Temperature monitoring method and device and electronic equipment

Abstract

The invention discloses a temperature monitoring method and device and electronic equipment. The method comprises the steps that temperature data and a state video of a storage battery pack are obtained, and the state video comprises video data of the storage battery pack in the charging process and the discharging process; the temperature data and the state video are analyzed, a data analysis result is obtained, and the data analysis result at least comprises real-time temperature data and predicted temperature data of the storage battery pack; and determining the abnormal degree of the storage battery pack according to a data analysis result, and executing early warning operation under the condition that the abnormal degree exceeds a preset threshold value. According to the method and the device, the technical problems that the abnormal change of the temperature cannot be fully identified due to the fact that the temperature monitoring method in the related technology mostly depends on periodic detection or fixed temperature threshold alarm, and especially when the storage battery pack is in different charging and discharging states, false alarm is generated due to different temperature fluctuation ranges are solved.

Description

Technical Field

[0001] This application relates to the field of power monitoring, and more particularly, to a temperature monitoring method, apparatus, and electronic device. Background Art

[0002] In modern communication networks, power rooms, as key facilities of communication operators, carry the important task of providing stable power supply for key communication equipment. Among them, the battery pack, as an energy storage device in the power room, its operating state is directly related to the stability of the entire room and the continuity of communication services. During the charging and discharging process of the battery, temperature is an extremely sensitive indicator. Abnormal temperature, especially in the high-temperature state, will not only accelerate the aging of the battery, reduce its service life, but also may cause thermal runaway, resulting in equipment damage or even fire, posing a serious threat to communication security.

[0003] In related technologies, the battery temperature monitoring method usually relies on regular manual inspections and an automatic alarm mechanism with static thresholds. This monitoring mode has significant limitations. On the one hand, due to the fixed and limited detection frequency, the system may miss the warning of sudden temperature changes and cannot respond to temperature anomalies in real time, especially lacking the immediate warning ability in emergency situations. On the other hand, the setting of fixed thresholds ignores the influence of important dynamic factors such as charge and discharge states and ambient temperature fluctuations, resulting in misjudgments in the early warning mechanism in practical applications, including false alarms and missed alarms, reducing the accuracy and reliability of temperature monitoring. In addition, lacking in-depth data analysis and prediction functions, traditional monitoring systems cannot effectively utilize historical temperature information to predict future temperature anomaly trends, limiting their application potential in preventive and intelligent monitoring.

[0004] To address the above problems, no effective solutions have been proposed yet. Summary of the Invention

[0005] Embodiments of this application provide a temperature monitoring method, apparatus, and electronic device to at least solve the technical problem that the temperature monitoring method in related technologies mostly relies on periodic detection or fixed temperature threshold alarms, and cannot fully identify abnormal temperature changes, especially when the battery pack is in different charge and discharge states, false alarms are generated due to different temperature fluctuation ranges.

[0006] According to one aspect of the embodiments of the present application, a temperature monitoring method is provided, including: obtaining temperature data and status videos of a battery pack, where the temperature data includes temperature values of the battery pack at different times, and the status videos include video data of the battery pack during the charging process and the discharging process; analyzing the temperature data and the status videos to obtain an analysis result of the data, where the analysis result of the data at least includes real-time temperature data and predicted temperature data of the battery pack, the real-time temperature data is used to represent the real-time temperature change of the battery pack, and the predicted temperature data is used to represent the temperature change trend of the battery pack within a future time period; determining the abnormal degree of the battery pack according to the analysis result of the data, and performing a warning operation when the abnormal degree exceeds a preset threshold.

[0007] Optionally, obtaining the temperature data of the battery pack includes: collecting the temperature data of the battery pack in the operating state through a sensor in an infrared camera at a preset time interval.

[0008] Optionally, analyzing the temperature data and the status videos to obtain an analysis result of the data includes: analyzing the temperature data and the status videos through a convolutional neural network to obtain real-time temperature data; analyzing the temperature data and the status videos through a long short-term memory network to obtain predicted temperature data.

