Data-driven method and apparatus for assessing the operational reliability of battery energy storage systems

Abstract

This invention relates to the field of battery energy storage system technology, and particularly to a data-driven method and apparatus for assessing the operational reliability of battery energy storage systems. The method includes: acquiring junction temperature fluctuation information and cycle count of the target energy storage converter to calculate the total component failure probability and wear failure probability, thereby constructing a converter reliability assessment model; acquiring energy storage sample data of the target battery module, voltage and temperature measurement points of all individual cells to calculate the safety warning probability based on data reconstruction error and consistency safety warning probability, thereby constructing a battery energy storage safety warning model; and establishing an operational reliability assessment model based on the converter reliability assessment model and the battery energy storage safety warning model. This solves the problem that existing battery energy storage system operational reliability assessments do not consider the PCS operating cycle count, battery monitoring data, and state imbalances, making it difficult to accurately assess the reliability of battery energy storage systems.

Description

Technical Field

[0001] This invention relates to the field of battery energy storage system operation technology, and in particular to a data-driven method and apparatus for evaluating the operational reliability of battery energy storage systems. Background Technology

[0002] A battery energy storage system (BESS) is a device used to store and release electrical energy. It typically includes an energy storage system, such as a battery pack or supercapacitor, and corresponding power electronic converters. Its design aims to provide efficient and reliable energy storage solutions to meet power demands in various scenarios. BESS offers advantages such as rapid response and flexibility, and has been widely used in balancing grid loads, mitigating fluctuations in renewable energy output, compensating for power outages, and storing emergency power, gradually becoming a mainstream trend in power systems.

[0003] The power conversion system (PCS) is one of the key components in a battery-electrical-electrode system (BESS). The PCS connects to the battery on its DC side and to the low-voltage side of the step-up transformer on its AC side. Each PCS corresponds to one energy storage unit. The PCS can perform bidirectional charging and discharging, operating in active inverter mode to convert DC to AC and supply power to the grid, or in active rectification mode to convert AC to DC and absorb energy from the grid for storage in the battery. In a BESS, the PCS regulates the flow of power and optimizes energy conversion efficiency. It is typically equipped with advanced control system components such as insulated-gate bipolar transistors (IGBTs), MOSFETs, and anti-parallel diodes to adapt to different power demands and provide stable power output. Its main functions include short-circuit protection, DC reverse connection protection, battery overcurrent protection, hardware fault protection, and overload protection. The reliability of a PCS is related to the reliability of its individual components. PCS failures are largely attributed to the failure of the power devices IGBTs and diodes used in the PCS. Therefore, it is necessary to evaluate the reliability of the IGBT modules in the PCS and then study measures and methods to extend the reliability of the PCS.

[0004] Studies have shown that due to variations in junction temperature magnitude and fluctuation intensity, the aluminum bonding wires and solder layers of each power device in a PCS (Power Packaging System) undergo long-term, frequent expansion and contraction. This cumulative effect eventually leads to thermal fatigue failure, which is the root cause of IGBT thermal fatigue failure. Since the operating power of a PCS is directly related to the junction temperature of its power devices, and thus depends on the junction temperature of its components, considering junction temperature fluctuations can accurately reflect the impact of real-time random power variations on PCS component failures. By combining actual energy storage system PCS operating data, this study proposes an energy storage system reliability assessment that takes into account the impact of junction temperature fluctuations. This is of great significance for developing reasonable energy storage system operation and maintenance strategies.

[0005] Given that the impact of the number of PCS operation cycles on its reliability assessment has not been fully studied, this aspect is crucial for evaluating its reliability. Furthermore, with the rapid development of BESS, its inherent safety issues are becoming increasingly prominent. However, battery reliability often involves a time span of tens of hours, therefore it is necessary to explore the potential patterns in its data. The reliability state variable response characteristics mainly include voltage, current, temperature, and state of charge.

[0006] Therefore, there is an urgent need for an operational reliability assessment scheme that can take into account the operating power of the energy storage converter and the impact of battery monitoring data on complex operating conditions and component health levels. Summary of the Invention

[0007] This invention provides a data-driven method and apparatus for evaluating the operational reliability of battery energy storage systems, which addresses the problem that existing reliability assessments of battery energy storage systems do not consider factors such as the number of PCS operating cycles, battery monitoring data, and state imbalances, making it difficult to accurately assess the reliability of battery energy storage systems.

[0008] A first aspect of the present invention provides a data-driven method for evaluating the operational reliability of a battery energy storage system, comprising the following steps:

[0009] The system acquires junction temperature fluctuation information and cycle count of the target energy storage converter to calculate the total component failure probability and wear failure probability of the target energy storage converter based on the junction temperature fluctuation information and cycle count. A converter reliability assessment model is constructed based on the total component failure probability and wear failure probability. Energy storage sample data of the target battery module is acquired to train a bidirectional long short-term memory network using the energy storage sample data, resulting in multiple reconstruction errors. These multiple reconstruction errors are input into a pre-constructed safety warning judgment model for discrimination, to calculate the safety warning probability based on the data reconstruction errors. The voltage and temperature measurement points of all individual cells in the target battery module are acquired to calculate the consistency safety warning probability based on the voltage and temperature measurement points of all individual cells. A battery energy storage safety warning model is constructed based on the safety warning probability based on the data reconstruction errors and the consistency safety warning probability. An operational reliability assessment model is established based on the converter reliability assessment model and the battery energy storage safety warning model, to perform an operational reliability assessment of the target battery energy storage system using the operational reliability assessment model.

[0010] Optionally, the step of acquiring the junction temperature fluctuation information and cycle number of the target energy storage converter, and calculating the total component failure probability and wear failure probability of the target energy storage converter based on the junction temperature fluctuation information, includes:

[0011] Calculate the junction temperature fluctuation information and cycle number of the target energy storage converter under different operating conditions; calculate the total failure probability of the component based on the junction temperature fluctuation information, and calculate the wear failure probability based on the cycle number.

