fault evolution process in real time based on the highly perceptible battery, and realizes the subsequent conventional battery fault evolution process.

2. The fault diagnosis and risk prediction method and system of the lithium battery energy storage system integrating power electronics in the embodiment of the present invention evaluates the battery current, voltage characteristics, electrochemical impedance, internal temperature and battery gas evolution under normal and fault conditions. Characteristic differences, construct two-level fault diagnosis methods of non-destructive and minimally damaging; establish a fault diagnosis database containing different fault characteristic quantities, and establish the judgment criteria for collaborative judgment using different fault characteristic quantities through research, so as to facilitate different faults in subsequent system operations Fault identification when there are differences in feature diagnosis.

3. The fault diagnosis and risk prediction method and system of the lithium battery energy storage system integrating power electronics in the embodiment of the present invention uses the established multi-dimensional coupling model of the battery to construct a digital image of the energy storage system, uses residual analysis, and combines multi-level faults Diagnostic methods realize model-based fault diagnosis and hazard warning, improve non-destructive diagnosis strategies, improve fault identification speed, and leave sufficient window period for fault post-processing.

Description of the drawings

Other features, objects and advantages of the present invention will become more apparent by reading the detailed description of the non-limiting embodiments with reference to the following drawings:

Figure 1 is a schematic overall flow diagram of a fault diagnosis and risk prediction method for a lithium battery energy storage system integrating power electronics according to an embodiment of the present invention;

Figure 2 is a schematic flow chart of constructing a high-sensitivity battery according to a preferred embodiment of the present invention;

Figure 3 is a schematic diagram of the online measurement process of battery cells and module level impedance according to a preferred embodiment of the present invention;

Figure 4 is a schematic structural diagram of online measurement of battery cell and module level impedance according to a preferred embodiment of the present invention;

Figure 5 is a schematic diagram of non-destructive diagnosis and lossy diagnosis fault characteristics according to a preferred embodiment of the present invention.

Detailed ways

The present invention will be described in detail below with reference to specific embodiments. The following examples will help those skilled in the art to further understand the present invention, but do not limit the present invention in any form. It should be noted that, for those of ordinary skill in the art, several modifications and improvements can be made without departing from the concept of the present invention. These all belong to the protection scope of the present invention.

Figure 1 is a schematic flowchart of the overall fault diagnosis and risk prediction method for a lithium battery energy storage system integrating power electronics in one embodiment of the present invention, which specifically includes the following steps:

S1. Construct a high-sensitivity battery;

S2. Carry out micro-short circuit fault simulation of actual battery units, extract fault performance characteristics, and determine non-destructive diagnosis thresholds;

S3. Conduct overcharge and over-discharge experiments on high-sensitivity batteries to cause artificial micro-short circuits, and use non-destructive fault diagnosis methods combined with non-destructive diagnostic thresholds to conduct real-time fault diagnosis;

S4. While causing a highly perceptible battery failure, monitor the internal thermal quantity information and characteristic gas information of the highly perceptible battery, and complete the selection of the temperature deviation threshold and gas concentration threshold;

S5. Based on the non-destructive diagnosis threshold, temperature deviation threshold and gas concentration threshold, monitor the evolution pattern of highly perceptible battery faults throughout the process, classify fault urgency, and build a fault database.

In this embodiment, high perceptibility means that in addition to the battery current, battery terminal voltage, and battery shell temperature (including battery tabs whose temperature can be measured) that can be measured by conventional batteries, additional measurable items including but not Limited to the following physical/thermal quantities: battery internal temperature, cell tab potential, battery cell internal pressure, battery release gas concentration, etc. Since the above quantities cannot be measured by conventional batteries, for this reason, in the invention, it will be possible Batteries that measure these quantities are defined as high-perceivable batteries.

The actual battery unit is a battery with the same specifications as the high-sensitivity battery, such as volume, capacity, current rate and other related specifications.

