A smart Internet of Things-based power storage system

Abstract

This invention relates to the field of power energy storage technology, specifically disclosing a power energy storage system based on the Internet of Things. The system includes: an acoustic emission detection module for acquiring stress wave information inside the energy storage battery; a sensor group including a gas sensor, a temperature sensor, and a pressure sensor for acquiring characteristic gas concentration information, temperature information, and pressure information of the energy storage battery, respectively; an energy storage battery parameter interface for acquiring operating information of the energy storage battery; and an energy storage management system for determining safety thresholds based on the information acquired by the acoustic emission detection module, the sensor group, and the energy storage battery parameter interface, and, if the threshold range is not exceeded, performing correlation analysis between the stress wave information, characteristic gas concentration information, pressure information, temperature information, and operating information, and predicting the risk of energy storage battery runaway based on the analysis results.

Description

Technical Field

[0001] This invention relates to the field of power energy storage technology, specifically to a power energy storage system based on the Internet of Things. Background Technology

[0002] Electrical energy storage systems (ESS) are technological systems that convert electrical energy into other forms of energy for storage and then convert it back into electrical energy for release when needed. They are a core component of modern energy systems (especially smart grids and renewable energy systems), addressing the mismatch between electricity supply and demand in terms of time, space, and intensity. Among these technologies, lithium-ion battery-based energy storage is currently the most mainstream and fastest-growing, boasting high energy and power density and fast response speed, and is widely used in industrial and commercial energy storage, as well as grid-side energy storage.

[0003] Existing power storage systems are equipped with monitoring modules to monitor and manage the operating status of energy storage batteries. By acquiring key performance indicators of the energy storage batteries (such as power, cycle life, charge and discharge efficiency, operating power parameters, etc.) and environmental variables (temperature, gas, pressure, etc.), the modules determine the safety status of the energy storage batteries by using threshold judgments based on the range of key performance indicators and influencing variables. In particular, acoustic emission detection technology can capture stress waves generated by internal material deformation, phase change, gas generation, etc., enabling non-invasive monitoring of lithium battery status with high timeliness and predictive early warning of risks to power storage systems.

[0004] When existing power energy storage systems use acoustic emission detection technology, they mainly rely on training AI models to identify the sound of safety valve rupture. Therefore, the accuracy of identification is somewhat insufficient. In addition, the existing threshold judgment method mainly judges energy storage batteries with obvious abnormal conditions. Moreover, since the real-time operating status of batteries varies, the sensitivity of judging energy storage batteries with potential risks is low. Therefore, how to improve the timeliness and accuracy of power energy storage systems in judging potential operational risks is the fundamental problem that this invention aims to solve. Summary of the Invention

[0005] The purpose of this invention is to provide a power storage system based on the Internet of Things (IoT) to solve the following technical problems:

[0006] How to improve the timeliness and accuracy of the assessment of potential operational risks in power storage systems.

[0007] The objective of this invention can be achieved through the following technical solutions:

[0008] A smart Internet of Things-based power storage system, the system comprising:

[0009] Acoustic emission detection module, used to acquire stress wave information inside the energy storage battery;

[0010] The sensor group includes a gas sensor, a temperature sensor, and a pressure sensor, which are used to acquire characteristic gas concentration information, temperature information, and pressure information of the energy storage battery of the energy storage power station, respectively.

[0011] The energy storage battery parameter interface is used to obtain the operating information of the energy storage battery;

[0012] The energy storage management system is used to determine safety thresholds based on information obtained from the acoustic emission detection module, sensor group, and energy storage battery parameter interface. When the threshold is not exceeded, the system performs correlation analysis with stress wave information, characteristic gas concentration information, pressure information, temperature information, and operating information, and predicts the risk of energy storage battery runaway based on the analysis results.

[0013] Through the above technical solution, correlation analysis is performed on stress wave information, characteristic gas concentration information, pressure information, temperature information and operation information. This process judges the abnormality of changes in stress wave information, characteristic gas concentration information, pressure information and temperature information based on the operation information of the energy storage battery. Therefore, based on the preliminary judgment, the risk of runaway of the energy storage battery can be predicted according to the correlation analysis results, which can greatly improve the sensitivity and accuracy of the prediction results.