[0012] Optionally, the step of calculating the total failure probability of the component based on the junction temperature fluctuation information and calculating the wear failure probability based on the number of cycles includes:

[0013] The thermal stress factor and temperature cycling factor of each component in the target energy storage converter are calculated based on the junction temperature fluctuation information. Based on the FIDES reliability guidelines, the failure rate of each component is calculated according to its preset basic failure rate parameter, thermal stress factor, and temperature cycling factor. The failure rates of each component are summed to obtain the total component failure probability of the target energy storage converter. Based on the Weibull distribution, the probability density of the aluminum bonding wires and the probability density of the solder layer in the target energy storage converter are calculated. The wear failure probability of the target energy storage converter at a given cycle number is calculated based on the cycle number, the probability density of the aluminum bonding wires, and the probability density of the solder layer.

[0014] Optionally, the step of acquiring energy storage sample data of the target battery module, and using the energy storage sample data to train the bidirectional long short-term memory network to obtain the reconstruction error, includes:

[0015] The energy storage sample data of the target battery module is obtained, and the energy storage sample data is processed to obtain normalized energy storage sample data. The normalized energy storage sample data is used to train a battery energy storage data feature mining model based on a bidirectional long short-term memory network to obtain a reconstructed dataset, and the multiple reconstruction errors are calculated based on the reconstructed dataset.

[0016] Optionally, the step of obtaining the voltage and temperature measurement points of all individual cells in the target battery module, and calculating the consistency safety warning probability based on the voltage and temperature measurement points of all individual cells, includes:

[0017] The system acquires the voltage of all individual cells in the target battery module to calculate the average voltage of the target battery module at a preset time. It then filters the maximum and minimum voltage values ​​of the target battery module at the preset time. Based on the maximum and minimum voltage values, it calculates the root mean square error of extreme voltage. The ratio of the root mean square error of extreme voltage to the average voltage is used as the voltage consistency coefficient. Next, it acquires the temperature measurement points of all individual cells in the target battery module to calculate the maximum and minimum temperature values ​​of the target battery module at the preset time. Based on the maximum and minimum temperature values, it calculates the root mean square error of extreme temperature, and uses this root mean square error as the temperature consistency standard deviation. Finally, it calculates the consistency safety warning probability based on the voltage consistency coefficient and the temperature consistency standard deviation.

[0018] Optionally, the expression for the operational reliability assessment model is:

[0019] P BESS =1-(1-P) fault,PCS (1-P) TR&SOB (1-P) fault,AE )

[0020] Among them, P BESS For integrated security early warning probability, P fault,PCS P represents the target energy storage converter failure rate. TR&SOB P is the battery failure probability based on data reconstruction error and consistency. fault,AE The failure probability of auxiliary equipment in the target battery energy storage system, excluding the target energy storage converter and the target battery module.

[0021] A second aspect of the present invention provides a data-driven battery energy storage system operational reliability assessment device, comprising:

[0022] A first acquisition module is used to acquire junction temperature fluctuation information and cycle count of the target energy storage converter, so as to calculate the total component failure probability and wear failure probability of the target energy storage converter based on the junction temperature fluctuation information and the cycle count; a first construction module is used to construct a converter reliability assessment model based on the total component failure probability and the wear failure probability; a training module is used to acquire energy storage sample data of the target battery module, so as to train a bidirectional long short-term memory network using the energy storage sample data to obtain multiple reconstruction errors; a discrimination module is used to input the multiple reconstruction errors into a pre-constructed safety early warning judgment model for discrimination, so as to... The system includes: a first module for calculating the safety warning probability based on data reconstruction error; a second acquisition module for acquiring the voltage and temperature measurement points of all individual cells in the target battery module, and calculating the consistency safety warning probability based on the voltage and temperature measurement points of all individual cells; a second construction module for constructing a battery energy storage safety warning model based on the safety warning probability based on data reconstruction error and the consistency safety warning probability; and an evaluation module for establishing an operational reliability evaluation model based on the converter reliability evaluation model and the battery energy storage safety warning model, and using the operational reliability evaluation model to evaluate the operational reliability of the target battery energy storage system.

[0023] A third aspect of the present invention provides an electronic device, including: a memory, a processor, and a computer program stored in the memory and executable on the processor, wherein the processor executes the program to implement the data-driven battery energy storage system operation reliability assessment method as described in the above embodiments.

[0024] A fourth aspect of the present invention provides a computer program product, which, when executed by a processor, implements the above-described data-driven battery energy storage system operation reliability assessment method.

[0025] A fifth aspect of the present invention provides a computer-readable storage medium storing a computer program that, when executed by a processor, implements the above-described data-driven battery energy storage system operation reliability assessment method.

[0026] This invention proposes a data-driven method and apparatus for assessing the operational reliability of a battery energy storage system (DESS). It establishes models for the thermal stress factor and junction temperature cycling factor of converter components based on the FIDES reliability guidelines. Then, it extracts the converter junction temperature cycling information using the rainflow counting method and introduces the influence of the number of cycles of the PCS (Power Consolidation System) based on the Weibull probability distribution on its failure probability, thereby constructing a converter reliability assessment model. Addressing battery monitoring data and state imbalance, it proposes a battery cluster reliability method based on reconstruction error calculation and the State of Balance (SOB) state to uncover potential patterns in battery operation data, thus constructing a battery energy storage safety early warning model. Finally, based on the converter reliability assessment model and the battery energy storage safety early warning model, it establishes an assessment model that considers both the influence of PCS pressure, temperature, and cycle count on the DESS's operational reliability and the influence of potential patterns in battery operation data on the DESS's operational reliability. This allows for more accurate operational reliability assessment results, more rational allocation of DESS resources, and maintenance of the overall performance of the DESS.