[0009] Optionally, analyzing the temperature data and the status videos through a convolutional neural network to obtain real-time temperature data includes: determining a first feature map corresponding to the temperature data and a second feature map corresponding to the status videos, where the first feature map is used to represent the thermal imaging features corresponding to the temperature data, and the second feature map is used to represent the status change features corresponding to each frame of the status videos; extracting features from the first feature map through a convolutional neural network to obtain the temperature distribution information of the battery pack, where the temperature distribution information includes the temperature hot spot area and the temperature gradient of the battery pack; extracting features from the second feature map through a convolutional neural network to obtain the status change information of the battery pack, where the status change information includes the color change, the surface state change of the battery pack, and the visual feature change of the battery pack in the charging state and the discharging state; determining the real-time temperature data according to the temperature distribution information and the status change information.

[0010] Optionally, analyze the temperature data and status video through a long short-term memory network to obtain predicted temperature data, including: converting the temperature data into time series data, and determining a feature vector corresponding to each frame image in the status video, where the feature vector is used to represent the temperature change characteristics of the battery pack and the state change characteristics of the battery pack in the charging state and the discharging state; fusing the time series data and the feature vector to obtain the real-time temperature series of the battery pack; obtaining the historical temperature data of the battery pack, and processing the historical temperature data and the real-time temperature series through a long short-term memory network to obtain predicted temperature data.

[0011] Optionally, determine the abnormality degree of the battery pack according to the data analysis result, and perform a warning operation when the abnormality degree exceeds a preset threshold, including: comparing the real-time temperature data and the historical temperature data to obtain a first comparison result, where the first comparison result is used to reflect the abnormality degree of the real-time temperature data; when the first comparison result indicates that the deviation value between the real-time temperature data and the historical temperature data exceeds the first preset threshold in the preset threshold, determine that the real-time temperature data is abnormal, and perform a first-level warning operation in the warning operation, where the first-level warning operation is used to generate a warning message and send it to the target object.

[0012] Optionally, determine the abnormality degree of the battery pack according to the data analysis result, and perform a warning operation when the abnormality degree exceeds a preset threshold, and further include: comparing the predicted temperature data and the real-time temperature data to obtain a second comparison result, where the second comparison result reflects the abnormality degree of the predicted temperature data; when the second comparison result indicates that the deviation value between the predicted temperature data and the real-time temperature data exceeds the second preset threshold in the preset threshold, determine that the predicted temperature data is abnormal, and perform a second-level warning operation in the warning operation, where the second-level warning operation is used to trigger an automatic temperature reduction mechanism.

[0013] According to another aspect of the embodiments of the present application, there is also provided a temperature monitoring device, including: an acquisition module, configured to acquire the temperature data and status video of the battery pack, where the temperature data includes the temperature values of the battery pack at different times, and the status video includes the video data of the battery pack during the charging process and the discharging process; an analysis module, configured to analyze the temperature data and the status video to obtain a data analysis result, where the data analysis result at least includes the real-time temperature data and the predicted temperature data of the battery pack, the real-time temperature data is used to represent the real-time temperature change of the battery pack, and the predicted temperature data is used to represent the temperature change trend of the battery pack in the future time period; a warning module, configured to determine the abnormality degree of the battery pack according to the data analysis result, and perform a warning operation when the abnormality degree exceeds a preset threshold.

[0014] According to another aspect of the embodiments of the present application, an electronic device is further provided, including: a memory and a processor, wherein the memory is used to store program instructions; the processor is connected to the memory and is used to execute the above temperature monitoring method.

[0015] According to yet another aspect of the embodiments of the present application, a non-volatile storage medium is further provided. The non-volatile storage medium includes a stored computer program, wherein the device where the non-volatile storage medium is located executes the above temperature monitoring method by running the computer program.

[0016] According to yet another aspect of the embodiments of the present application, a computer program product is further provided, including computer instructions that implement the above temperature monitoring method when executed by a processor.

