Multi-source mixed signal energy storage power station risk early warning system and method

Abstract

The application discloses a multi-source mixed signal energy storage power station risk early warning system and method, and the method comprises the following steps: acquiring the battery group of the energy storage power station to obtain multi-source data, including infrared video data, battery operation parameters and alarm data; identifying the surface temperature of the battery and external heat sources by using the infrared video data, and establishing an external thermal runaway risk early warning model by monitoring the temperature change and identifying the external heat sources; establishing an internal risk early warning model based on the battery operation parameters and the alarm data; determining the weights of the external thermal runaway risk early warning model and the internal risk early warning model, combining the models by weighted summation to obtain an energy storage power station risk early warning model; acquiring the multi-source data of the battery group of a target energy storage power station, inputting the data into the risk early warning model, and outputting a risk early warning result. The application uses multi-source samples and multi-dimensional models to early warn the thermal runaway of the battery of the energy storage power station, and guarantees the accuracy of the battery health score and the effectiveness of the thermal runaway early warning in the real operating environment.

Description

TECHNICAL FIELD

[0001] The present application relates to the technical field of data processing, in particular to a multi-source mixed signal energy storage power station risk early warning system and method. BACKGROUND

[0002] Energy storage technology is a key technology to support the construction of new power systems, but the fire and explosion problems of energy storage power stations caused by lithium-ion batteries have become the main pain points restricting the development of the energy storage industry, and the safe operation risk prevention and control of energy storage stations has become increasingly important.

[0003] At present, the safety warning of domestic and foreign energy storage batteries mainly extracts current, voltage and other electrical signals to realize the rapid detection of the health status of the battery. Limited by the single battery warning model, the battery fault characteristics and fault cycle evolution prediction in the warning process are not perfect. Even if the result is detected quickly, the model is difficult to accurately judge the location, mode, severity, subsequent evolution rate and thermal runaway time of the fault, and the algorithm cannot provide a reasonable response strategy. SUMMARY

[0004] Therefore, in order to solve the problems existing in the prior art, the present application provides a multi-source mixed signal energy storage power station risk early warning system and method, which guarantees the accuracy of the battery health score and the effectiveness of the thermal runaway early warning in the real operating environment.

[0005] To achieve the above purpose, the present application provides the following technical scheme:

[0006] In a first aspect, the present application provides a multi-source mixed signal energy storage power station risk early warning method, comprising:

[0007] Obtaining the multi-source data of the battery pack of the energy storage power station, the multi-source data including: infrared video data, operating parameters of the battery and alarm data;

[0008] Using the infrared video data to identify the surface temperature of the battery and external heat sources, and establishing an external thermal runaway risk early warning model by monitoring the temperature change of the battery pack in different working states and identifying the presence or absence of external heat sources;

[0009] Establishing an internal risk early warning model based on the operating parameters and alarm data of the battery;

[0010] Determining the weights of the external thermal runaway risk early warning model and the internal risk early warning model, and combining the models by weighted summation to obtain the final energy storage power station risk early warning model;

[0011] Obtaining the multi-source data of the battery pack of the target energy storage power station, and inputting the data into the energy storage power station risk early warning model to output a risk early warning result.

[0012] Further, the step of identifying the surface temperature of the battery and the external heat source using the infrared video data comprises:

[0013] Fusing the visible light and infrared light images in the infrared video data to obtain a fused image;

[0014] Graying the fused image and correcting it to obtain a normalized image, and using the maximum inter-class variance method to perform image binarization processing to obtain a high-temperature region image;

[0015] Extracting the contour in the light image and combining it with the high-temperature region image to perform image cutting to obtain the high-temperature distribution of each battery pack;

[0016] Using a pixel accumulation method to locate a rectangular frame, taking the long side of the rectangular frame as the direction, accumulating the continuous pixels in the column of the overall image, and screening out the column whose continuous pixels are equal to the length of the rectangular frame, and at the same time, taking the short side of the rectangular frame as the reference, locating the pixel coordinates of the four corners of the rectangular frame;

[0017] Positioning the position of the ROI according to the relative position relationship between the rectangular frame and the temperature value and segmenting to obtain the maximum and minimum temperature values of the image;

[0018] Obtaining the temperature value in the image by comparing the proportional relationship between the size of the pixel point in the image and the position of the pixel value in the rectangular frame and the maximum and minimum values;

[0019] Taking the sum of the pixel points in the high-temperature region of each battery pack as the area size, using Gaussian approximation to fit the area change, and obtaining the area change rate.

