Method and system for dynamically monitoring and evaluating service life of battery of energy storage power station

Abstract

The invention discloses an energy storage power station battery life dynamic monitoring evaluation method and system, and relates to the technical field of electrochemical energy storage power station safety monitoring, and the method comprises the steps: obtaining the nanoparticle concentration, characteristic gas data and electrochemical parameters in a battery cabin of a to-be-monitored energy storage power station; carrying out pretreatment and data fusion on the nanoparticle concentration, the characteristic gas data and the electrochemical parameters to obtain multi-source fusion data; establishing a battery life prediction model of the to-be-monitored energy storage power station based on the long short-term memory network, and training to obtain a trained battery life prediction model; and inputting the multi-source fusion data into the trained battery life prediction model to obtain the battery residual life and the battery health state of the to-be-monitored energy storage power station. According to the method, the technical problems of response lag, single data dimension and incapability of realizing dynamic real-time analysis in the prior art are relieved.

Description

Technical Field

[0001] The present invention relates to the technical field of safety monitoring of electrochemical energy storage power stations, and specifically to a method and system for dynamically monitoring and evaluating the battery life of an energy storage power station. Background Technique

[0002] Lithium-ion batteries are electrochemical systems with complex non-linear characteristics. Their thermal runaway process will exhibit different signal characteristics in multiple physical fields. At present, the battery life monitoring of energy storage power stations mainly relies on traditional single-threshold warning devices. By monitoring parameters such as the voltage, current, and temperature of the battery, the early key characteristics in the evolution process of battery thermal runaway are extracted to judge the health status of the battery, so as to achieve early active safety warning. This safety warning method based on signal characteristics mostly uses single-parameter threshold warning (such as voltage and temperature exceeding the limit to trigger an alarm). This traditional system relies on passive sensors to detect sudden changes in smoke or temperature, and the response lags behind the thermal runaway stage (for example, it cannot give an early warning in time within a few seconds after the battery safety valve opens).

[0003] The safety warning method based on statistics does not consider the specific causes of battery failure, but uses battery big data to establish a statistical algorithm. By inputting a large amount of data, abnormal signal values or the change trend of abnormal signals in the data are captured, so as to predict the battery health degree in school. Finally, an alarm for abnormal values is given according to statistical principles. This method is highly dependent on data and requires continuous updating and optimization of the algorithm to adapt to new battery types and working conditions.

[0004] Most of the existing technologies are passive detections, lacking the ability to actively detect internal microparticles of the battery (such as nanoscale particles and electrolyte volatile gases generated by thermal runaway), and the evaluation algorithms are mostly fixed-period calculations (such as updating the health degree every 60 minutes), and cannot achieve dynamic real-time analysis. Summary of the Invention

[0005] The purpose of the present invention is to provide a method and system for dynamically monitoring and evaluating the battery life of an energy storage power station in order to solve at least one of the above technical problems.

[0006] In the first aspect, an embodiment of the present invention provides a method for dynamically monitoring and evaluating the battery life of an energy storage power station, including: obtaining the concentration of nano-particles, characteristic gas data, and electrochemical parameters in the battery compartment of the energy storage power station to be monitored; preprocessing and data fusion of the concentration of nano-particles, the characteristic gas data, and the electrochemical parameters to obtain multi-source fusion data; establishing a battery life prediction model for the energy storage power station to be monitored based on a long short-term memory network and training to obtain a trained battery life prediction model; inputting the multi-source fusion data into the trained battery life prediction model to obtain the remaining battery life and battery health status of the energy storage power station to be monitored.

[0007] Further, the characteristic gas data includes: hydrogen concentration, carbon monoxide concentration, carbon dioxide concentration, and volatile organic compound concentration; the electrochemical parameters include cell voltage, cell current, cell internal resistance, and cell temperature inconsistency.

[0008] Further, preprocess and data fusion are performed on the nanoparticle concentration, the characteristic gas data, and the electrochemical parameters to obtain multi-source fusion data, including: filtering and denoising the nanoparticle concentration, the characteristic gas data, and the electrochemical parameters, eliminating environmental interference, and performing time series synchronization based on timestamp alignment to obtain the multi-source fusion data.