[0017] In the embodiments of the present application, by obtaining the temperature data and status videos of the battery pack, wherein the temperature data includes the temperature values of the battery pack at different times, and the status videos include the video data of the battery pack during the charging process and the discharging process; analyzing the temperature data and the status videos to obtain an analysis result of the data, wherein the analysis result of the data at least includes the real-time temperature data and the predicted temperature data of the battery pack, the real-time temperature data is used to represent the real-time temperature change of the battery pack, and the predicted temperature data is used to represent the temperature change trend of the battery pack within a future time period; determining the abnormal degree of the battery pack according to the analysis result of the data, and performing a warning operation when the abnormal degree exceeds a preset threshold, the purpose of accurately monitoring the temperature change of the battery pack, predicting and warning potential high-temperature risks in advance is achieved, thereby realizing the technical effects of improving the safety level of the power machine room, ensuring the stable operation of power equipment, and reducing equipment failures and maintenance costs caused by abnormal temperatures, and further solving the technical problem that the temperature monitoring methods in the related art mostly rely on periodic detection or fixed temperature threshold alarms and cannot fully identify abnormal temperature changes, especially when the battery pack is in different charge and discharge states, false alarms are generated due to different temperature fluctuation ranges. Description of the Drawings

[0018] The drawings described herein are used to provide a further understanding of the present application and constitute a part of the present application. The illustrative embodiments of the present application and their descriptions are used to explain the present application and do not constitute an improper limitation to the present application. In the drawings:

[0019] Figure 1 is a hardware structure diagram of a computer terminal for implementing the temperature monitoring method according to the embodiments of the present application;

[0020] Figure 2 is a flowchart of a temperature monitoring method according to the embodiments of the present application;

[0021] Figure 3It is a schematic diagram of a system architecture according to an embodiment of the present application;

[0022] Figure 4 It is a structural diagram of a temperature monitoring device according to an embodiment of the present application. Detailed implementation manners

[0023] In order to enable those skilled in the art of the present technology to better understand the solutions of the present application, the technical solutions in the embodiments of the present application will be clearly and completely described below in conjunction with the accompanying drawings in the embodiments of the present application. Obviously, the described embodiments are only a part of the embodiments of the present application, rather than all of the embodiments. Based on the embodiments in the present application, all other embodiments obtained by those of ordinary skill in the art without making creative efforts shall fall within the protection scope of the present application.

[0024] It should be noted that the terms "first", "second", etc. in the specification and claims of the present application and the above-mentioned drawings are used to distinguish similar objects, and do not necessarily need to be used to describe a specific order or sequence. It should be understood that such data can be interchanged under appropriate circumstances so that the embodiments of the present application described herein can be implemented in an order other than those illustrated or described herein. In addition, the terms "comprising" and "having" and any variations thereof are intended to cover non-exclusive inclusion. For example, a process, method, system, product or device that includes a series of steps or units does not necessarily need to be limited to those steps or units clearly listed, but may include other steps or units not clearly listed or inherent to these processes, methods, products or devices.

[0025] First, some nouns or terms that appear in the process of explaining the embodiments of the present application are applicable to the following explanations:

[0026] Convolutional Neural Network (CNN): A neural network specifically used for image and video processing. It extracts features from input data through convolutional layers and reduces the feature dimension through pooling layers to identify patterns and structures in images. In the present application, CNN is used to analyze the surface temperature image of the battery captured by the infrared camera to identify abnormal temperature areas.

[0027] Long Short-Term Memory (LSTM): A recurrent neural network used for processing time series data, which is good at processing time series data with long-term dependencies. It solves the long-term dependence problem by introducing gating mechanisms (such as forget gates, input gates, output gates), allowing the network to selectively remember or forget information. In the present application, LSTM is used to analyze the temperature trend of the battery pack and predict future temperature changes to early warn of possible overheating risks.

[0028] Time series analysis: A technique for statistical or predictive analysis of data that changes over time, typically used to identify trends, periodic changes, and outliers in the data.

[0029] AI video analysis: Refers to the use of artificial intelligence technologies, such as computer vision and deep learning, to automatically analyze video content, identify objects, detect behaviors, analyze trends, etc.

[0030] To solve the problem of poor temperature monitoring efficiency in related technologies, an embodiment of the present application provides a temperature monitoring method, which can run on Figure 1 the computer terminal shown below. The following describes this computer terminal.