[0020] Further, the process of fusing the visible light and infrared light images in the infrared video data comprises:

[0021] Loading all parameters of the infrared video acquisition device, and using the input distance to construct a virtual chessboard;

[0022] Transforming the virtual chessboard into the image space of the two acquisition devices for matching, and using the matched points to obtain a transformation matrix and a transformation mapping table;

[0023] Converting the image according to the transformation matrix and the transformation mapping table and inputting;

[0024] Eliminating the low-temperature region in the infrared light image, transforming the visible light image to the position of the thermal imager for image fusion, converting the image and outputting to obtain the fused image.

[0025] Further, the process of establishing an external thermal runaway risk early warning model comprises:

[0026] If at least one of the following conditions is monitored, it is determined that the battery pack has a risk of thermal runaway: an external heat source is close to the battery pack, the highest temperature point is located at the positive and negative electrodes, and the highest temperature point of the battery pack is located at the positive and negative electrodes while the area change rate of high temperature is greater than a preset threshold.

[0027] Further, the alarm data is converted into discrete digital data by one-hot encoding, and is combined with the battery operating state parameter data leading to thermal runaway.

[0028] By clustering analysis on the combined data, different types of risk situations are obtained.

[0029] The risk situations are labeled by an expert database to obtain an internal risk early warning model based on a clustering algorithm.

[0030] Real-time data of the battery pack operation are input into the model to determine whether the battery will have thermal runaway and obtain the cause of the thermal runaway and the risk level result.

[0031] According to different causes of thermal runaway and risk level results, the early warning result is converted into a corresponding risk score.

[0032] Further, the risk score is determined by the location of the thermal runaway occurrence point, the difference between the current temperature and the expected thermal runaway temperature, and the risk level.

[0033] Further, the entropy weight method is used to determine the weights of the external thermal runaway risk early warning model and the internal risk early warning model, and the final energy storage power station risk early warning model is obtained by weighted summation of the model combination. According to the size of the final early warning result, the threshold of each level is divided to carry out graded risk early warning.

[0034] In a second aspect, the embodiments of the present application provide a multi-source mixed signal energy storage power station risk early warning system, which comprises:

[0035] A multi-source data acquisition module is used to acquire multi-source data of the battery pack of the energy storage power station, and the multi-source data includes infrared video data, operating parameters of the battery, and alarm data.

[0036] An external thermal runaway risk early warning model establishment module is used to identify the surface temperature of the battery and external heat sources by using the infrared video data, and to establish an external thermal runaway risk early warning model by monitoring the temperature change of the battery pack in different working states and identifying the presence or absence of external heat sources.

[0037] An internal risk early warning model establishment module is used to establish an internal risk early warning model based on the operating parameters and alarm data of the battery.

[0038] The energy storage power station risk early warning model establishing module is configured to determine weights of the external thermal runaway risk early warning model and the internal risk early warning model, and combine the models by weighted summation to obtain the final energy storage power station risk early warning model.

[0039] The risk early warning result output module is configured to acquire multi-source data of a battery pack of a target energy storage power station, and input the multi-source data into the energy storage power station risk early warning model to output a risk early warning result.

[0040] In a third aspect, an embodiment of the present application provides a computer device, which includes at least one processor and a memory connected to the at least one processor in communication, wherein the memory stores instructions executable by the at least one processor, and the instructions are executed by the at least one processor to enable the at least one processor to perform the energy storage power station risk early warning method of multi-source mixed signals according to the first aspect of the present application.

[0041] In a fourth aspect, an embodiment of the present application provides a computer readable storage medium, which stores computer instructions, and the computer instructions are used to enable a computer to perform the energy storage power station risk early warning method of multi-source mixed signals according to the first aspect of the present application.