[0009] Further, a battery life prediction model for the energy storage power station to be monitored is established based on a long short-term memory network, including: establishing a first quantitative relationship between the characteristic gas data and the battery health state:

[0010]

[0011] In the formula, C i (t) is the gas concentration at the current moment, and C0 is the gas concentration at each gas in the initial state of the battery; establish a second quantitative relationship between the electrochemical parameters and the battery health state:

[0012]

[0013] In the formula, w and j i are matrix coefficients, and f ji (t) is each electrochemical variable; based on the first quantitative relationship and the second quantitative relationship, establish a comprehensive battery health state evaluation model:

[0014] SOH(t) = β1·SOH EC (t) + β2·SOH gos (t)

[0015] In the formula, β1 and β2 are weight coefficients; based on the comprehensive battery health state evaluation model and the long short-term memory network, establish the battery life prediction model; among them, the loss function of the battery life prediction model during training includes:

[0016]

[0017] In the formula, L is the loss function, and SOH true is the true value of the battery health state in the training set.

[0018] Further, the method further includes: determining the warning level information of the energy storage power station to be monitored based on the nanoparticle concentration and the characteristic gas data.

[0019] Second aspect, an embodiment of the present invention further provides a dynamic monitoring and evaluation system for the battery life of an energy storage power station, including: an acquisition module, a preprocessing module, a building module, and a prediction module; wherein, the acquisition module is used to acquire the nanoparticle concentration, characteristic gas data, and electrochemical parameters in the battery compartment of the energy storage power station to be monitored; the preprocessing module is used to preprocess and fuse the nanoparticle concentration, the characteristic gas data, and the electrochemical parameters to obtain multi-source fusion data; the building module is used to build a battery life prediction model for the energy storage power station to be monitored based on a long short-term memory network and perform training to obtain a trained battery life prediction model; the prediction module is used to input the multi-source fusion data into the trained battery life prediction model to obtain the remaining battery life and battery health status of the energy storage power station to be monitored.

[0020] Further, the preprocessing module is further used to filter and denoise the nanoparticle concentration, the characteristic gas data, and the electrochemical parameters, eliminate environmental interference, and perform time series synchronization based on timestamp alignment to obtain the multi-source fusion data.

[0021] Further, it further includes an early warning module, which is used to determine the early warning level information of the energy storage power station to be monitored based on the nanoparticle concentration and the characteristic gas data.

[0022] Third aspect, an embodiment of the present invention further provides an electronic device, including: a memory, a processor, and a computer program stored on the memory and executable on the processor, and when the processor executes the computer program, it implements the method provided by the embodiment of the present invention.

[0023] Fourth aspect, an embodiment of the present invention further provides a computer-readable storage medium, where the computer-readable storage medium stores computer instructions, and when the computer instructions are executed by a processor, they implement the method provided by the embodiment of the present invention.

[0024] The present invention provides a method and system for dynamically monitoring and evaluating the battery life of an energy storage power station. By combining multi-dimensional real-time collection of nanoparticles, characteristic gases, and electrochemical parameters, it breaks through the limitations of traditional single-parameter monitoring. It captures the time series changes of the battery state through a long short-term memory network, predicts the battery health in real time, improves the evaluation accuracy, and can trigger an early warning at the very early stage of thermal runaway based on the microparticle concentration and characteristic gas concentration, improving the response speed and alleviating the technical problems of response lag, single data dimension, and inability to achieve dynamic real-time analysis existing in the prior art. Description of the Drawings

[0025] To more clearly illustrate the technical solutions in the embodiments of the present application or the prior art, the following will briefly introduce the drawings required for use in the specific embodiments or the description of the prior art. Obviously, the drawings in the following description are some embodiments of the present application. For those of ordinary skill in the art, without creative efforts, other drawings can also be obtained based on these drawings.

[0026] Figure 1 It is a flowchart of a method for dynamically monitoring and evaluating the battery life of an energy storage power station provided by an embodiment of the present invention;

[0027] Figure 2 It is a schematic diagram of a system for dynamically monitoring and evaluating the battery life of an energy storage power station provided by an embodiment of the present invention. Specific Embodiments

[0028] The following will clearly and completely describe the technical solutions in the embodiments of the present invention with reference to the drawings in the embodiments of the present invention. Obviously, the described embodiments are only some embodiments of the present invention, rather than all embodiments. Based on the embodiments of the present invention, all other embodiments obtained by those of ordinary skill in the art without creative efforts belong to the scope of protection of the present invention.

[0029] Embodiment 1

[0030] Figure 1 It is a flowchart of a method for dynamically monitoring and evaluating the battery life of an energy storage power station provided by an embodiment of the present invention. As Figure 1 shown, the method specifically includes the following steps:

[0031] Step S102, obtain the concentration of nanoparticles, characteristic gas data, and electrochemical parameters in the battery compartment of the energy storage power station to be monitored.