[0031] The temperature monitoring method embodiment provided by the embodiment of the present application can be executed on a mobile terminal, a computer terminal, or a similar computing device. Figure 1 The following shows a hardware structure block diagram of a computer terminal for implementing the temperature monitoring method. As Figure 1 shown, the computer terminal 10 may include one or more (shown as 102a, 102b,..., 102n in the figure) processors (the processor may include, but is not limited to, a processing device such as a microprocessor MCU or a programmable logic device FPGA), a memory 104 for storing data, and a transmission module 106 for communication functions connected by wired and / or wireless networks. In addition, it may further include: a display, a keyboard, a cursor control device, an input / output interface (I / O interface), a universal serial bus (USB) port (which may be included as one of the ports of the I / O interface), a network interface, and a BUS bus. Those of ordinary skill in the art can understand that Figure 1 the structure shown is only schematic and does not limit the structure of the above-mentioned electronic device. For example, the computer terminal 10 may further include more or fewer components than Figure 1 shown, or have a different configuration from Figure 1 shown.

[0032] It should be noted that the above one or more processors and / or other data processing circuits are generally referred to as "data processing circuits" in this article. The data processing circuit may be embodied in whole or in part as software, hardware, firmware, or any combination thereof. In addition, the data processing circuit may be a single independent processing module, or be incorporated in whole or in part into any one of the other elements in the computer terminal 10. As involved in the embodiment of the present application, the data processing circuit is used for processor control (such as the selection of a variable resistance terminal path connected to an interface).

[0033] The memory 104 can be used to store software programs and modules of application software, such as the program instructions / data storage device corresponding to the temperature monitoring method in the embodiments of the present application. The processor executes various functional applications and data processing by running the software programs and modules stored in the memory 104, that is, the above-mentioned temperature monitoring method is implemented. The memory 104 may include a high-speed random access memory, and may also include a non-volatile memory, such as one or more magnetic storage devices, flash memories, or other non-volatile solid-state memories. In some instances, the memory 104 may further include a memory remotely disposed relative to the processor, and these remote memories can be connected to the computer terminal 10 through a network. Examples of the above-mentioned network include but are not limited to the Internet, intranets, local area networks, mobile communication networks, and combinations thereof.

[0034] The transmission module 106 is used to receive or send data via a network. Specific examples of the above-mentioned network may include a wireless network provided by a communication provider of the computer terminal 10. In one instance, the transmission module 106 includes a network adapter (Network Interface Controller, NIC), which can be connected to other network devices through a base station and thus can communicate with the Internet. In one instance, the transmission module 106 can be a radio frequency (RF) module, which is used to communicate with the Internet wirelessly.

[0035] The display can be, for example, a touch-screen liquid crystal display (LCD), which enables a user to interact with the user interface of the computer terminal 10.

[0036] It should be noted here that in some alternative embodiments, the above Figure 1 shown computer terminal may include hardware elements (including circuits), software elements (including computer code stored on a computer-readable medium), or a combination of both hardware elements and software elements. It should be pointed out that Figure 1 is only an example of a specific specific instance and is intended to illustrate the types of components that may exist in the above computer terminal.

[0037] Under the above operating environment, an embodiment of a temperature monitoring method is provided in the embodiments of the present application. It should be noted that the steps shown in the flowchart of the accompanying drawings can be executed in a computer system such as a set of computer-executable instructions, and although the logical order is shown in the flowchart, in some cases, the steps shown or described can be executed in a different order than here.

[0038] Figure 2 is a flowchart of a temperature monitoring method according to an embodiment of the present application. As Figure 2 shown, the method includes the following steps:

[0039] Step S202: Obtain the temperature data and status video of the battery pack. The temperature data includes the temperature values of the battery pack at different times, and the status video includes the video data of the battery pack during the charging and discharging processes.

[0040] Specifically, obtaining the temperature data of the battery pack includes: collecting the temperature data of the battery pack in the operating state through the sensor in the infrared camera at a preset time interval.