[0027] Additional aspects and advantages of the invention will be set forth in part in the description which follows, and in part will be obvious from the description, or may be learned by practice of the invention. Attached Figure Description

[0028] The above and / or additional aspects and advantages of the present invention will become apparent and readily understood from the following description of the embodiments taken in conjunction with the accompanying drawings, wherein:

[0029] Figure 1 A flowchart illustrating a data-driven method for evaluating the operational reliability of a battery energy storage system, as provided in an embodiment of the present invention.

[0030] Figure 2 This is a schematic diagram illustrating the specific execution of the data-driven battery energy storage system operation reliability assessment method provided in this embodiment of the invention.

[0031] Figure 3 This is a schematic diagram of a battery module SOB safety early warning framework provided in an embodiment of the present invention;

[0032] Figure 4 This is a block diagram of a data-driven battery energy storage system operation reliability assessment device provided in an embodiment of the present invention.

[0033] Figure 5 This is a schematic diagram of an electronic device provided in an embodiment of the present invention. Detailed Implementation

[0034] Embodiments of the present invention are described in detail below, examples of which are illustrated in the accompanying drawings, wherein the same or similar reference numerals denote the same or similar elements or elements having the same or similar functions throughout. The embodiments described below with reference to the accompanying drawings are exemplary and intended to explain the present invention, and should not be construed as limiting the present invention.

[0035] The following description, with reference to the accompanying drawings, illustrates a data-driven method and apparatus for evaluating the operational reliability of a battery energy storage system according to embodiments of the present invention.

[0036] Figure 1 This is a flowchart illustrating a data-driven method for evaluating the operational reliability of a battery energy storage system, as provided in an embodiment of the present invention.

[0037] like Figure 1 As shown, the data-driven method for assessing the operational reliability of a battery energy storage system includes the following steps:

[0038] In step S101, the junction temperature fluctuation information and cycle number of the target energy storage converter are obtained, so as to calculate the total component failure probability and wear failure probability of the target energy storage converter based on the junction temperature fluctuation information and cycle number.

[0039] In some embodiments, junction temperature fluctuation information and cycle count of the target energy storage converter are obtained to calculate the total component failure probability and wear failure probability of the target energy storage converter based on the junction temperature fluctuation information, including:

[0040] Calculate the junction temperature fluctuation and cycle number of the target energy storage converter under different operating conditions;

[0041] The total failure probability of components is calculated based on junction temperature fluctuation information, and the wear failure probability is calculated based on the number of cycles.

[0042] In some embodiments, the total failure probability of components is calculated based on junction temperature fluctuation information, and the wear failure probability is calculated based on the number of cycles, including:

[0043] Calculate the thermal stress factor and temperature cycling factor of each component in the target energy storage converter based on junction temperature fluctuation information;

[0044] Based on the FIDES reliability guidelines, the failure rate of each component is calculated according to the preset basic failure rate parameters, thermal stress factor and temperature cycling factor of each component.

[0045] The total failure probability of the target energy storage converter is obtained by summing the failure rates of each component.

[0046] Based on the Weibull distribution, the probability density of aluminum bonding lines and solder layer of the target energy storage converter are calculated.

[0047] The wear failure probability of the target energy storage converter under the specified number of cycles is calculated based on the cycle number, the probability density of aluminum bonding wires, and the probability density of solder layers.

[0048] like Figure 2 As shown, in actual implementation, the root causes of PCS failures in the target energy storage converter lie in equipment status, charging and discharging power, and module temperature. Therefore, this embodiment of the invention analyzes the impact of module temperature on the failure rate of PCS components, using thermal stress factor and temperature cycling factor as metrics. According to the FIDES reliability guidelines, the thermal stress factor ρ of components in the PCS under the i-th junction temperature fluctuation state is... TH,i It can be represented as:

[0049]

[0050] Where α is a constant, and its value varies depending on the component, E a =9.89×10 -20 To activate the energy constant, T mean,i Let be the average temperature of the PCS component in state i.

[0051] The component temperature cycling factor ρ of the PCS in the i-th state of junction temperature fluctuation TC,i The model expression is:

[0052]

[0053] Where, n i n is the junction temperature cycle number under component operating state i, n0 is the reference fluctuation cycle number, and t con,i Let θ be the continuous running time in running state i. i Let ΔT be the junction temperature fluctuation cycle time under operating state i. i ΔT0 represents the junction temperature fluctuation amplitude under PCS component operating state i, and ΔT0 represents the reference temperature fluctuation amplitude. max,i This represents the maximum value of the junction temperature fluctuation reached by the PCS component in the i-th state.

[0054] The starting and ending values ​​T0 of the junction temperature fluctuation cycle of the PCS module were calculated using the rainflow counting method. f Using formulas (3)-(5), the fluctuation amplitude ΔT of the junction temperature fluctuation is respectively... i Continuous running time t con Average temperature T mean Maximum temperature T max Estimate the corresponding number of junction temperature fluctuations, n. i Add 1:

[0055]

[0056]

[0057]

[0058] Combining thermal stress factor and temperature cycling factor, and based on the FIDES reliability guidelines, a failure rate model under specific operating conditions can be obtained to calculate the failure rate of each component in the PCS:

[0059]

[0060] Where, λ TH , λ TC These are the basic failure rates corresponding to the thermal stress factor and the temperature cycling factor, respectively, n. TH ρ represents the total number of iterations. pm The impact of PCS component manufacturing quality, ρ pr The impact of reliability quality management and control levels throughout the component lifecycle, ρ in This is the overstress contribution factor for components. It should be noted that the calculated failure rate is in FITs (Failure Intakes), which represents the failure rate per 10... 9 Number of malfunctions per hour.