[0032] Specifically, the concentration of nanoparticles (0.01 - 20 μm) and characteristic gas data in the battery compartment are collected in real time through an active inhalation gas monitoring device (such as an aspirating particulate sensor, a red and blue light dual-source detector, etc., meeting GB15631-2008); the electrochemical parameters of the battery are obtained synchronously.

[0033] In some alternative embodiments provided by the embodiments of the present invention, the characteristic gas data includes: hydrogen concentration, carbon monoxide concentration, carbon dioxide concentration, and volatile organic compound concentration; the electrochemical parameters include battery voltage, battery current, battery internal resistance, and battery temperature inconsistency.

[0034] Step S104, perform preprocessing and data fusion on the nanoparticle concentration, characteristic gas data, and electrochemical parameters to obtain multi-source fusion data.

[0035] Specifically, the concentration of nanoparticles, the characteristic gas data, and the electrochemical parameters are filtered and denoised to eliminate environmental interference (such as dust and water mist), and time series synchronization is performed based on timestamp alignment to obtain multi-source fusion data.

[0036] Step S106: Establish a battery life prediction model for the energy storage power station to be monitored based on a long short-term memory network and perform training to obtain a trained battery life prediction model.

[0037] Step S108: Input the multi-source fusion data into the trained battery life prediction model to obtain the remaining battery life and the battery health state of the energy storage power station to be monitored.

[0038] Specifically, step S106 includes the following steps:

[0039] Step S1061: Establish a first quantization relationship between the characteristic gas data and the battery health state:

[0040]

[0041] In the formula, C i (t) is the gas concentration at the current moment, and C0 is the gas concentration of each gas in the initial state of the battery; SOH gos is the battery health state associated with the characteristic gas data.

[0042] Step S1062: Establish a second quantization relationship between the electrochemical parameters and the battery health state:

[0043]

[0044] In the formula, w and j i are matrix coefficients, and f ji (t) is each electrochemical variable; SOH EC is the battery health state associated with the electrochemical parameters.

[0045] Step S1063: Based on the first quantization relationship and the second quantization relationship, establish a comprehensive battery health state evaluation model:

[0046] SOH(t) = β1·SOH EC (t) + β2·SOH gos (t)

[0047] In the formula, β1 and β2 are weight coefficients determined during model training; SOH is the comprehensive battery health state evaluation model.

[0048] Step S1064: Based on the comprehensive battery health state evaluation model and the long short-term memory network, establish a battery life prediction model; among them, the loss function during the training of the battery life prediction model includes:

[0049]

[0050] Wherein, L is the loss function, and SOH true is the true value of the state of health of the battery in the training set.

[0051] In some optional embodiments provided by the embodiments of the present invention, the battery life prediction model is a double-layer LSTM neural network model: h t = LSTM(x t , h t-1 ); wherein,

[0052] The first layer extracts temporal features: β i = Softmax(W β h t + b β );

[0053] The second layer performs dynamic weight allocation, and its closed-loop control strategy is:

[0054]

[0055] Specifically, the training process of the battery life prediction model includes: constructing a real-time database with multi-source fusion data, converting the multi-source fusion data into a temporal feature data matrix, splitting the temporal feature data matrix, selecting 80% to construct a weight coefficient matrix for the machine learning algorithm, selecting the data of the first 30 timestamps before the current moment to construct a data slider, predicting the state of health of the battery at the subsequent specified timestamp, and comparing the state of health of the battery with the grading standard to obtain the health level. Among them, the update period of the cloud intelligent analysis platform ≤ 30 minutes.

[0056] Specifically, the embodiments of the present invention use an LSTM neural network to construct a dynamic battery life prediction model according to the battery health balcony comprehensive evaluation model, combine dynamic health features, calculate the remaining battery life and the state of health of the battery in real time, and set the update period to no more than 30 minutes.

[0057] Specifically, the method provided by the embodiments of the present invention further includes: determining the early warning level information of the energy storage power station to be monitored based on the nanoparticle concentration and characteristic gas data. For example, setting grading early warning thresholds (early warning, alarm, fire alarm) according to the safety regulations of GB / T42288-2022, and linking with the fire protection system (starting the exhaust, fire extinguishing devices, etc.).