"Non-destructive" refers to a way in which diagnosis can be made without destructive testing. "Minimal damage" refers to the use of overcharge and over-discharge, which has actually caused internal damage to the battery. However, since the battery has not been disassembled, the degree of damage is limited. In order to distinguish between non-destructive diagnosis/testing, it is called " "minor loss"

In a preferred embodiment of the present invention, executing S1, selecting battery cell and module specifications for the energy storage system, and designing high-sensitivity battery cells and battery modules with the same specifications may include the following steps, as shown in Figure 2:

S1.1: Use a full-spectrum battery research and testing platform to conduct trial production of battery cells, embed temperature, gas, pressure and other sensors in it, and realize internal information transmission of battery cells through transmission media such as inert materials and optical fibers;

S1.2: After completing the battery cell packaging, perform a cycle test on the high-sensitivity battery cells and battery cells of the same specification to detect battery parameters such as battery capacity, internal resistance, open circuit voltage, and thermal resistance;

S1.3: Through iterative optimization of the cell structure, the characteristics of the high-perceivable cell and the actually used energy storage cell in terms of capacity, internal resistance, and open circuit voltage curve are basically consistent;

S1.4: On the basis of completing the high-perceptibility battery core, according to the selected battery module specifications, the high-perception battery core is assembled and modularly packaged, and smart sensors are also pre-embedded before module packaging (actually The sensor is placed outside the battery core, and then multiple battery cells are connected in series and parallel and then packaged into modules), and the internal information of the battery is transmitted through transmission media such as inert materials and optical fibers.

The highly perceptible battery obtained in this embodiment can monitor the battery fault evolution process in real time, which is helpful to realize the subsequent conventional battery fault evolution process.

In a preferred embodiment of the present invention, executing S2 may include the following steps:

S2.1: By selecting a conventional battery module (a conventional battery module is a battery module that corresponds to the special battery module made in S1.4 without any other operations. Pay attention to the battery module in this conventional battery module at this time. Multiple external short-circuit resistors (1Ω/10Ω/100Ω) with different resistance values are added to the cell tabs (the core is also the specification selected in S1) to simulate the micro-short circuit condition of the energy storage unit;

S2.2: Control the power interface converter of the energy storage unit to inject voltage ripple of specific frequency and waveform into the micro-short-circuit battery module composed of S2.1, conduct online measurement of battery cell and module-level impedance, and compare the measured impedance with that before fault simulation The difference in impedance, complete the impedance information deviation comparison, conduct relevant research on the relationship between the impedance difference and the short-circuit resistance value, and determine the micro-short-circuit fault deviation impedance threshold; the specific frequency refers to one or several frequencies within the range of 0.1 to 1000Hz. Optional, frequency is 0.1Hz, 0.2Hz, 0.5Hz, 1Hz, 2Hz, 5Hz, 10Hz, 20Hz, 50Hz, 100Hz, 200Hz, 500Hz, 1000Hz.

Further, the power conversion interface of the energy storage unit is controlled to inject a voltage ripple of specific frequency and waveform into the micro-short-circuit battery module to conduct online measurement of battery cell and module-level impedance. The specific steps include the following steps, as shown in Figure 3:

S2.2.1: According to the required waveform and frequency of the excitation voltage, inject the duty cycle disturbance signal of the corresponding waveform and frequency into the PWM control signal of the power interface converter of the energy storage unit, and inject a specific frequency and frequency into the micro-short-circuit battery module. The voltage ripple of the waveform is shown in Figure 4;

S2.2.2: Sampling battery cell and module battery voltage and current signals;

S2.2.3: After Fourier operation is performed on the battery voltage and current sampling signals, the amplitude and phase of the voltage and current sampling signals at different frequencies are obtained;