[0041] In the above step S202, the temperature data is the temperature values of the battery pack at different times collected in real time by the infrared camera. These data provide a direct and real-time feedback of the battery temperature and are the basic information source of the monitoring system. Among them, the infrared camera is installed directly above or on the side of the battery pack in the computer room to fully cover the surface of the battery pack, so as to ensure that the temperature status of the entire battery pack can be monitored without dead angles. The camera is equipped with a high-precision sensor that can frequently collect temperature data at intervals of several seconds and instantly generate thermal imaging maps. These images and data are then quickly transmitted to the central data processing system for analysis and presented in real time on the monitoring platform. Due to the high-sensitivity characteristics of its sensor, the infrared camera can keenly capture minute temperature fluctuations, which makes the temperature collection not only frequent but also extremely accurate, providing a solid foundation for subsequent temperature anomaly detection and early warning.

[0042] It should be noted that when transmitting the temperature image data to the central data processing unit, compression and encryption measures can be taken to ensure the efficiency and security of data transmission.

[0043] The status video is a series of video data of the battery pack during the charging and discharging processes recorded by the AI camera. Among them, the AI camera can not only shoot videos at a high frame rate to ensure that every instant of the state change is recorded, but also identify and understand the surface characteristics, color changes and any potential abnormal behaviors of the battery during the charging and discharging states through real-time analysis of the video stream.

[0044] Step S204: Analyze the temperature data and status video to obtain the data analysis result. The data analysis result at least includes the real-time temperature data and predicted temperature data of the battery pack. The real-time temperature data is used to represent the real-time temperature change of the battery pack, and the predicted temperature data is used to represent the temperature change trend of the battery pack in the future time period.

[0045] In the above step S204, the temperature data and status video can be processed by a CNN model to provide real-time temperature change information of the battery pack under the operating state, that is, the above real-time temperature data. This real-time temperature data is used to directly monitor the current temperature state and is the basis for judging immediate anomalies. Further, based on big data analysis, especially the LSTM model, the temperature data and status video can be processed to predict the temperature change trend of the battery pack in the future time period, that is, the above predicted temperature data. This predicted temperature data is used to identify abnormal increase or decrease trends of the battery pack and provide a basis for early warning.

[0046] Step S206: Determine the abnormal degree of the battery pack according to the data analysis result, and perform a warning operation when the abnormal degree exceeds a preset threshold.

[0047] In the above step S206, the judgment of the abnormal degree is based on the real-time temperature data and the predicted temperature data, and combines the historical temperature information and the current charge and discharge state of the battery pack. If the analysis result shows that the abnormal degree of the temperature exceeds the preset threshold, that is, the temperature abnormally rises beyond the normal range or the predicted temperature will approach or even exceed the safety threshold, the system will immediately perform a warning operation.

[0048] Through the above steps S202 to S206, the purpose of accurately monitoring the temperature change of the battery pack, predicting and warning potential high-temperature risks in advance is achieved, thereby realizing the technical effects of improving the safety level of the power machine room, ensuring the stable operation of power equipment, and reducing equipment failures and maintenance costs caused by abnormal temperatures. Furthermore, it solves the technical problem that the temperature monitoring methods in related technologies mostly rely on periodic detection or fixed temperature threshold alarms and cannot fully identify abnormal temperature changes, especially when the battery pack is in different charge and discharge states, false alarms occur due to different temperature fluctuation ranges. The following is a detailed description.

[0049] In the above step S204, the temperature data and status video are analyzed to obtain a data analysis result, including: analyzing the temperature data and status video through a convolutional neural network to obtain real-time temperature data; analyzing the temperature data and status video through a long short-term memory network to obtain predicted temperature data.

[0050] In the embodiment of the present application, the system comprehensively uses two deep learning technologies, the CNN network and the LSTM network, to deeply analyze the collected temperature data and status video. This process is the key to realizing intelligent temperature monitoring and warning of the battery pack.