[0058] In some optional embodiments provided by the embodiments of the present invention, a three-level linkage early warning system is established as follows:

[0059] 1. Early warning level (±15% mutation of nanoparticle concentration / 10 min), start the exhaust system;

[0060] 2. Alarm level (H2 > 200 ppm and temperature rise rate > 5 °C / min), cut off the circuit;

[0061] 3. Fire alarm level (CO concentration shows exponential growth), trigger the release of perfluorocyclohexanone fire extinguishing agent.

[0062] Optionally, the method provided by the embodiments of the present invention further includes: displaying real-time data curves, health heat maps and maintenance suggestions through a WEB interface; generating monthly / annual statistical reports to support battery replacement decisions.

[0063] As can be seen from the above description, the present invention provides a method for dynamically monitoring and evaluating the battery life of an energy storage power station. Compared with the prior art, it has the following technical effects:

[0064] (1) Active multi-source data fusion: Combining multi-dimensional real-time collection of nanoparticle, characteristic gas and electrochemical parameters, breaking through the limitations of traditional single-parameter monitoring;

[0065] (2) Dynamic algorithm optimization: Adopting a short-cycle (≤ 30 minutes) update mechanism, capturing the temporal changes of battery status through an LSTM neural network, and real-time predicting the battery health degree to improve the evaluation accuracy;

[0066] (3) Early warning mechanism: Based on the microparticle concentration and characteristic gas concentration, triggering an early warning at the very early stage of thermal runaway, with a response speed improved by more than 90% compared with the prior art.

[0067] Embodiment 2

[0068] Figure 2 It is a schematic diagram of a system for dynamically monitoring and evaluating the battery life of an energy storage power station according to an embodiment of the present invention. As Figure 2 shown, the system includes: an acquisition module 10, a preprocessing module 20, a building module 30 and a prediction module 40.

[0069] Specifically, the acquisition module 10 is used to acquire the nanoparticle concentration, characteristic gas data and electrochemical parameters in the battery compartment of the energy storage power station to be monitored.

[0070] The preprocessing module 20 is used to preprocess and fuse the nanoparticle concentration, characteristic gas data and electrochemical parameters to obtain multi-source fusion data.

[0071] The building module 30 is used to establish a battery life prediction model for the energy storage power station to be monitored based on a long short-term memory network and perform training to obtain a trained battery life prediction model.

[0072] The prediction module 40 is used to input the multi-source fusion data into the trained battery life prediction model to obtain the remaining battery life and battery health status of the energy storage power station to be monitored.

[0073] Specifically, the preprocessing module 20 is further configured to perform filtering and noise reduction on the nanoparticle concentration, characteristic gas data, and electrochemical parameters, eliminate environmental interference, and perform time series synchronization based on timestamp alignment to obtain multi-source fusion data.

[0074] Specifically, as Figure 2 shown, the system provided by the embodiment of the present invention further includes an early warning module 50, which is configured to determine the early warning level information of the energy storage power station to be monitored based on the nanoparticle concentration and characteristic gas data.

[0075] In some optional embodiments provided by the embodiments of the present invention, the energy storage power station battery life dynamic monitoring and evaluation system is deployed on a cloud platform. The acquisition module 10 obtains the nanoparticle concentration, characteristic gas data, and electrochemical parameters through an active inhalation gas monitoring device (such as a pipeline, a detector host, etc.), and then transmits the detection data to the cloud platform through the communication module for network transmission. Preferably, the communication module supports RS485 / Ethernet / 4G. Then, the remaining battery life and the battery health status are predicted through the cloud platform and transmitted to the monitoring computer in the control room, and the concentration situation and early warning information can be displayed in real time on the monitor.

[0076] Optionally, the on-line monitoring host should have a capacitive touch screen, provide a good man-machine interaction interface, support on-site operation and testing, and can display the real-time change curves of the pyrolysis particle concentration and characteristic gas in real time, facilitating the viewing of the change trend of the monitored quantity. And a cloud intelligent analysis platform to form a complete dynamic monitoring network.

[0077] The present invention also provides an electronic device, including: a memory, a processor, and a computer program stored on the memory and executable on the processor. When the processor executes the computer program, the method provided by the embodiment of the present invention is implemented.

[0078] The present invention also provides a computer-readable storage medium, which stores computer instructions. When the computer instructions are executed by a processor, the method provided by the embodiment of the present invention is implemented.

[0079] For those skilled in the art, it is obvious that the present invention is not limited to the details of the above exemplary embodiments, and without departing from the spirit or basic characteristics of the present invention, the present invention can be implemented in other specific forms. Therefore, in any regard, the embodiments should be regarded as exemplary and non-limiting. The scope of the present invention is defined by the appended claims rather than the above description. Therefore, it is intended to embrace all changes within the meaning and scope of the equivalent elements of the claims in the present invention. Any reference signs in the claims should not be regarded as limiting the claimed rights.