[0014] Furthermore, the pressure sensor is a piezoresistive thin-film sensor, which is installed on the surface of each energy storage battery; the temperature sensor is a distributed fiber optic temperature measurement module, which is arranged to cover all monitoring points of the energy storage power station; and the gas sensors are evenly distributed at various points inside the energy storage power station.

[0015] The above technical solutions enable the accurate acquisition of characteristic gas concentration information, temperature information, and energy storage battery pressure information.

[0016] Furthermore, the correlation analysis process includes:

[0017] The internal space of the energy storage power station is divided into regions according to the monitoring range of the gas sensor. The stress wave information of each region is input into the AI ​​model to obtain the risk type of the energy storage battery.

[0018] Through formula The gas matching coefficient g for each region is calculated, where m is the number of monitored gas species, j∈[1,m]. Let j be the monitoring concentration of the j-th gas. Let the basic concentration threshold of the j-th gas be... Indicates selection The maximum value among 0, The weighting coefficient for the j-th gas corresponding to the risk type Type of the energy storage battery;

[0019] The real-time operating coefficient R(t) of the energy storage battery in the current area is obtained based on the operating data of the energy storage battery in each area; the real-time state coefficient S(t) of each energy storage battery is determined based on the pressure and temperature information; and the synchronization coefficient X between the real-time state coefficient S(t) and the real-time operating coefficient R(t) is obtained.

[0020] The gas matching coefficient g is compared with the preset threshold gt:

[0021] If g≥gt, then it is determined that there is a risk fault of type Type in the energy storage battery in the region, and the energy storage battery with risk fault is determined according to the synchronization coefficient X.

[0022] If g < gt, then determine whether there is a risky energy storage battery based on the magnitude of the synchronization coefficient X, and if the determination is yes, identify the energy storage battery with the risky failure.

[0023] By using the above technical solution, the synchronization coefficient X can be used to determine the synchronicity between the changes in the real-time state coefficient S(t) and the real-time operating coefficient R(t). The higher the synchronicity, the more likely the change in the real-time state coefficient S(t) is caused by the fluctuation of the real-time operating coefficient R(t), which in turn indicates that the probability of other fault causes causing the fluctuation of R(t) is low. Therefore, the safety of the energy storage battery is judged to be high.

[0024] Furthermore, the process of obtaining the synchronization coefficient includes:

[0025] Select the real-time state coefficient S(t) and real-time operation coefficient R(t) corresponding to n time points at fixed time intervals;

[0026] Through formula The synchronization coefficient X is calculated.

[0027] Where i∈[1,n], Let i be the i-th time point.

[0028] The above technical solution enables the determination of the degree of synchronization by the magnitude of the synchronization coefficient.

[0029] Furthermore, the process of obtaining the real-time operating coefficient R(t) of the current regional energy storage battery includes:

[0030] Through formula The real-time operating coefficient R(t) is calculated and obtained;

[0031] Where q is the total number of parameters of the energy storage battery, k∈[1,q], Let f be the real-time value of the k-th parameter, and f be a defined function. When within the basic threshold range ,otherwise, , for The corresponding unit value, , The boundary of the basic threshold, Indicates selection and The minimum value in, The influence weight of the k-th parameter.

[0032] The above technical solution enables the determination of the current operating status of energy storage batteries.

[0033] Furthermore, the process of obtaining the real-time operating coefficient R(t) of the current regional energy storage battery includes:

[0034] Through formula Calculate the real-time operating coefficient R(t) of the energy storage battery in the current area;

[0035] in, Let x be the real-time temperature of the x-th monitoring point. Let x be the baseline temperature value for the x-th monitoring point. Of all monitoring points during the monitoring period The maximum value, This is a temperature unit value. This is the baseline value for the rate of temperature rise. Of all monitoring points during the monitoring period and the maximum value of 0, , As a fixed parameter adjustment coefficient, F(t) represents the real-time pressure on the battery surface. To monitor the maximum value of F(t) within the specified time period, This is the pressure unit value.