[0051] Among them, the CNN is mainly responsible for processing real-time video streams. By identifying and analyzing the temperature distribution characteristics on the surface of the battery, such as color changes and heat map patterns, accurate real-time temperature data can be obtained. This kind of analysis can promptly capture the occurrence of any temperature anomaly. Further, the LSTM focuses on the time series characteristics of the temperature data. By learning the fluctuation rules of historical temperature data and the temperature change patterns under charge and discharge states, it predicts the temperature trend of the battery in the short term in the future, that is, the above-mentioned predicted temperature data. This combination of the real-time analysis of CNN and the future prediction of LSTM not only ensures the system's immediate response to the current temperature anomaly but also can prospectively identify potential temperature rise risks and achieve effective early warning.

[0052] Optionally, the temperature data and the status video are analyzed through a convolutional neural network to obtain real-time temperature data, including: determining a first feature map corresponding to the temperature data, and determining a second feature map corresponding to the status video, where the first feature map is used to represent the thermal imaging features corresponding to the temperature data, and the second feature map is used to represent the status change features corresponding to each frame of the image in the status video; extracting features from the first feature map through a convolutional neural network to obtain the temperature distribution information of the battery pack, where the temperature distribution information includes the temperature hot spot area and the temperature gradient of the battery pack; extracting features from the second feature map through a convolutional neural network to obtain the status change information of the battery pack, where the status change information includes the color change, the surface state change of the battery pack, and the visual feature change of the battery pack under the charge state and the discharge state; determining the real-time temperature data based on the temperature distribution information and the status change information.

[0053] In the embodiment of the present application, through the analysis of the temperature data and the status video by the CNN network, the system can intelligently identify and extract the temperature distribution characteristics and the status change characteristics on the surface of the battery pack, and then determine the real-time temperature data. The specific process can be as follows:

[0054] First, the temperature data obtained by the infrared camera is converted into a thermal imaging image, and then a first feature map is formed. This first feature map intuitively reflects the temperature distribution on the surface of the battery, and different colors correspond to different temperature values, providing rich visual information for subsequent feature extraction and serving as the basic data for the CNN to perform temperature analysis. At the same time, the battery status video captured by the AI camera is converted into a series of image frames, and each frame of the image corresponds to a second feature map, which is used to capture the surface change characteristics of the battery. Through the analysis of color and surface state, the second feature map can reveal the visual feature changes of the battery during the charge and discharge processes, including the light and dark changes of the color and the subtle differences in the surface texture.

[0055] Specifically, different layers in the CNN network can be used to extract different features of each frame of the image. For example, starting from the original image data, multi-level features such as edges, textures, and object shapes can be gradually extracted through convolutional layers, and then dimensionality reduction processing is performed through pooling layers to enhance the robustness and computational efficiency of the model. Finally, the generated multi-level feature maps (such as the above-mentioned first feature map and second feature map) comprehensively integrate the details and comprehensive information of the image, providing a solid foundation for subsequent tasks such as image classification, object detection, and state recognition, demonstrating the powerful functions and application potential of deep learning in the field of image analysis. Among them, in some complex video analysis tasks, the system will only extract key frames with significant changes to reduce the processing volume and improve efficiency. For example, key frames are saved at important change moments in the video (such as the start, end, or acceleration instant of an action), reducing the processing of repeated frames.

[0056] It should be noted that before forming the first feature map and the second feature map, image denoising and size adjustment need to be performed on each frame of the temperature image and the status video. Specifically, for example, basic denoising processing (such as filtering and normalization) is performed on each image data to standardize factors such as image brightness and color, reduce unnecessary noise interference, and ensure consistent processing conditions for each image. Also, each image data is uniformly scaled to a specified size to meet the input requirements of the CNN model. For example, a common size is 224x224 pixels or other standard resolutions.

[0057] Secondly, the CNN analyzes the first feature map to extract the temperature distribution information of the battery pack, including but not limited to the temperature hot spot area (i.e., the part where the temperature abnormally rises) and the temperature gradient (the difference in temperature change) of the battery pack. At the same time, the CNN processes the second feature map to extract the state change information of the battery pack, including but not limited to the color change and surface state change of the battery pack. These information reflect the health status of the battery during the charging and discharging process, helping to determine whether the temperature change matches the charging and discharging state and avoiding false alarms caused by normal charging and discharging.