[0080] In addition, it should be understood that although this specification is described according to embodiments, not every embodiment only contains an independent technical solution. This narrative way of the specification is only for clarity. Those skilled in the art should regard the specification as a whole, and the technical solutions in each embodiment can also be appropriately combined to form other embodiments that can be understood by those skilled in the art.

Claims

1. A method for dynamic monitoring and evaluation of battery life of an energy storage power station, characterized in that: include: Obtain the nanoparticle concentration, characteristic gas data and electrochemical parameters in the battery compartment of the energy storage power station to be monitored; Preprocessing and data fusion of the nanoparticle concentration, the characteristic gas data and the electrochemical parameters to obtain multi-source fusion data; Establishing a battery life prediction model for the energy storage power station to be monitored based on a long short-term memory network, and performing training to obtain a trained battery life prediction model; The multi-source fusion data is input into the trained battery life prediction model to obtain the remaining battery life and battery health status of the energy storage power station to be monitored.

2. The method according to claim 1, characterized in that: The characteristic gas data include: hydrogen concentration, carbon monoxide concentration, carbon dioxide concentration and volatile organic compound concentration; the electrochemical parameters include battery voltage, battery current, battery internal resistance and battery temperature inconsistency.

3. The method according to claim 1, characterized in that: The nanoparticle concentration, the characteristic gas data and the electrochemical parameters are preprocessed and data fused to obtain multi-source fused data, including: filtering and denoising the nanoparticle concentration, the characteristic gas data and the electrochemical parameters, eliminating environmental interference, and performing time series synchronization based on timestamp alignment to obtain the multi-source fused data.

4. The method according to claim 1, characterized in that: A battery life prediction model for the energy storage power station to be monitored is established based on a long short-term memory network, including: Establishing a first quantitative relationship between the characteristic gas data and the battery health status: In the formula, C i (t) is the gas concentration at the current moment, C0 is the gas concentration in the initial state of the battery; A second quantitative relationship between the electrochemical parameter and the battery health state is established: In the formula, w and j i is the matrix coefficient, f ji (t) is each electrochemical variable; Based on the first quantitative relationship and the second quantitative relationship, a comprehensive evaluation model for the battery health status is established: SOH(t)=β1·SOH EC (t)+β2·SOH gos (t) In the formula, β1 and β2 are weight coefficients; Based on the battery health status comprehensive evaluation model and the long short-term memory network, the battery life prediction model is established; wherein the loss function of the battery life prediction model during training includes: Where L is the loss function, SOH true is the true value of the battery health status in the training set.

5. The method according to claim 1, characterized in that: The method further includes: determining the warning level information of the energy storage power station to be monitored based on the nanoparticle concentration and the characteristic gas data.

6. A dynamic monitoring and evaluation system for battery life of energy storage power station, characterized in that: include: Acquisition module, preprocessing module, establishment module and prediction module; among them, The acquisition module is used to obtain the nanoparticle concentration, characteristic gas data and electrochemical parameters in the battery compartment of the energy storage power station to be monitored; The preprocessing module is used to preprocess and fuse the nanoparticle concentration, the characteristic gas data and the electrochemical parameters to obtain multi-source fusion data; The establishment module is used to establish a battery life prediction model of the energy storage power station to be monitored based on a long short-term memory network, and perform training to obtain a trained battery life prediction model; The prediction module is used to input the multi-source fusion data into the trained battery life prediction model to obtain the remaining battery life and battery health status of the energy storage power station to be monitored.

7. The system according to claim 6, characterized in that: The preprocessing module is also used to filter and reduce noise on the nanoparticle concentration, the characteristic gas data and the electrochemical parameters, eliminate environmental interference, and perform time series synchronization based on timestamp alignment to obtain the multi-source fusion data.

8. The system according to claim 6, characterized in that: It also includes an early warning module, which is used to determine the early warning level information of the energy storage power station to be monitored based on the nano-particle concentration and the characteristic gas data.

9. An electronic device, characterized in that: include: A memory, a processor, and a computer program stored in the memory and executable on the processor, wherein the processor implements the method according to any one of claims 1 to 5 when executing the computer program.

10. A computer-readable storage medium, characterized in that: The computer-readable storage medium stores computer instructions, and when the computer instructions are executed by a processor, the method according to any one of claims 1 to 5 is implemented.

Images