[0042] The technical scheme of the present application has the following advantages:

[0043] The energy storage power station risk early warning system and method of multi-source mixed signals include the following steps: acquiring multi-source data of a battery pack of an energy storage power station, including infrared video data, operating parameters of the battery, and alarm data; identifying surface temperature of the battery and external heat sources by using the infrared video data, and establishing an external thermal runaway risk early warning model by monitoring temperature changes and identifying external heat sources; establishing an internal risk early warning model based on the operating parameters of the battery and the alarm data; determining weights of the external thermal runaway risk early warning model and the internal risk early warning model, and combining the models by weighted summation to obtain an energy storage power station risk early warning model; and acquiring multi-source data of a battery pack of a target energy storage power station, and inputting the multi-source data into the risk early warning model to output a risk early warning result. The present application uses multi-source samples and multi-dimensional models to perform early warning on thermal runaway of a battery of an energy storage power station, and ensures accuracy of a battery health score and effectiveness of thermal runaway early warning in a real operating environment. BRIEF DESCRIPTION OF DRAWINGS

[0044] In order to more clearly illustrate the specific embodiments of the present application or the technical solutions in the prior art, the following will briefly introduce the drawings needed to be used in the specific embodiments or prior art description. Obviously, the drawings in the following description are some embodiments of the present application, and other drawings can also be obtained by those skilled in the art without creative labor.

[0045] Figure 1A flow chart of one specific example of the multi-source mixed signal energy storage power station risk early warning method provided in the embodiment of the present application;

[0046] Figure 2 A module composition diagram of one example of the multi-source mixed signal energy storage power station risk early warning system provided in the embodiment of the present application;

[0047] Figure 3 A composition diagram of one specific example of the computer device provided in the embodiment of the present application. DETAILED DESCRIPTION

[0048] The technical solutions of the present application will be described clearly and completely below with reference to the drawings. Obviously, the described embodiments are part of the embodiments of the present application, rather than all the embodiments. Based on the embodiments in the present application, all other embodiments obtained by those skilled in the art without creative labor fall within the scope of protection of the present application.

[0049] In addition, the technical features involved in different embodiments of the present application described below can be combined with each other as long as they do not conflict with each other.

[0050] Embodiment 1

[0051] The multi-source mixed signal energy storage power station risk early warning method provided in the embodiment of the present application, as shown in the figure, comprises the following steps: Figure 1

[0052] Step S1: Obtain the multi-source data of the battery pack of the energy storage power station, which includes infrared video data, operating parameters of the battery and alarm data.

[0053] The embodiment of the present application is based on the early warning server on the side of the plant station, installs a high-definition camera and a thermal imager, a temperature sensor in the energy storage battery cabin to collect light signals and heat, and interfaces with the energy storage battery management system to collect battery information. The types of collected data include image heat distribution data, alarm data, real-time voltage of battery monomer, real-time current of battery monomer, real-time temperature of battery monomer, real-time voltage of battery cluster, real-time current of battery cluster, etc.

[0054] Step S2: Identify the surface temperature of the battery and the external heat source by using the infrared video data, establish an external thermal runaway risk early warning model by monitoring the temperature change of the battery pack in different working states and identifying the presence or absence of external heat sources.

[0055] ​Specifically, by means of the high-definition camera and the thermal imager, the thermal radiation information in the infrared light image is combined with the contour information, texture information and gradient information in the visible light image in a manner of infrared light and visible light image fusion, so as to realize the collection of the surface temperature of all the batteries in the energy storage power station and the accurate identification of the external heat source. By monitoring the temperature change of the battery pack in different working states of charging and discharging and identifying the presence or absence of the external heat source, an external thermal runaway risk early warning model is established.

[0056] In the embodiment of the present application, the step of identifying the surface temperature of the battery and the external heat source by means of the infrared video data comprises:

[0057] Step S21: fuse the visible light and the infrared light image in the infrared video data to obtain a fused image. Specifically, the pixel gray value weighted average method is selected to fuse the image in the present application, which can effectively improve the signal-to-noise ratio of image fusion, the algorithm is simple and fast, the accurate registration and fusion of images can be realized based on OpenCV, and the image fusion step is as follows:

[0058] (1) load all the parameters of the infrared video acquisition device, and construct a virtual chessboard by means of the input distance; wherein the input distance can be reasonably set according to actual needs, which is not specifically limited here.