S2.2.4: Calculate battery impedance based on battery voltage and current, where the impedance modulus is equal to the ratio of voltage and current amplitude, and the impedance phase angle is equal to the phase difference between voltage and current. For example, suppose the Fourier algorithm is used to extract the battery unit voltage and current. The amplitudes at the signal frequency lω ₁ are c _{l_U} and c _{l_I} respectively, and the phases are respectively and/> The impedance information of the battery unit at frequency lω ₁ can be calculated:/>

S2.3: Control the power interface converter of the energy storage unit to make the micro-short-circuit battery module run in the same working conditions as the actual energy storage system, and compare the data reported by the converter and battery management system with the voltage, current and temperature data under normal working conditions. , select several quantities with obvious changes as characteristic quantities, study the relationship between characteristic quantities and short-circuit resistance, and determine the outlier factor deviation threshold;

S2.4: Use the electrothermal coupling model of the energy storage unit to construct a digital mirror model of the energy storage system to ensure that the mirror model and the actual physical model operate under the same temperature and current conditions, and perform state deviation on the output voltage of the mirror model and the actual physical model Monitor, conduct relevant research on the relationship between voltage residual and short-circuit resistor resistance, and complete the selection of voltage residual threshold.

This embodiment uses the established battery multi-dimensional coupling model to build a digital image of the energy storage system, uses residual analysis, and combines multi-level fault diagnosis methods to achieve model-based fault diagnosis and danger warning, improve the non-destructive diagnosis strategy, and improve the fault identification speed. Leave sufficient window period for troubleshooting.

In a preferred embodiment of the present invention, S3 is executed. The schematic diagram of non-destructive diagnosis is shown in Figure 5, which may include the following steps:

S3.1: Conduct overcharge and over-discharge experiments on battery modules containing highly perceptible cells, and use multiple fault diagnosis methods in S2 to conduct real-time monitoring of micro-short circuits, and continuously record the characteristic quantities of each fault in the overcharge and over-discharge experiments. (referring to changes in impedance deviation, voltage and current deviation, and voltage residual);

S3.2: When it is determined based on the fault diagnosis criteria obtained in S2 that a micro-short circuit has occurred in the battery, stop the overcharge and over-discharge experiment, and charge and discharge the battery module until its state of charge is in the range of 40% to 60%. .

In a preferred embodiment of the present invention, S4 is executed. The schematic diagram of minimal damage diagnosis is shown in Figure 5, which may include the following steps:

S4.1: Record the highly perceptible internal and external thermal information of the battery after non-destructive fault diagnosis, and record the temperature deviation at this time as the fault threshold;

S4.2: Record the concentration changes of characteristic gases inside and outside the highly perceptible battery after non-destructive fault diagnosis, and record the concentration at this time as the fault threshold.

In a preferred embodiment of the present invention, performing step 5 may include the following steps:

S5.1: By operating the post-fault highly perceptible battery within the normal SOC range, continuously record the above five fault characteristic quantities obtained by various methods (impedance information deviation, voltage and current characteristic quantity deviation, actual energy storage system voltage Residual deviation, temperature deviation and gas concentration deviation) changes; at the same time, set up a control test to run the high-sensitivity battery under derated working conditions after a fault, and continuously record the changes in fault characteristic quantities of various methods;

Furthermore, the working condition simulation experiment can be carried out on three high-sensitivity battery modules containing micro-short-circuit cells. One of them adopts non-derating operation mode, and the other two modules adopt derating by 1/3 and derating by 2 respectively. /3 operating mode;

Furthermore, the fault characteristic quantities of various methods include fault deviation impedance, fault deviation voltage and current in S2, voltage residual amount with the mirror model, and battery internal and external temperature deviation and characteristic gas concentration in S4;

S5.2: Compare the fault evolution rules under normal operation and derated operation, study the impact of operation at different powers on the evolution of battery module micro-short circuits, divide the fault evolution stages and fault emergency levels based on the research results, and construct a fault database.