[0058] Finally, based on feature extraction, the temperature distribution information and the state change information are comprehensively analyzed to determine the real-time temperature data. Specifically, by comparing the current temperature hot spot with historical data, it is possible to judge in real time whether there is overheating or local abnormality in the battery pack. At the same time, combined with the state change information, the system can more accurately identify the normal temperature fluctuation range during the charging and discharging state, as well as any abnormal changes that may exceed this range, thereby obtaining more accurate real-time temperature data to provide a basis for subsequent early warning and maintenance decisions.

[0059] Optionally, the temperature data and status video are analyzed through a long short-term memory network to obtain predicted temperature data, including: converting the temperature data into time series data, and determining a feature vector corresponding to each frame of the status video, where the feature vector is used to represent the temperature change characteristics of the battery pack and the state change characteristics of the battery pack in the charging state and the discharging state; fusing the time series data and the feature vector to obtain the real-time temperature sequence of the battery pack; obtaining the historical temperature data of the battery pack, and processing the historical temperature data and the real-time temperature sequence through the long short-term memory network to obtain the predicted temperature data.

[0060] In the embodiments of the present application, through the prediction ability of LSTM, the system can comprehensively utilize real-time and historical data to detect and warn of temperature anomalies of the battery pack in advance, providing more accurate temperature monitoring and maintenance decision support. The specific process can be as follows:

[0061] First, the temperature data obtained by the infrared camera is converted into time series data. At the same time, the battery status video captured by the AI camera is analyzed frame by frame, and the temperature change characteristics and morphological characteristics in the charging and discharging states of each frame of the image are converted into feature vectors. These feature vectors contain comprehensive information on the surface color, texture, and temperature distribution of the battery, and are the basis for LSTM to perform complex time series analysis.

[0062] Secondly, the time series data is preprocessed, including data synchronization (ensuring the time consistency of the infrared temperature data and the status video), data standardization (for example, scaling the temperature data to a unified numerical range, such as between 0 and 1, to meet the model input requirements), and feature extraction (for example, selecting the features most relevant to temperature change from the feature vectors of each frame in the status video), to ensure the quality and format consistency of the data, providing an optimized data set for model training.

[0063] Furthermore, the system fuses the time series temperature data with the feature vectors of the status video to generate the real-time temperature sequence of the battery pack. This fusion process takes into account the interaction between temperature and visual features, and can more comprehensively reflect the temperature change characteristics of the battery pack in different charging and discharging states, providing multi-dimensional data support for subsequent in-depth analysis and prediction.

[0064] Next, obtain the historical temperature data and related status characteristics of the battery pack as the basis for training the LSTM model. Through training, the LSTM model can learn and understand the long-term dependence relationship of temperature data and the correlation of status characteristics, and shows advantages especially when dealing with complex temperature fluctuations in the charging state and discharging state. The trained LSTM model takes the real-time temperature sequence and the current feature vector as inputs to predict the temperature change in the next period of time. This prediction ability enables the system to not only monitor the temperature in real time but also give early warnings of temperature anomalies, providing sufficient time for maintenance personnel to take measures to avoid failures.

[0065] Finally, based on the prediction processing of the LSTM model, the system generates predicted temperature data, which contains the future temperature trend and potential anomalies and is the key basis for the warning module to make intelligent alarm decisions. The prediction results are combined with the real-time temperature data to form a complete temperature monitoring and warning mechanism, ensuring the safety and stability of the operation of the battery pack in the computer room.

[0066] In the above step S206, determine the abnormal degree of the battery pack according to the data analysis result, and perform a warning operation when the abnormal degree exceeds a preset threshold, including: comparing the real-time temperature data and the historical temperature data to obtain a first comparison result, where the first comparison result is used to reflect the abnormal degree of the real-time temperature data; when the first comparison result indicates that the deviation value between the real-time temperature data and the historical temperature data exceeds the first preset threshold in the preset threshold, determine that the real-time temperature data is abnormal and perform a first-level warning operation in the warning operation, where the first-level warning operation is used to generate a warning message and send it to the target object, such as the computer room operation and maintenance personnel or the remote monitoring platform.