[0061] PCS components are generally series-connected and non-repairable systems, while IGBT modules include IGBTs and diodes. Therefore, a PCS failure rate calculation model based on the FIDES reliability guidelines is used to calculate the sum of the failure rates of each component to obtain the total PCS failure rate F based on the FIDES reliability guidelines. FIDES for:

[0062] F FIDES =P FIDES,IGBT +P FIDES,Diode (7)

[0063] In the formula, P FIDES,IGBT P FIDES,Diode The failure rates of IGBTs and Diodes, respectively, are based on the FIDES reliability guidelines and are calculated using formula (6).

[0064] The failure rate of PCS components is not constant during the stable failure period. It varies with time and cannot meet the basic conditions of the constant failure rate model. Therefore, it is necessary to further study the impact of the number of cycles on its operating conditions.

[0065] For PCS (Power Control System), the failure of its component materials is a random event, and there is an uncertain relationship between fatigue level and failure. Theoretically, PCS failures are largely attributed to the IGBTs (Integrated Power Bicycles) used in its applications. However, the two main failure causes of PCS modules are aluminum bond wire detachment and solder layer cracking, and failure can occur at any time. The aging process can be described by a Weibull distribution. Based on the study of aging characterization methods, the reliability of IGBTs was analyzed.

[0066] Assuming the failure probability distribution functions of the aluminum bonding wires, solder layer materials, and IGBTs conform to the Weibull distribution model, the specific expressions are as follows:

[0067]

[0068] In the formula, m bw m sl These are the Weibull distribution shape parameters for the aluminum bonding wires and the solder layer, respectively; η bw η sl , respectively, represent the scale parameters of the Weibull distribution of the aluminum bonding wire and the solder layer; N is the number of cycles; f bw (N), f sl (N), f IGBT (N) represents the failure probability density of the aluminum bonding wire, solder layer, and IGBT, respectively.

[0069] The cumulative failure probability functions for aluminum bonding wires and solder layers in the IGBT module are as follows:

[0070]

[0071] In the formula, F bw (N), F sl (N) represents the cumulative failure probability of the aluminum bonding wire and the solder layer, respectively.

[0072] The specific expression for the influence of the PCS on its failure probability with the number of cycles is as follows:

[0073] F CL (N)=1-[1-F bw (N)][1-F sl (N)](10)

[0074] In the formula, F CL (N) represents the PCS wear probability under different cycle numbers.

[0075] In step S102, a converter reliability assessment model is constructed based on the total failure probability of components and the wear failure probability.

[0076] like Figure 2As shown, in actual implementation, by combining equations (7) and (10), the failures of the PCS equipment caused by the number of cycles can be derived. Combined with the FIDES reliability guideline model, a PCS overall reliability model considering the influence of the number of cycles and the thermal stress factor and temperature cycling factor can be established, namely, the converter reliability assessment model:

[0077] P fault,PCS =1-[1-F FIDES ][1-F CL (N)] (11)

[0078] In step S103, energy storage sample data of the target battery module is obtained, and the bidirectional long short-term memory network is trained using the energy storage sample data to obtain multiple reconstruction errors.

[0079] In some embodiments, energy storage sample data of the target battery module is acquired to train a bidirectional long short-term memory network using the energy storage sample data, thereby obtaining reconstruction error, including:

[0080] Acquire energy storage sample data of the target battery module and perform planning processing on the energy storage sample data to obtain normalized energy storage sample data;

[0081] A battery energy storage data feature mining model based on a bidirectional long short-term memory network was trained using normalized energy storage sample data to obtain a reconstructed dataset, and multiple reconstruction errors were calculated based on the reconstructed dataset.

[0082] like Figure 2 As shown, in the actual execution process, in order to eliminate the influence of the dimensions of each feature data when calculating the reconstruction error and to ensure the training accuracy of the basic reconstruction model, it is necessary to perform data standardization processing on the target battery module, using the mean square error standardization z-score method. The n battery energy storage portions of the obtained data are used as the training set, and the remaining battery energy storage portions are used as the test set, i.e.:

[0083]

[0084] In the formula, s* represents the normalized energy storage sample data, s sam For the energy storage sample data of the target battery module, s μ s is the mean of the energy storage sample data. σ This represents the standard deviation of the battery energy storage sample.

[0085] Furthermore, an ensemble model is constructed based on the Bi-LSTM-based battery energy storage data feature mining method. The training and optimization objective of the ensemble model is to minimize the reconstruction error of the battery energy storage input and output data, i.e.:

[0086]

[0087] In the formula, For input S in With output The reconstruction error, n va For S in and Dimension, s in,i,t For S in The element value of the i-th dimension at time t, s out,i,t for The element value of the i-th dimension at time t is calculated by the Bi-LSTM network method, where T is the total number of time intervals in the running time.

[0088] Input n training sets into an ensemble model, train n ensemble models in sequence to calculate multiple reconstructed data, and form a reconstructed dataset to calculate the reconstructed error based on the reconstructed dataset.

[0089] In step S104, multiple reconstruction errors are input into a pre-built safety warning judgment model for discrimination, so as to calculate the safety warning probability based on the data reconstruction error.

[0090] In actual implementation, multiple reconstruction errors are input into a pre-built security early warning judgment model, allowing these multiple reconstruction errors to be compared and analyzed with a given reconstruction error threshold K to determine the output of each integrated model. This results in multiple judgments, among which the pre-constructed security warning judgment model is:

[0091]

[0092] In the formula, The integrated model's judgment result on battery energy storage is given by K, which is the threshold for determining a safety warning.

[0093] Furthermore, the safety warning probability P of each integrated model is calculated based on multiple judgment results. fault,BM ,Right now:

[0094]

[0095] In the formula, y i The result of the i-th integrated model's judgment on the battery energy storage is calculated by equation (14).