[0059] (2) transform the virtual chessboard into the image space of the two acquisition devices for matching, and use the matched points to obtain the transformation matrix and the transformation mapping table;

[0060] (3) input the image according to the transformation matrix and the transformation mapping table;

[0061] (4) eliminate the area with lower temperature in the infrared light image, transform the visible light image to the position of the thermal imager for image fusion, and output the fused image after image coding.

[0062] Step S22: gray the fused image and correct it (for example, Gamma correction) to obtain a normalized image, and use the maximum inter-class variance method to perform image binarization processing to obtain a high-temperature area image;

[0063] Step S23: extract the contour in the visible light image, combine it with the high-temperature area image, and then perform image cutting to obtain the high-temperature distribution of each battery pack; the battery pack can be identified by extracting the contour in the present application, and it is judged whether there is an external heat source in the image.

[0064] Step S24: The rectangular frame is positioned by using a pixel accumulation method, the whole image is accumulated by column according to the long side of the rectangular frame, the column with continuous pixels equal to the length of the rectangular frame is screened out, and the pixel coordinates of the four corners of the rectangular frame are positioned according to the short side of the rectangular frame.

[0065] Step S25: The position of the ROI is positioned and segmented according to the relative position relationship between the rectangular frame and the temperature value, and the maximum and minimum temperature values of the image are obtained.

[0066] Step S26: The temperature value in the image is obtained by comparing the size of the pixel point in the image with the proportional relationship of the maximum and minimum values of the pixel value in the rectangular frame.

[0067] Step S27: The total sum of the pixel points in the high-temperature area of each battery group is taken as the area size, the area change is fitted by using Gaussian approximation, and the area change rate is obtained.

[0068] Step S3: An internal risk early warning model is established based on the operation parameters and alarm data of the battery.

[0069] In order to quantify the possibility of thermal runaway risk, the thermal runaway risk probability score is represented by , and the thermal runaway risk is judged based on the following dimensions:

[0070] 1. Whether there is an external heat source close to the battery group, if yes, it is considered that there is a thermal runaway risk,

[0071] 2. Whether the highest temperature point is located at the positive and negative electrodes, if not, it is considered that a short circuit occurs in the battery group, and there may be puncture and other situations, and there is a thermal runaway risk,

[0072] 3. If the highest point temperature is located at the positive and negative electrodes, and the high-temperature area change rate ΔS> ΔS0, wherein ΔS0 is an empirical value obtained through multiple experiments, the value is related to the battery model and material, it is considered that there is a thermal runaway risk, wherein T max is the maximum temperature of the battery group, and T0 is the minimum value of the highest temperature of the battery when the thermal runaway occurs.

[0073] 4. If none of the above situations occurs, it is considered that the battery group has no thermal runaway risk,

[0074] Step S4: The weights of the external thermal runaway risk early warning model and the internal risk early warning model are determined, the final energy storage power station risk early warning model is obtained by weighted summation.

[0075] The embodiment of the application aims at the problems of battery aging and short circuit, researches the thermal runaway mechanism of the battery, finds the main features affecting the thermal runaway of the energy storage battery by combining mechanism analysis and feature data identification, divides them into various abnormal types by a clustering algorithm, labels them with early warning type based on the expert experience database on the cloud, and increases the influence of real-time alarm data to obtain an internal risk early warning model based on the operation parameters and alarm data of the energy storage battery, and the specific steps are as follows:

[0076] Step S41: convert the alarm data into discrete digital data by one-hot encoding, and combine the battery operation state parameter data leading to thermal runaway;

[0077] Step S42: obtain different types of risk situations by clustering analysis (for example, based on K-means clustering) on the combined data;

[0078] Step S43: label the risk situations by the expert database to obtain an internal risk early warning model based on the clustering algorithm;

[0079] Step S44: input the real-time data of the battery pack operation into the model to determine whether the battery will have thermal runaway and obtain the cause of the thermal runaway and the risk level result;

[0080] Step S45: convert the early warning result into a corresponding risk score according to the cause of the thermal runaway and the risk level result, and the size of the risk score is determined by the position of the thermal runaway occurrence point, the difference between the current temperature and the expected thermal runaway temperature, and the risk level.