[0036] The above technical solution can comprehensively assess the changes in external factors of energy storage batteries by considering temperature magnitude, temperature rise rate, and battery surface pressure.

[0037] Furthermore, when g ≥ gt, the process for determining a potential energy storage battery failure includes:

[0038] Compare the synchronization coefficient X corresponding to all energy storage batteries with the preset threshold Xt:

[0039] If there are energy storage batteries with X < Xt, then all energy storage batteries with X < Xt are identified as energy storage batteries with risk of failure.

[0040] If there is no energy storage battery with X < Xt, then the energy storage battery with the smallest synchronization coefficient X is judged to be the energy storage battery with a risk of failure.

[0041] The above technical solution enables the identification of energy storage batteries that experience risky failures when g ≥ gt.

[0042] Furthermore, when g < gt, the process for determining a potential energy storage battery failure includes:

[0043] The synchronization coefficient X is compared with the preset threshold Xt:

[0044] If there are energy storage batteries with X < Xt, then all energy storage batteries with X < Xt are identified as energy storage batteries with risk of failure.

[0045] If there is no energy storage battery with X < Xt, then all energy storage batteries in this area are considered to be in normal condition.

[0046] The above technical solution enables an energy storage battery to determine whether a risk fault has occurred when g < gt.

[0047] The beneficial effects of this invention are:

[0048] (1) The present invention performs correlation analysis on stress wave information, characteristic gas concentration information, pressure information and temperature information with operation information. This process judges the abnormality of changes in stress wave information, characteristic gas concentration information, pressure information and temperature information based on the operation information of the energy storage battery. Therefore, based on the preliminary judgment, the risk of runaway of the energy storage battery can be predicted according to the correlation analysis results, which can greatly improve the sensitivity and accuracy of the prediction results. Attached Figure Description

[0049] The invention will now be further described with reference to the accompanying drawings.

[0050] Figure 1 This is a logical block diagram of the power energy storage system based on the smart Internet of Things of this invention. Detailed Implementation

[0051] The technical solutions of the embodiments of the present invention will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only some embodiments of the present invention, and not all embodiments. Based on the embodiments of the present invention, all other embodiments obtained by those skilled in the art without creative effort are within the scope of protection of the present invention.

[0052] In one embodiment, a smart Internet of Things-based power storage system is provided; please refer to [link / reference]. Figure 1As shown, the system includes an acoustic emission detection module, a sensor group, a battery parameter interface, and an energy storage management system. The sensor group includes gas sensors, temperature sensors, and pressure sensors. The pressure sensors are piezoresistive thin-film sensors installed on the surface of each battery. The temperature sensors utilize distributed fiber optic temperature measurement modules, distributed across all monitoring points within the energy storage power station, providing high detection accuracy. The gas sensors are evenly distributed throughout the energy storage power station, capable of detecting characteristic gases such as CO and H2. Through the sensor group, the system can acquire information on the concentration of characteristic gases, temperature, and pressure of the batteries. The acoustic emission detection module employs existing technology to acquire stress wave information within the batteries. Its structure includes a piezoelectric broadband sensor (frequency response range 20–500 kHz). The system comprises a kHz-level acoustic emission detection module, a conductive clamp, a preamplifier, an acoustic emission detection module, and a processing terminal. Based on FFT (Fast Fourier Transform) and wavelet analysis, it extracts time-domain / frequency-domain features and combines AI algorithms to classify signals, thereby determining the type of safety risk present in the energy storage battery. Examples include casing rupture and impedance mismatch caused by gas generation between electrode layers. The energy storage battery parameter interface is used to acquire the battery's operational information. Finally, the energy storage management system uses the information acquired from the acoustic emission detection module, sensor group, and energy storage battery parameter interface to determine safety thresholds, achieving a preliminary assessment of the energy storage battery's risks. This assessment process can identify… The energy storage battery exhibits obvious faults. Therefore, when the judgment result exceeds the threshold range, it indicates that the corresponding energy storage battery is abnormal. When the judgment does not exceed the threshold range, this embodiment performs correlation analysis based on stress wave information, characteristic gas concentration information, pressure information, temperature information, and operating information. This process judges the abnormality of changes in stress wave information, characteristic gas concentration information, pressure information, and temperature information based on the operating information of the energy storage battery. Therefore, based on the preliminary judgment, the risk of runaway of the energy storage battery can be predicted according to the correlation analysis results, which can greatly improve the sensitivity and accuracy of the prediction results.