[0096] Since the target battery module BC is composed of several battery modules, the thermal runaway warning result of the target battery module BC, i.e., the safety warning probability based on data reconstruction error, is:

[0097]

[0098] In the formula, Pfault,BC P is the probability of BC security warning based on data reconstruction error. fault,BC,1 P fault,BC,2 … Based on the data reconstruction error n BM The probability of a safety warning for a battery module.

[0099] In step S105, the voltage and temperature measurement points of all individual cells in the target battery module are obtained to calculate the consistency safety warning probability based on the voltage and temperature measurement points of all individual cells.

[0100] In some embodiments, the voltage and temperature measurement points of all individual cells in the target battery module are obtained to calculate the consistency safety warning probability based on the voltage and temperature measurement points of all individual cells, including:

[0101] Obtain the voltage of all individual cells in the target battery module to calculate the average voltage of the target battery module at a preset time.

[0102] Filter the maximum and minimum voltage values ​​of the target battery module at a preset time;

[0103] Calculate the root mean square error of the extreme voltage based on the maximum and minimum voltage values;

[0104] The ratio of the root mean square error of extreme voltage to the average voltage is used as the voltage consistency coefficient.

[0105] Obtain the temperature measurement points of all individual cells in the target battery module to calculate the maximum and minimum temperature values ​​of the target battery module at a preset time.

[0106] The root mean square error of extreme temperature is calculated based on the maximum and minimum temperature values, and the root mean square error of extreme temperature is used as the standard deviation of temperature consistency.

[0107] The probability of a safety warning based on consistency is calculated using the voltage consistency coefficient and the temperature consistency standard deviation.

[0108] like Figure 2 As shown, in actual operation, the target battery module involves several chemical reactions during charging and discharging. If these internal chemical reactions are balanced, the battery energy storage will have high consistency. Uneven battery energy storage can lead to overcharging / over-discharging, not only shortening its lifespan but also potentially causing thermal runaway, increasing the probability of system failure, and reducing its reliability. Figure 3 As shown, when evaluating the operational reliability of a battery energy storage system, it is also necessary to consider consistency indicators such as the current, voltage, temperature, state of charge, internal resistance, and health status of the battery module. However, due to limitations in data availability, this embodiment of the invention only establishes a consistency model for the voltage and temperature of the battery module and does not analyze other parameters.

[0109] The voltage consistency assessment of the battery module should calculate the average voltage of the target battery module at time t, then determine the maximum and minimum values ​​of the terminal voltage of each individual battery cell at time t, and calculate the ratio of the root mean square error of the extreme voltage to the average voltage as the evaluation index of voltage consistency. The specific model is shown in the following formulas (17)-(22):

[0110]

[0111]

[0112] u max,t ={u 1,t ,u 2,t ,...,u n,t}(19)

[0113] u min,t =min{u 1,t ,u 2,t ,...,u n,t}(20)

[0114]

[0115]

[0116] In the formula, u av,t Let n be the average voltage of each individual cell at time t. cell u represents the total number of individual cells in the target battery module. n,t Let n be the voltage of each individual cell n at time t. u is the average value of the terminal voltage curve during the charging and discharging phases of the target battery module. max,t u min,t These represent the maximum and minimum voltage values ​​of each individual cell at time t, respectively, and σ u The root mean square error of the extreme terminal voltage during the charging and discharging phase of the target battery module, μ u Battery module voltage consistency coefficient.

[0117] The temperature consistency assessment of the battery module should calculate the maximum and minimum temperatures of all temperature measurement points of the battery module at time t, and calculate the root mean square error of the extreme temperature as the evaluation index of temperature consistency. The specific model is shown in the following formulas (23)-(25):

[0118] T max,t =max{T 1,t ,T 2,t ,...,T m,t}(twenty three)

[0119] T min,t=min{T 1,t ,T 2,t ,...,T m,t} (twenty four)

[0120]

[0121] In the formula, T max,t T min,t Let T be the maximum and minimum temperature values ​​at all temperature measurement points of the target battery module at time t. 1,t T 2,t ... T m,t For the temperature of the target battery module at measurement points 1, 2, ..., m, σ temp The root mean square of the extreme temperature error during the charging and discharging phase of the target battery module; the larger the value, the worse the temperature consistency.

[0122] The unweighted scores of formulas (22) and (25) are weighted to obtain the SOB safety warning probability used to evaluate the consistency of the target battery module, as follows:

[0123]

[0124]

[0125] In the formula, These are the weighting coefficients for the battery module voltage consistency coefficient and the temperature consistency standard deviation, respectively. U V T These are the unweighted safety warning scores for the battery module's voltage consistency coefficient and temperature consistency standard deviation, respectively. The numbers are 1, 2, ..., n respectively. BM The weighted consistency safety warning probability of the target battery module, P SOB,BC The probability of a consistency safety warning for the target battery module BC.

[0126] In step S106, a battery energy storage safety early warning model is constructed based on the safety early warning probability based on data reconstruction error and the consistency safety early warning probability.

[0127] Specifically, formulas (16) and (27) are combined to form the battery energy storage safety early warning model P. TR&SOB .

[0128] In step S107, an operational reliability assessment model is established based on the converter reliability assessment model and the battery energy storage safety early warning model, so as to use the operational reliability assessment model to assess the operational reliability of the target battery energy storage system.

[0129] In some embodiments, the expression for running the reliability assessment model is:

[0130] P BESS =1-(1-P) fault,PCS (1-P) TR&SOB (1-P) fault,AE )

[0131] Among them, P BESS For integrated security early warning probability, P fault,PCS P represents the target energy storage converter failure rate. TR&SOB P is the battery failure probability based on data reconstruction error and consistency. fault,AE The failure probability of auxiliary equipment in the target battery energy storage system, excluding the target energy storage converter and the target battery module.