[0081] Since the output results of each model are different, the influence is also different, and they cannot be directly superimposed, so different weights need to be assigned to the results of each model. The embodiment of the application determines the weights of the external thermal runaway risk early warning model and the internal risk early warning model by using the entropy weight method, combines the models by weighted summation to obtain the final energy storage power station risk early warning model, and according to the size of the final early warning result, the thresholds of each level are drawn according to the early warning type to realize graded early warning.

[0082] The entropy weight method is an objective weighting method, and the meaning of entropy value is that the greater the change degree between each scheme and the same index data, the greater the amount of information reflected, and the greater the corresponding weight value, and vice versa.

[0083] Suppose X represents a certain risk that causes event X to occur, and P(X) represents the probability of the occurrence of this risk, which can be defined as: I(X) = -ln[P(X)], where I(X) represents the amount of information. If the risks of event X that may occur are: x1, x2... x n , then the information entropy of event X can be defined as:

[0084]

[0085] For the j-th risk indicator, its information entropy is calculated using the following formula:

[0086]

[0087] Information utility value d j =1-e j The higher the information utility value, the greater the amount of information. After normalizing the information utility value, the entropy weight of each risk indicator is obtained as follows:

[0088]

[0089] The prediction scores of the external thermal runaway risk warning model and the internal risk warning model are combined. Substituting into the above formula, the entropy weights ν1 and ν2 of each model are calculated respectively. By combining the models through weighted summation, the final multi-source combined early warning model is obtained, as shown in the following formula:

[0090]

[0091] By comparing the magnitude of the final warning result with the warning type, thresholds at each level can be defined to achieve tiered warnings.

[0092] Step S5: Obtain multi-source data of the battery pack of the target energy storage power station, input it into the energy storage power station risk warning model, and output the risk warning result.

[0093] In practical applications, a physical connection is established between the server where the energy storage power station risk warning model is located and the application server where infrared video data and battery information are located via a bus. The power station-side application server inputs multi-source data with time stamps into the risk warning model by calling the risk warning model interface, and receives and displays the risk warning results output by the energy storage power station risk warning model in real time.

[0094] Example 2

[0095] This invention provides a risk early warning system for energy storage power stations based on multi-source hybrid signals, such as... Figure 2 As shown, it includes:

[0096] The multi-source data acquisition module 1 is used to acquire multi-source data from the battery pack of the energy storage power station. The multi-source data includes infrared video data, battery operating parameters and alarm data. This module executes the method described in step S1 of embodiment 1, which will not be repeated here.

[0097] The external thermal runaway risk early warning model establishing module 2 is configured to identify the surface temperature of the battery and external heat sources by using the infrared video data, and to establish an external thermal runaway risk early warning model by monitoring the temperature change of the battery pack in different working states and identifying the presence or absence of external heat sources. The module performs the method described in step S2 of embodiment 1, and thus will not be described again.

[0098] The internal risk early warning model obtaining module 3 is configured to establish an internal risk early warning model based on the operating parameters and alarm data of the battery. The module performs the method described in step S3 of embodiment 1, and thus will not be described again.

[0099] The energy storage power station risk early warning model establishing module 4 is configured to determine the weights of the external thermal runaway risk early warning model and the internal risk early warning model, and to obtain a final energy storage power station risk early warning model by weighted summation. The module performs the method described in step S4 of embodiment 1, and thus will not be described again.

[0100] The risk early warning result output module 5 is configured to obtain the multi-source data of the battery pack of the target energy storage power station, and to input the data into the energy storage power station risk early warning model to output a risk early warning result. The module performs the method described in step S4 of embodiment 1, and thus will not be described again.

[0101] The multi-source mixed signal energy storage power station risk early warning system provided by the present application uses multi-source samples and multi-dimensional models to early warn the thermal runaway of the battery of the energy storage power station, and ensures the accuracy of the battery health score and the effectiveness of the thermal runaway early warning in the real operating environment.