[0053] In one embodiment, a correlation analysis process is provided, including: dividing the internal space of the energy storage power station into regions according to the monitoring range of gas sensors; inputting stress wave information of each region into an AI model to obtain the energy storage battery risk type (Type); wherein the AI ​​model establishes samples by processing multiple sets of audio recordings of abnormal energy storage batteries, and trains them using basic models such as LSTM and GAN to obtain the AI ​​model; the specific training process is existing technology; the closest risk type is obtained from the AI ​​model as the energy storage battery risk type (Type); and then the process is performed using a formula. The gas matching coefficient g for each region is calculated, where m is the number of monitored gas species, j∈[1,m]. Let j be the monitoring concentration of the j-th gas. Let be the basic concentration threshold for the j-th gas. It should be noted that the basic threshold is the corresponding gas concentration under the standard environment of the energy storage power station, and its value is less than the safety threshold in the threshold determination process. Indicates selection The maximum value in 0, therefore when When the value is less than or equal to 0, its effect is not considered. The weighting coefficient for the j-th gas corresponding to the risk type Type of the energy storage battery is determined by adjusting the gas concentration based on the generation state of various gases when the risk type occurs, according to empirical data. This allows for different weighting distribution strategies to be established for different energy storage battery risk types, improving the accuracy and sensitivity of the judgment. Then, the real-time operating coefficient R(t) of the energy storage batteries in each region is obtained based on their operating data. The real-time state coefficient S(t) of each energy storage battery is determined based on pressure and temperature information. Finally, the synchronization coefficient X between the real-time state coefficient S(t) and the real-time operating coefficient R(t) is obtained. The magnitude of the synchronization coefficient X is used to determine the weighting coefficient. This allows us to determine the synchronicity between the changes in the real-time state coefficient S(t) and the real-time operating coefficient R(t). Higher synchronicity indicates that the change in the real-time state coefficient S(t) is caused by fluctuations in the real-time operating coefficient R(t), suggesting a lower probability of other fault causes causing R(t) fluctuations. Therefore, the safety of the energy storage battery is considered high. The gas matching coefficient g is compared with a preset threshold gt, which is obtained by fitting test data. If g ≥ gt, it indicates an abnormal concentration of the characteristic gas, thus indicating a risk fault of type Type for the energy storage battery in that area. This is determined based on the synchronization coefficient X. For energy storage batteries exhibiting potential risks, timely identification and replacement are crucial to ensuring the overall safety of the energy storage power station. The process for identifying risky batteries involves comparing the synchronization coefficient X of all energy storage batteries with a preset threshold Xt. The preset threshold Xt is selected based on empirical data and is set between 0.8 and 0.9. If any energy storage battery has a value X < Xt, then all energy storage batteries with X < Xt are identified as having potential risks. If no energy storage battery has X < Xt, then the energy storage battery with the smallest synchronization coefficient X is identified as having potential risks. This process ensures timely identification and replacement of risky energy storage batteries. The judgment of the battery is as follows: if g < gt, it indicates that the concentration of the characteristic gas is normal. Therefore, the magnitude of the synchronization coefficient X is used to determine whether there is a risky energy storage battery. The synchronization coefficient X is compared with the preset threshold Xt: if there is an energy storage battery with X < Xt, then all energy storage batteries with X < Xt are judged as having a risky energy storage battery; if there is no energy storage battery with X < Xt, then all energy storage batteries in the area are judged to be in normal condition. Through the above process, based on the interaction between the characteristic gas concentration state and the acoustic emission detection module, and combined with the synchronization coefficient X of the energy storage battery, the accurate and timely judgment of the risky energy storage battery can be achieved.