[0132] like Figure 2 As shown, in actual implementation, the target battery energy storage system (BESS) also involves several auxiliary devices, including battery energy switches, battery energy hubs, battery energy adapters, and battery energy network cards. Therefore, it is necessary to obtain the failure probability P of other auxiliary devices besides BC and PCS using known methods. fault,AE .

[0133] Furthermore, the PCS failure rate P is calculated based on the aforementioned converter reliability assessment model. fault,PCS Based on the aforementioned battery energy storage safety early warning model, the battery failure probability P based on data reconstruction error and SOB is calculated. TR&SOB An operational reliability assessment model is established based on the PCS failure rate, the battery failure probability based on data reconstruction error and SOB, and the failure probability of other auxiliary equipment. This model is used to assess the operational reliability of the target battery energy storage system. The operational reliability assessment model is as follows:

[0134] P BESS =1-(1-P) fault,PCS (1-P) TR&SOB (1-P) fault,AE )

[0135] In the formula, P BESS For operational reliability results, P fault,PCS For PCS failure rate, P TR&SOB To determine the battery failure probability based on data reconstruction error and SOB, P fault,AE The failure probability of BESS auxiliary equipment other than BC and PCS.

[0136] According to the data-driven battery energy storage system reliability assessment method proposed in this embodiment, a model of converter component thermal stress factor and junction temperature cycling factor based on the FIDES reliability guideline is established. Then, the junction temperature cycling information of the converter is extracted by the rainflow counting method, and the influence of the number of cycles of the PCS based on the Weibull probability distribution on its failure probability is introduced, thereby constructing a converter reliability assessment model. In view of battery monitoring data and state imbalance, a battery cluster reliability method based on reconstruction error calculation and balanced state SOB is proposed to explore the potential patterns of battery operation data, thereby constructing a battery energy storage safety early warning model. Finally, based on the converter reliability assessment model and the battery energy storage safety early warning model, an assessment model is established that considers both the influence of PCS pressure, temperature, and cycle number on the DESS operation reliability and the influence of the potential patterns of battery operation data on the DESS operation reliability. This allows for more accurate operation reliability assessment results, more reasonable allocation of DESS resources, and maintenance of the overall performance of the DESS.

[0137] Next, referring to the accompanying drawings, a data-driven battery energy storage system operation reliability assessment device proposed according to an embodiment of the present invention is described.

[0138] Figure 4 This is a block diagram of a data-driven battery energy storage system operation reliability assessment device according to an embodiment of the present invention.

[0139] like Figure 4 As shown, the data-driven battery energy storage system operation reliability assessment device 40 includes: a first acquisition module 401, a first construction module 402, a training module 403, a discrimination module 404, a second acquisition module 405, a second construction module 406, and an assessment module 407.

[0140] The system comprises the following modules: First Acquisition Module 401 acquires junction temperature fluctuation information and cycle count of the target energy storage converter to calculate the total component failure probability and wear failure probability of the converter. First Construction Module 402 constructs a converter reliability assessment model based on the total component failure probability and wear failure probability. Training Module 403 acquires energy storage sample data of the target battery module to train a bidirectional long short-term memory network using this data, generating multiple reconstruction errors. Discrimination Module 404 inputs these reconstruction errors into a pre-constructed safety warning judgment model for discrimination, calculating the safety warning probability based on the data reconstruction errors. Second Acquisition Module 405 acquires the voltage and temperature measurement points of all individual cells in the target battery module to calculate the consistency safety warning probability based on these data. Second Construction Module 406 constructs a battery energy storage safety warning model based on the safety warning probability based on the data reconstruction errors and the consistency safety warning probability. The evaluation module 407 is used to establish an operational reliability evaluation model based on the converter reliability evaluation model and the battery energy storage safety early warning model, so as to use the operational reliability evaluation model to evaluate the operational reliability of the target battery energy storage system.

[0141] It should be noted that the foregoing explanation of the embodiment of the data-driven battery energy storage system operation reliability assessment method also applies to the data-driven battery energy storage system operation reliability assessment device of this embodiment, and will not be repeated here.

[0142] According to the data-driven battery energy storage system reliability assessment device proposed in this embodiment, a model of converter component thermal stress factor and junction temperature cycling factor based on the FIDES reliability guideline is established. Then, the junction temperature cycling information of the converter is extracted by the rainflow counting method, and the influence of the number of cycles of the PCS based on the Weibull probability distribution on its failure probability is introduced, thereby constructing a converter reliability assessment model. In response to battery monitoring data and state imbalance, a battery cluster reliability method based on reconstruction error calculation and balanced state SOB is proposed to explore the potential patterns of battery operation data, thereby constructing a battery energy storage safety early warning model. Finally, based on the converter reliability assessment model and the battery energy storage safety early warning model, an assessment model is established that considers both the influence of PCS pressure, temperature, and cycle number on the DESS operation reliability and the influence of the potential patterns of battery operation data on the DESS operation reliability. This allows for more accurate operation reliability assessment results, more reasonable allocation of DESS resources, and maintenance of the overall performance of the DESS.

[0143] Figure 5 This is a schematic diagram of an electronic device provided in an embodiment of the present invention. The electronic device may include:

[0144] The memory 501, the processor 502, and the computer program stored on the memory 501 and capable of running on the processor 502.

[0145] When the processor 502 executes the program, it implements the data-driven battery energy storage system operation reliability assessment method provided in the above embodiments.

[0146] Furthermore, electronic devices also include:

[0147] Communication interface 503 is used for communication between memory 501 and processor 502.

[0148] The memory 501 is used to store computer programs that can run on the processor 502.

[0149] The memory 501 may include high-speed RAM memory, and may also include non-volatile memory, such as at least one disk storage device.