The fault database refers to the judgment criteria for different fault characteristic quantities. Generally, battery current, voltage characteristic quantities, electrochemical impedance, internal temperature and battery gas evolution are judged on the difference of these characteristic quantities. If it exceeds the above-mentioned collected threshold range, Then it is judged that a corresponding fault has occurred. However, in actual situations, it may happen that after individual feature quantities exceed the threshold, other feature quantities have not yet exceeded the threshold. Therefore, it is necessary to establish judgment criteria for collaborative judgment.

The judgment criteria during collaborative judgment can be set according to priority. For example, if characteristic gas is collected, it means that a fault must have occurred. At this time, if other characteristic quantities do not exceed the threshold, the threshold can be considered to be reduced. It can also be judged by accuracy, that is, which fault characteristic quantity is more accurate depends on which one.

The fault database established in this embodiment can be directly used when establishing the same energy storage system, and can also be used to provide reference for other established energy storage systems.

In a preferred embodiment of the present invention, a fault database about the impedance part is provided. The internal short-circuit impedance is >300Ω; 100~300Ω; 10~100Ω; <10Ω. The first one is basically equivalent to normal. The second type can be considered as a minor fault and can be transferred to key monitoring. The third type can be considered as a more serious fault and needs to be derated. The last one means that it is very dangerous and must be removed immediately.

Based on the same inventive concept, in other embodiments of the present invention, a lithium battery energy storage system fault diagnosis and risk prediction system integrating power electronics is provided, including:

Sensing module, which constructs a highly perceptible battery;

Non-destructive threshold module, which simulates micro-short circuit faults of actual battery cells, extracts fault performance characteristics, and determines non-destructive diagnosis thresholds;

A micro-destructive test module, which conducts overcharge and over-discharge experiments on the high-perceivability battery to cause artificial micro-short circuits, and uses non-destructive fault diagnosis methods combined with the non-destructive diagnostic threshold to perform fault diagnosis in real time;

The minimal loss threshold module, while triggering the high-perceivability battery failure, monitors the internal thermal quantity information and characteristic gas information of the high-perceivability battery, and completes the selection of the temperature deviation threshold and the gas concentration threshold;

Database module, which monitors the evolution rules of highly perceptible battery faults throughout the process, divides fault emergency procedures, and builds a fault database.

For details of each module/unit in the above examples of the present invention, reference can be made to the implementation technology of the steps corresponding to the fault diagnosis and risk prediction method of the lithium battery energy storage system based on integrated power electronics in the above embodiments, which will not be described again here.

Based on the same inventive concept, in other embodiments of the present invention, a terminal is provided, including a memory, a processor, and a computer program stored in the memory and executable on the processor. When the processor executes the program It can be used to perform the fault diagnosis and risk prediction method of the lithium battery energy storage system integrated with power electronics, or to run the fault diagnosis and risk prediction system of the lithium battery energy storage system integrated with power electronics.

Based on the same inventive concept, in other embodiments of the present invention, a computer-readable storage medium is provided, on which a computer program is stored. When the program is executed by a processor, it can be used to execute the lithium battery integrating power electronics. Energy storage system fault diagnosis and risk prediction method, or, running the lithium battery energy storage system fault diagnosis and risk prediction system integrating power electronics.

This Embodiment The specific embodiments of the present invention have been described above. It should be understood that the present invention is not limited to the specific embodiments described above. Those skilled in the art can make various variations or modifications within the scope of the claims, which does not affect the essence of the present invention. The above preferred features can be used in any combination as long as they do not conflict with each other.