[0054] The process of obtaining the synchronization coefficient includes: firstly, selecting the real-time state coefficient S(t) and real-time operation coefficient R(t) corresponding to n time points at fixed time intervals. The fixed time interval is set according to the selection, and then using the formula... The synchronization coefficient X is calculated; where i∈[1,n], For the i-th time point, the closer the obtained X value is to 1, the higher the synchronization between the real-time state coefficient S(t) and the real-time operating coefficient R(t). This indicates that the change in the real-time state coefficient S(t) is caused by the fluctuation of the real-time operating coefficient R(t) rather than by other faults. Therefore, it can be judged that the energy storage battery has a high safety level.

[0055] In one embodiment, a process for obtaining the real-time operating coefficient R(t) of the current regional energy storage battery is provided, including: firstly, through the formula... The real-time operating coefficient R(t) is calculated; where q is the total number of parameters of the energy storage battery, k∈[1,q], Let f be the real-time value of the k-th parameter, and f be a defined function. When within the basic threshold range ,otherwise, , Indicates selection and The minimum value in, for The corresponding unit value, based on The unit is determined, for example The unit is mA, then 1mA , The boundary of the basic threshold is set based on empirical data of the corresponding parameters. The influence weight of the k-th parameter is set according to the relative range of the corresponding parameter and the strength of the correlation between the corresponding parameter and the real-time state coefficient S(t) in the empirical data. Therefore, the current operating state of the energy storage battery can be judged by the magnitude of the real-time operating coefficient R(t).

[0056] In one embodiment, a process for obtaining the real-time operating coefficient R(t) of the current regional energy storage battery is provided, including: firstly, using the formula... The real-time operating coefficient R(t) of the energy storage battery in the current region is calculated; where, Let x be the real-time temperature of the x-th monitoring point. This is the baseline temperature value for the x-th monitoring point, which is set according to the temperature control standards of the energy storage power station. It should be noted that the baseline temperature value is lower than the corresponding safety threshold. Of all monitoring points during the monitoring period The maximum value, This is a temperature unit value, with a value of 1℃. This is a baseline value for the temperature rise rate, which is lower than the safety threshold corresponding to the temperature rise rate. The specific value is selected and set based on empirical data. Of all monitoring points during the monitoring period And the maximum value of 0, F(t) is the real-time pressure on the battery surface. To monitor the maximum value of F(t) within the specified time period, This is the pressure unit value, which is 1 N. , To fix the parameter tuning coefficients, it is based on and The numerical range and the degree of influence in empirical data are selected and set. On the basis of consistency of numerical range, the greater the degree of influence, the greater the corresponding weight coefficient. Therefore, by obtaining the real-time operating coefficient R(t), the change status of external factors of energy storage battery can be judged by comprehensively considering the temperature magnitude, temperature rise rate and battery surface pressure state.

[0057] The foregoing has provided a detailed description of one embodiment of the present invention, but this description is merely a preferred embodiment and should not be construed as limiting the scope of the invention. All equivalent variations and modifications made within the scope of the claims of this invention should still fall within the patent coverage of this invention.