[0150] If the memory 501, processor 502, and communication interface 503 are implemented independently, then the communication interface 503, memory 501, and processor 502 can be interconnected via a bus to complete communication between them. The bus can be an Industry Standard Architecture (ISA) bus, a Peripheral Component Interconnect (PCI) bus, or an Extended Industry Standard Architecture (EISA) bus, etc. Buses can be categorized as address buses, data buses, control buses, etc. For ease of representation, Figure 5 The bus is represented by a single thick line, but this does not mean that there is only one bus or one type of bus.

[0151] Optionally, in a specific implementation, if the memory 501, processor 502, and communication interface 503 are integrated on a single chip, then the memory 501, processor 502, and communication interface 503 can communicate with each other through an internal interface.

[0152] Processor 502 may be a central processing unit (CPU), an application specific integrated circuit (ASIC), or one or more integrated circuits configured to implement embodiments of the present invention.

[0153] This invention also provides a computer program product, which, when executed by a processor, implements the above-described data-driven battery energy storage system operation reliability assessment method.

[0154] This invention also provides a computer-readable storage medium storing a computer program that, when executed by a processor, implements the above-described data-driven battery energy storage system operation reliability assessment method.

[0155] In the description of this specification, the references to terms such as "one embodiment," "some embodiments," "example," "specific example," or "some examples," etc., refer to specific features, structures, materials, or characteristics described in connection with that embodiment or example, which are included in at least one embodiment or example of the present invention. In this specification, the illustrative expressions of the above terms do not necessarily refer to the same embodiment or example. Furthermore, the specific features, structures, materials, or characteristics described may be combined in any suitable manner in one or more embodiments or examples. Moreover, without contradiction, those skilled in the art can combine and integrate the different embodiments or examples described in this specification, as well as the features of different embodiments or examples.

[0156] Furthermore, the terms "first" and "second" are used for descriptive purposes only and should not be construed as indicating or implying relative importance or implicitly specifying the number of indicated technical features. Thus, a feature defined as "first" or "second" may explicitly or implicitly include at least one of that feature. In the description of this invention, "N" means at least two, such as two, three, etc., unless otherwise explicitly specified.

[0157] Any process or method description in the flowchart or otherwise herein can be understood as representing a module, segment, or portion of code comprising one or N executable instructions for implementing custom logic functions or processes, and the scope of preferred embodiments of the invention includes additional implementations in which functions may be performed not in the order shown or discussed, including substantially simultaneously or in reverse order depending on the functions involved, as should be understood by those skilled in the art to which embodiments of the invention pertain.

[0158] The logic and / or steps represented in the flowchart or otherwise described herein, for example, can be considered as a sequenced list of executable instructions for implementing logical functions, and can be embodied in any computer-readable medium for use by, or in conjunction with, an instruction execution system, apparatus, or device (such as a computer-based system, a processor-included system, or other system that can fetch and execute instructions from, an instruction execution system, apparatus, or device). For the purposes of this specification, "computer-readable medium" can be any means that can contain, store, communicate, propagate, or transmit programs for use by, or in conjunction with, an instruction execution system, apparatus, or device. More specific examples (a non-exhaustive list) of computer-readable media include: an electrical connection having one or more wires (electronic device), a portable computer disk drive (magnetic device), random access memory (RAM), read-only memory (ROM), erasable and editable read-only memory (EPROM or flash memory), fiber optic devices, and portable optical disc read-only memory (CDROM). Alternatively, the computer-readable medium may be paper or other suitable media on which the program can be printed, since the program can be obtained electronically by optically scanning the paper or other medium, followed by editing, interpreting, or otherwise processing as necessary, and then stored in a computer memory.

[0159] It should be understood that various parts of the present invention can be implemented in hardware, software, firmware, or a combination thereof. In the above embodiments, the N steps or methods can be implemented in software or firmware stored in memory and executed by a suitable instruction execution system. If implemented in hardware, as in another embodiment, it can be implemented using any one or a combination of the following techniques known in the art: discrete logic circuits having logic gates for implementing logical functions on data signals, application-specific integrated circuits (ASICs) having suitable combinational logic gates, programmable gate arrays (PGAs), field-programmable gate arrays (FPGAs), etc.

[0160] Those skilled in the art will understand that all or part of the steps of the methods described in the above embodiments can be implemented by a program instructing related hardware. The program can be stored in a computer-readable storage medium, and when executed, it includes one or a combination of the steps of the method embodiments.

[0161] Furthermore, the functional units in the various embodiments of the present invention can be integrated into a processing module, or each unit can exist physically separately, or two or more units can be integrated into a module. The integrated module can be implemented in hardware or as a software functional module. If the integrated module is implemented as a software functional module and sold or used as an independent product, it can also be stored in a computer-readable storage medium.

[0162] The storage medium mentioned above can be a read-only memory, a disk, or an optical disk, etc. Although embodiments of the present invention have been shown and described above, it is to be understood that the above embodiments are exemplary and should not be construed as limiting the present invention. Those skilled in the art can make changes, modifications, substitutions, and variations to the above embodiments within the scope of the present invention.

Claims

1. A data-driven method for evaluating the operational reliability of a battery energy storage system, characterized in that, Includes the following steps: Obtain junction temperature fluctuation information and cycle count of the target energy storage converter, and calculate the total component failure probability and wear failure probability of the target energy storage converter based on the junction temperature fluctuation information and the cycle count; A converter reliability assessment model is constructed based on the total failure probability of the components and the wear failure probability. Acquire energy storage sample data of the target battery module, and use the energy storage sample data to train the bidirectional long short-term memory network to obtain multiple reconstruction errors; The multiple reconstruction errors are input into a pre-built security warning judgment model for discrimination, so as to calculate the security warning probability based on the data reconstruction errors; The voltage and temperature measurement points of all individual cells in the target battery module are obtained, and the consistency safety warning probability is calculated based on the voltage and temperature measurement points of all individual cells. A battery energy storage safety early warning model is constructed based on the safety early warning probability based on data reconstruction error and the consistency safety early warning probability. An operational reliability assessment model is established based on the converter reliability assessment model and the battery energy storage safety early warning model, so as to use the operational reliability assessment model to assess the operational reliability of the target battery energy storage system.