Claims (10)

1. A lithium battery energy storage system fault diagnosis and risk prediction method integrating power electronics, which is characterized by including: Constructing a high-perceivability battery; Carry out micro-short-circuit fault simulation of actual battery units, extract fault performance characteristics, and determine non-destructive diagnosis thresholds; Conduct overcharge and over-discharge experiments on the high-sensitivity battery to cause artificial micro-short circuits, and use non-destructive fault diagnosis methods combined with the non-destructive diagnostic threshold to perform fault diagnosis in real time; While triggering the high-perceivability battery failure, monitor the internal thermal quantity information and characteristic gas information of the high-perceivability battery, and complete the selection of the temperature deviation threshold and the gas concentration threshold; Based on the non-destructive diagnosis threshold, the temperature deviation threshold and the gas concentration threshold, the highly perceptible battery fault evolution rules are monitored throughout the process, fault emergency procedures are divided, and a fault database is constructed. 2. The fault diagnosis and risk prediction method of the lithium battery energy storage system integrating power electronics according to claim 1, characterized in that the structure of the highly perceptible battery includes: Select energy storage system battery cell and module specifications; Make high-perceivability electric cores and complete packaging, embed temperature sensors, gas sensors and pressure sensors in the electric cores, and set up transmission media to realize the external transmission of internal information of the electric cores; Perform cycle testing on the packaged high-sensitivity cells and the selected energy storage system battery cells to detect battery parameters, which include capacity, internal resistance, open circuit voltage and thermal resistance; Optimize and iterate the high-perceivability battery cell to make it consistent with the battery parameter curve of the selected energy storage system battery cell; According to the module specifications of the energy storage system battery, the high-sensitivity cells are assembled and modularly packaged; temperature sensors, gas sensors and pressure sensors are pre-embedded before module packaging, and transmission media are set to implement the module Internal information is transmitted externally. 3. A method for fault diagnosis and risk prediction of a lithium battery energy storage system integrating power electronics according to claim 1, characterized in that the micro-short circuit fault simulation of the actual battery unit is performed, the fault performance characteristics are extracted, and the non-destructive method is determined. Diagnostic thresholds, including: Select conventional battery modules with the same specifications as the battery cells and modules of the energy storage system; A micro-short-circuit battery module is formed by adding multiple short-circuit resistors with different resistance values to the cell tabs in the conventional battery module; Control the power interface converter of the energy storage unit to inject a voltage ripple of specific frequency and waveform into the micro-short-circuit battery module to conduct online measurement of battery cell and module level impedance, compare the difference between the measured impedance and the impedance before fault simulation, and complete the impedance information Compare the deviations to obtain the impedance difference, obtain the relationship between the impedance difference and the short-circuit resistor resistance, and determine the micro-short-circuit fault deviation impedance threshold through the relationship; Control the power interface converter of the energy storage unit to make the micro-short-circuit battery module run in the same working condition as the actual energy storage system, and compare the data reported by the converter and battery management system with the voltage, current and temperature data under normal working conditions to select Arrange the first several quantities from large to small in variation amplitude value as characteristic quantities, obtain the relationship between the characteristic quantities and the resistance value of the short-circuit resistor, and determine the voltage, current and temperature deviation threshold; Utilize the electrothermal coupling model of the energy storage unit to construct a digital mirror model of the energy storage system to ensure that the mirror model and the actual physical model operate under the same temperature and current conditions, and monitor the state deviation between the output of the mirror model and the actual physical model. , obtain the relationship between the voltage residual and the short-circuit resistor resistance, and complete the voltage residual threshold selection. 