[0102] Embodiment 3

[0103] The embodiment of the present application provides a computer device, such as Figure 3As shown, it comprises at least one processor 401, such as a CPU (Central Processing Unit), at least one communication interface 403, a memory 404, and at least one communication bus 402. The communication bus 402 is used to realize the connection and communication between the components. The communication interface 403 can include a display, a keyboard, and can also include a standard wired interface and a wireless interface. The memory 404 can be a high-speed RAM (Random Access Memory) or a non-volatile memory such as at least one disk memory. The memory 404 can also be at least one storage device located away from the aforementioned processor 401. The processor 401 can execute the multi-source mixed signal energy storage power station risk early warning method of embodiment 1. The memory 404 stores a set of program codes, and the processor 401 calls the program codes stored in the memory 404 to execute the multi-source mixed signal energy storage power station risk early warning method of embodiment 1.

[0104] The communication bus 402 can be a PCI (peripheral component interconnect) bus or an EISA (extended industry standard architecture) bus, etc. The communication bus 402 can be divided into an address bus, a data bus, and a control bus, etc. For the sake of representation, Figure 3 Only one line is used in the figure, but it does not mean that there is only one bus or one type of bus.

[0105] The memory 404 can include a volatile memory such as a RAM (random-access memory), and can also include a non-volatile memory such as a flash memory, a hard disk (HDD) or a solid-state disk (SSD), and can also include a combination of the above types of memories.

[0106] The processor 401 can be a central processing unit (CPU), a network processor (NP), or a combination of the CPU and the NP.

[0107] The processor 401 can further include a hardware chip. The hardware chip can be an application-specific integrated circuit (ASIC), a programmable logic device (PLD), or a combination thereof. The PLD can be a complex programmable logic device (CPLD), a field-programmable gate array (FPGA), a generic array logic (GAL), or any combination thereof.

[0108] Optionally, the memory 404 is further configured to store program instructions. The processor 401 can invoke the program instructions to implement the method for risk early warning of a multi-source hybrid signal energy storage power station as described in Embodiment 1.

[0109] The embodiments of the present application further provide a computer readable storage medium, and the computer readable storage medium stores computer executable instructions. The computer executable instructions can execute the method for risk early warning of a multi-source hybrid signal energy storage power station as described in Embodiment 1. The storage medium can be a magnetic disc, an optical disc, a read-only memory (ROM), a random access memory (RAM), a flash memory, a hard disk drive (HDD), a solid-state drive (SSD), or the like. The storage medium can further include a combination of the above-mentioned storage mediums.

[0110] Obviously, the above embodiments are merely examples for clarity and do not limit the embodiments. Based on the above description, those skilled in the art can make other different forms of changes or modifications. All the embodiments do not need to be exhausted, and the obvious changes or modifications still fall within the protection scope of the present application.

Claims

1. A risk early warning method for energy storage power stations based on multi-source hybrid signals, characterized in that, include: Acquire multi-source data from the battery packs of the energy storage power station. The multi-source data includes: infrared video data, battery operating parameters and alarm data; This method utilizes infrared video data to identify the surface temperature of a battery and external heat sources. By monitoring the temperature changes of the battery pack under different operating conditions and identifying the presence or absence of external heat sources, an external thermal runaway risk early warning model is established. The steps for identifying the battery surface temperature and external heat sources using infrared video data include: fusing visible light and infrared light images from the infrared video data to obtain a fused image; converting the fused image to grayscale and then correcting it to obtain a normalized image; performing image binarization using the maximum inter-class variance method to obtain the high-temperature region image; extracting the contours from the visible light image and combining them with the high-temperature region image, then segmenting the image to obtain each... High-temperature distribution of the battery pack: A pixel accumulation method is used to locate rectangular boxes. Using the long side of the rectangle as the direction, consecutive pixels are accumulated column by column across the entire image. Columns with consecutive pixels equal to the length of the rectangle are selected. Simultaneously, using the short side of the rectangle as a reference, the pixel coordinates of the four corners of the rectangle are located. The position of the Region of Interest (ROI) is located and segmented based on the relative position of the rectangle and the temperature values, obtaining the maximum and minimum temperature values ​​of the image. The temperature values ​​in the image are obtained by comparing the pixel size in the image with the ratio of the pixel position in the rectangle to the maximum and minimum values. The total number of pixels in the high-temperature region of each battery pack is used as its area size. Gaussian approximation is used to fit the area change to obtain its area change rate. An internal risk warning model is established based on battery operating parameters and alarm data. Alarm data is converted into discrete digital data through unique thermal encoding and then combined with battery operating status parameters that lead to thermal runaway. By performing cluster analysis on the merged data, different types of risk scenarios can be identified. By labeling risk scenarios using an expert database, an internal risk warning model based on clustering algorithms is obtained; The real-time data of the battery pack operation is input into the model to determine whether the battery will experience thermal runaway, the causes of thermal runaway, and the risk level. Based on the different causes of thermal runaway and the different risk levels, the warning results are converted into corresponding risk scores; The weights of the external thermal runaway risk warning model and the internal risk warning model are determined, and the models are combined by weighted summation to obtain the final energy storage power station risk warning model. The multi-source data of the battery pack of the target energy storage power station is obtained and input into the risk warning model of the energy storage power station to output the risk warning result.