2. The data-driven battery energy storage system operation reliability assessment method according to claim 1, characterized in that, The step of acquiring junction temperature fluctuation information and cycle count of the target energy storage converter, and calculating the total component failure probability and wear failure probability of the target energy storage converter based on the junction temperature fluctuation information, includes: Calculate the junction temperature fluctuation information and cycle number of the target energy storage converter under different operating conditions; The total failure probability of the component is calculated based on the junction temperature fluctuation information, and the wear failure probability is calculated based on the number of cycles.

3. The data-driven battery energy storage system operation reliability assessment method according to claim 2, characterized in that, The step of calculating the total failure probability of the component based on the junction temperature fluctuation information and calculating the wear failure probability based on the number of cycles includes: Calculate the thermal stress factor and temperature cycling factor of each component in the target energy storage converter based on the junction temperature fluctuation information. Based on the FIDES reliability guidelines, the failure rate of each component is calculated according to the preset basic failure rate parameters, thermal stress factor and temperature cycling factor of each component. The total failure probability of the target energy storage converter is obtained by summing the failure rates of each component. Based on the Weibull distribution, the probability density of the aluminum bonding wires and the probability density of the solder layer of the target energy storage converter are calculated. The wear failure probability of the target energy storage converter under the specified number of cycles is calculated based on the number of cycles, the probability density of the aluminum bonding wire, and the probability density of the solder layer.

4. The data-driven battery energy storage system operation reliability assessment method according to claim 1, characterized in that, The acquisition of energy storage sample data of the target battery module, and the use of the energy storage sample data to train the bidirectional long short-term memory network to obtain the reconstruction error, includes: The energy storage sample data of the target battery module is obtained, and the energy storage sample data is processed to obtain normalized energy storage sample data. A battery energy storage data feature mining model based on a bidirectional long short-term memory network is trained using the normalized energy storage sample data to obtain a reconstructed dataset, and the multiple reconstruction errors are calculated based on the reconstructed dataset.

5. The data-driven battery energy storage system operation reliability assessment method according to claim 1, characterized in that, The step of acquiring the voltage and temperature measurement points of all individual cells in the target battery module, and calculating the consistency safety warning probability based on the voltage and temperature measurement points of all individual cells, includes: The voltage of all individual cells in the target battery module is obtained to calculate the average voltage of the target battery module at a preset time. Filter the maximum and minimum voltage values ​​of the target battery module at the preset time; Calculate the root mean square error of the extreme voltage based on the maximum voltage value and the minimum voltage value; The ratio of the root mean square error of the extreme voltage to the average voltage is used as the voltage consistency coefficient; The temperature measurement points of all individual cells in the target battery module are obtained to calculate the maximum and minimum temperature values ​​of the target battery module at the preset time. The root mean square error of extreme temperature is calculated based on the maximum temperature value and the minimum temperature value, and the root mean square error of extreme temperature is used as the standard deviation of temperature consistency. The probability of a consistency safety warning is calculated based on the voltage consistency coefficient and the temperature consistency standard deviation.

6. The data-driven battery energy storage system operation reliability assessment method according to claim 1, characterized in that, The expression for the operational reliability assessment model is: P BESS =1-(1-P fault,PCS )(1-P TR&SOB )(1-P fault,AE ) Among them, P BESS For integrated security early warning probability, P fault,PCS P represents the target energy storage converter failure rate. TR&SOB P is the battery failure probability based on data reconstruction error and consistency. fault,AE The failure probability of auxiliary equipment in the target battery energy storage system, excluding the target energy storage converter and the target battery module.

7. A data-driven battery energy storage system operation reliability assessment device, characterized in that, include: The first acquisition module is used to acquire junction temperature fluctuation information and cycle number of the target energy storage converter, so as to calculate the total component failure probability and wear failure probability of the target energy storage converter based on the junction temperature fluctuation information and the cycle number. The first construction module is used to construct a converter reliability assessment model based on the total failure probability of the components and the wear failure probability. The training module is used to acquire energy storage sample data of the target battery module, and to use the energy storage sample data to train the bidirectional long short-term memory network to obtain multiple reconstruction errors. The discrimination module is used to input the multiple reconstruction errors into a pre-built security early warning judgment model for discrimination, so as to calculate the security early warning probability based on the data reconstruction errors; The second acquisition module is used to acquire the voltage and temperature measurement point of all individual cells in the target battery module, so as to calculate the consistency safety warning probability based on the voltage and temperature measurement point of all individual cells. The second construction module is used to construct a battery energy storage safety early warning model based on the safety early warning probability based on data reconstruction error and the consistency safety early warning probability. The evaluation module is used to establish an operational reliability evaluation model based on the converter reliability evaluation model and the battery energy storage safety early warning model, so as to use the operational reliability evaluation model to evaluate the operational reliability of the target battery energy storage system.

8. An electronic device, characterized in that, include: The system includes a memory, a processor, and a computer program stored in the memory and executable on the processor, wherein the processor executes the program to implement the data-driven battery energy storage system operation reliability assessment method as described in any one of claims 1-6.

9. A computer program product, characterized in that, When the computer program / instruction is executed by the processor, it implements the data-driven battery energy storage system operation reliability assessment method according to any one of claims 1-6.

10. A computer-readable storage medium having a computer program stored thereon, characterized in that, The program is executed by the processor to implement the data-driven battery energy storage system operation reliability assessment method as described in any one of claims 1-6.

Images