4. A method for fault diagnosis and risk prediction of a lithium battery energy storage system integrating power electronics according to claim 3, characterized in that the power conversion interface of the control energy storage unit injects specific frequencies and waveforms into the micro-short-circuit battery module. The voltage ripple of the battery cell and module level impedance can be measured online, including: According to the required waveform and frequency of the excitation voltage, the duty cycle disturbance signal of the corresponding waveform and frequency is injected into the PWM control signal of the power interface converter of the energy storage unit, and the micro-short-circuit battery module is injected with a specific frequency and waveform. voltage ripple; Sampling battery cell and module voltage and current signals; After Fourier operation is performed on the battery voltage and current sampling signals, the amplitude and phase of the voltage and current sampling signals at different frequencies are obtained; Calculate the battery impedance according to the battery voltage and current, that is: assuming that the Fourier algorithm is used to extract the battery cell voltage and current signal at the frequency lω _1. The amplitudes are c _{l_U} and c _{l_I} respectively, and the phases are respectively and/> The impedance information of the battery unit at frequency lω ₁ can be calculated:/> 5. A method for fault diagnosis and risk prediction of a lithium battery energy storage system integrating power electronics according to claim 1, characterized in that the overcharge and overdischarge experiments on the highly perceptible battery cause artificial micro-discharge. Short circuit, and use the non-destructive fault diagnosis method combined with the non-destructive diagnosis threshold to perform fault diagnosis in real time, including: Conduct overcharge and over-discharge experiments on the high-perceivability battery, use multiple fault diagnosis methods to conduct real-time monitoring of micro-short circuits, and continuously record the changes in each fault characteristic quantity during the overcharge and over-discharge experiments; When it is determined based on the non-destructive diagnosis threshold that a micro-short circuit has occurred in the high-sensitivity battery, the overcharge and over-discharge experiment will be stopped, and the high-sensitivity battery will be charged and discharged until its state of charge is in the range of 40% to 60%. . 6. A method for fault diagnosis and risk prediction of a lithium battery energy storage system integrating power electronics according to claim 1, characterized in that while triggering a highly perceptible battery fault, the highly perceptible battery is monitored. Internal thermal quantity information and characteristic gas information are used to complete the selection of temperature deviation thresholds and gas concentration thresholds, including: After recording non-destructive fault diagnosis, monitor the internal and external thermal information of the highly perceptible battery, and record the temperature deviation at this time as the fault threshold; After recording non-destructive fault diagnosis, monitor the concentration changes of characteristic gases inside and outside the highly perceptible battery, and record the concentration at this time as the fault threshold. 7. A method for fault diagnosis and risk prediction of a lithium battery energy storage system integrating power electronics according to claim 1, characterized in that the entire process monitors the highly perceptible battery fault evolution rules and divides fault emergency procedures. , build a fault database, including: By running the highly perceptible battery after the fault within the normal SOC range, the changes in fault characteristic quantities of various methods are continuously recorded; at the same time, a control test is set up to run the high perceptible battery after the fault under derated working conditions. Continuously record the changes in fault characteristics of various methods; Compare the fault evolution rules under normal operation and derated operation, study the impact of operation at different powers on the evolution of battery module micro-short circuits, and divide the fault evolution stages and fault emergency levels based on the research results to construct a fault database. 8. A lithium battery energy storage system fault diagnosis and risk prediction system integrating power electronics, which is characterized by including: Sensing module, which constructs a highly perceptible battery; Non-destructive threshold module, which simulates micro-short circuit faults of actual battery cells, extracts fault performance characteristics, and determines non-destructive diagnosis thresholds; A micro-destructive test module, which conducts overcharge and over-discharge experiments on the high-perceivability battery to cause artificial micro-short circuits, and uses non-destructive fault diagnosis methods combined with the non-destructive diagnostic threshold to perform fault diagnosis in real time; The minimal loss threshold module, while triggering the high-perceivability battery failure, monitors the internal thermal quantity information and characteristic gas information of the high-perceivability battery, and completes the selection of the temperature deviation threshold and the gas concentration threshold; Database module, which monitors the evolution rules of highly perceptible battery faults throughout the process, divides fault emergency procedures, and builds a fault database. 9. A terminal, including a memory, a processor and a computer program stored in the memory and executable on the processor, characterized in that when the processor executes the program, it can be used to execute any of claims 1-7. The method of claim 8, or the system of claim 8. 10. A computer-readable storage medium with a computer program stored thereon, characterized in that, when executed by a processor, the program can be used to perform the method of any one of claims 1-7, or to run the method of claim 1 The system described in 8.