Claims

1. A power storage system based on a smart Internet of Things, characterized by, The system comprises: an acoustic emission detection module for acquiring stress wave information inside the energy storage battery; a sensor group comprising a gas sensor, a temperature sensor and a pressure sensor for respectively acquiring characteristic gas concentration information, temperature information and pressure information of the energy storage battery; an energy storage battery parameter interfacing end for acquiring operation information of the energy storage battery; an energy storage management system for performing safety threshold judgment based on the information acquired by the acoustic emission detection module, the sensor group and the energy storage battery parameter interfacing end, and performing correlation analysis based on the stress wave information, the characteristic gas concentration information, the pressure information and the temperature information and the operation information when it is judged that the threshold range is not exceeded, and predicting the risk of out-of-control of the energy storage battery based on the analysis result; the pressure sensor is a piezoresistive thin film sensor arranged on the surface of each energy storage battery; the temperature sensor is a distributed optical fiber temperature measurement module arranged at monitoring sites of the energy storage station; and the gas sensors are uniformly distributed at various points inside the energy storage station; the correlation analysis process comprises: dividing the internal space of the energy storage station into regions according to the monitoring range of the gas sensor, inputting the stress wave information of each region into an AI model to acquire the risk type Type of the energy storage battery; The gas matching coefficient g of each area is calculated by the formula , wherein m is the number of types of monitored gases, j ∈ [1, m], is the monitored concentration of the jth gas, is the concentration-based threshold value of the jth gas, represents the selection of the maximum value in 0, is the weight coefficient of the jth gas corresponding to the energy storage battery risk type Type. acquiring the real-time operation coefficient R(t) of the energy storage battery in the current region based on the operation data of the energy storage battery in each region, determining the real-time state coefficient S(t) of each energy storage battery based on the pressure information and the temperature information, and acquiring the synchronization coefficient X of the real-time state coefficient S(t) and the real-time operation coefficient R(t); comparing the gas matching coefficient g with a preset threshold value gt: if g≥gt, it is judged that there is a risk failure of the risk type Type of the energy storage battery in the region, and the energy storage battery with the risk failure is determined based on the synchronization coefficient X; if g<gt, it is determined whether there is an energy storage battery with a risk failure based on the size of the synchronization coefficient X, and the energy storage battery with the risk failure is determined when it is judged that there is.

2. The power storage system based on the intelligent Internet of Things according to claim 1, characterized in that, the process of acquiring the synchronization coefficient comprises: selecting the real-time state coefficient S(t) and the real-time operation coefficient R(t) corresponding to n time points at fixed time intervals; The synchronization coefficient X is calculated by the formula X = (A - B) / (A + B) where i e [1, n], is the ith time point. 3.The power storage system based on the intelligent Internet of Things according to claim 1, wherein, the process of acquiring the real-time operation coefficient R(t) of the energy storage battery in the current region comprises: The real-time operation coefficient R(t) is calculated by the formula ​ wherein q is the total number of parameters of the energy storage battery, k ∈ [1, q], is the real-time value of the kth parameter, f is a defined function, and when , , otherwise, , is the corresponding unit value, , is the boundary of the basic threshold value, represents selecting the minimum value in and , is the influence weight of the kth parameter.

4. The power storage system based on the intelligent Internet of Things according to claim 1, characterized in that, the process of acquiring the real-time state coefficient S(t) of the energy storage battery in the current region comprises: The real-time state coefficient S(t) of the current area energy storage battery is calculated by the formula ​ wherein, is the real-time temperature of the xth monitoring point, is the base temperature value of the xth monitoring point, is the maximum value of the temperature unit value of all monitoring points in the monitoring period, is the maximum value of the temperature unit value of all monitoring points in the monitoring period, is the temperature unit value, is the base value of the temperature rise rate, is the maximum value of the temperature unit value of all monitoring points in the monitoring period, is the maximum value of the temperature unit value of all monitoring points in the monitoring period, , is a fixed parameter adjustment coefficient, F(t) is the real-time pressure on the surface of the battery, is the maximum value of F(t) in the monitoring period, is the pressure unit value.

5. The power storage system based on the intelligent Internet of Things according to claim 1, characterized in that, when g≥gt, the process of judging the energy storage battery with the risk failure comprises: comparing all the synchronization coefficients X of the energy storage batteries with a preset threshold value Xt: if there is an energy storage battery with X<Xt, all the energy storage batteries with X<Xt are judged as the energy storage battery with the risk failure; if there is no energy storage battery with X<Xt, the energy storage battery with the smallest synchronization coefficient X is judged as the energy storage battery with the risk failure. 6.The power storage system based on the intelligent Internet of Things according to claim 1, wherein, when g<gt, the process of judging the energy storage battery with the risk failure comprises: comparing the synchronization coefficient X with a preset threshold value Xt: if there is an energy storage battery with X<Xt, all the energy storage batteries with X<Xt are judged as the energy storage battery with the risk failure; if there is no energy storage battery with X<Xt, it is judged that all the energy storage batteries in the region are in normal state.

Images