2. The method for risk early warning of energy storage power stations using multi-source hybrid signals according to claim 1, characterized in that, The process of fusing visible light and infrared light images from infrared video data includes: Load all parameters of the infrared video acquisition device and construct a virtual chessboard using the input distance; The virtual chessboard is transformed into the image space of the two acquisition devices for matching, and the transformation matrix and transformation mapping table are obtained using the matched points. The image is transcoded and input based on the transformation matrix and transformation mapping table; The low-temperature areas in the infrared image are removed, the visible light image is transformed to the position of the thermal imager for image fusion, and the image is transcoded and output to obtain the fused image.

3. The risk early warning method for energy storage power stations using multi-source hybrid signals according to claim 1, characterized in that, The process of establishing an external thermal runaway risk early warning model includes: If at least one of the following conditions is detected: an external heat source is approaching the battery pack, the highest temperature point is located at the positive or negative electrode, or the highest temperature point of the battery pack is located at the positive or negative electrode and the rate of change of the high-temperature area is greater than a preset threshold, then it is determined that the battery pack has a risk of thermal runaway.

4. The risk warning method for energy storage power stations using multi-source hybrid signals according to claim 1, wherein the risk score is determined by the location of the thermal runaway point, the difference between the current temperature and the expected thermal runaway temperature, and the risk level.

5. The risk early warning method for energy storage power stations using multi-source hybrid signals according to claim 1, characterized in that, The weights of the external thermal runaway risk warning model and the internal risk warning model are determined by using the entropy weight method. The models are combined by weighted summation to obtain the final risk warning model for the energy storage power station. Based on the magnitude of the final warning result, the thresholds of each level are determined according to the warning type to carry out graded risk warning.

6. A risk early warning system for energy storage power stations using multi-source hybrid signals, characterized in that, The method based on claim 1 includes: The multi-source data acquisition module is used to acquire multi-source data from the battery pack of the energy storage power station. The multi-source data includes: infrared video data, battery operating parameters and alarm data. The module for establishing an external thermal runaway risk warning model is used to identify the surface temperature of the battery and external heat sources using infrared video data. By monitoring the temperature changes of the battery pack under different operating conditions and identifying the presence or absence of external heat sources, an external thermal runaway risk warning model is established. The internal risk warning model building module is used to build an internal risk warning model based on battery operating parameters and alarm data. The energy storage power station risk warning model establishment module is used to determine the weights of the external thermal runaway risk warning model and the internal risk warning model, and to obtain the final energy storage power station risk warning model by combining the models through weighted summation. The risk warning result output module is used to acquire multi-source data of the battery pack of the target energy storage power station, input it into the risk warning model of the energy storage power station, and output the risk warning result.

7. A computer device, characterized in that, include: At least one processor, and a memory communicatively connected to the at least one processor, wherein the memory stores instructions executable by the at least one processor, the instructions being executed by the at least one processor to cause the at least one processor to perform the multi-source hybrid signal energy storage power station risk warning method according to any one of claims 1-5.

8. A computer-readable storage medium, characterized in that, The computer-readable storage medium stores computer instructions for causing the computer to execute the multi-source mixed signal energy storage power station risk early warning method according to any one of claims 1-5.

Images