[0067] In the embodiment of the present application, by setting a reasonable first preset threshold, the system can accurately judge the abnormal situation of the current temperature of the battery pack, avoid false alarms or missed alarms, and ensure the effectiveness and timeliness of the warning information, which is a key step in realizing real-time temperature monitoring and warning of the battery pack based on infrared imaging and AI. Among them, the generation and timely sending of the warning information provide a valuable time window for dealing with potential battery overheating risks, ensuring the equipment safety and stable operation of the power computer room.

[0068] Specifically, data distribution analysis, difference analysis, and time window comparison analysis methods can be used to compare the real-time temperature data and the historical temperature data.

[0069] I. Data distribution analysis: That is, a statistical method used to compare the distributions of the real-time temperature data and the historical temperature data to determine whether the real-time temperature data deviates from the historical distribution.

[0070] The specific expression is as follows:

[0071]

[0072] In the formula, T r represents real-time temperature data; z represents the score of the standard normal distribution (z-score), which is used to measure the deviation degree of the real-time temperature data from the mean value of the historical temperature data; μ h represents the mean value of the historical temperature data; σ h represents the standard deviation of the historical temperature data.

[0073] When |z| > k (generally k = 2 or k = 3 is taken), the real-time temperature data is determined to be abnormal.

[0074] II. Difference analysis: It is used to calculate the difference ΔT between the real-time temperature data and the mean value of the historical temperature data to measure the deviation degree of the real-time temperature data.

[0075] The specific expression is as follows:

[0076] ΔT = T r - μ h

[0077] In the formula, it represents the difference between the real-time temperature data and the mean value of the historical temperature data. When ΔT exceeds the preset threshold T threshold (set based on the historical fluctuation range), the real-time temperature data is determined to be abnormal.

[0078] III. Time window comparison: Since the temperature values in the historical temperature data usually have periodic change characteristics (such as day and night, seasonal fluctuations, etc.), the difference ΔT between the real-time temperature data and the historical temperature data within the same time window can be compared through a sliding window window .

[0079] The specific expression is as follows:

[0080]

[0081] In the formula, n is the window size (such as the number of data points in the past 1 hour). When |ΔT window | exceeds the preset threshold, the real-time temperature data is determined to be abnormal.

[0082] Optionally, determine the degree of abnormality of the battery pack according to the data analysis result, and perform a warning operation when the degree of abnormality exceeds a preset threshold. It further includes: comparing the predicted temperature data and the real-time temperature data to obtain a second comparison result, where the second comparison result reflects the degree of abnormality of the predicted temperature data; when the second comparison result indicates that the deviation value between the predicted temperature data and the real-time temperature data exceeds a second preset threshold among the preset thresholds, determine that the predicted temperature data is abnormal and perform a secondary warning operation in the warning operation, where the secondary warning operation is used to trigger an automatic cooling mechanism.

[0083] In the embodiment of the present application, by setting the second preset threshold, the system can effectively distinguish normal prediction fluctuations from potential fault signs, so as to respond in time before the battery temperature is abnormal, reduce equipment damage and operation risks, and is an important part of the system's intelligent temperature monitoring and warning. Among them, the secondary warning can not only generate warning information, but also activate an automatic cooling mechanism, such as starting a ventilation system or standby cooling equipment, to directly intervene to control the temperature rise and prevent overheating risks, ensuring the stable operation of key equipment in the power machine room.

[0084] Specifically, the real-time temperature data and the predicted temperature data can be compared by using residual analysis and trend scoring method. Among them, residual analysis focuses on the difference between the actual temperature and the predicted temperature, while trend scoring focuses on the rate of temperature change.

[0085] I. Residual analysis: By comparing the difference between the real-time data and the predicted data, it is judged whether the actual temperature significantly deviates from the predicted temperature, so as to identify potential abnormal situations.

[0086] The specific expression is as follows:

[0087] Residual=T r -T p

[0088] In the formula, T r represents the real-time temperature data, T p represents the predicted temperature data, and Residual represents the difference between the predicted temperature data and the real-time temperature data. When the residual exceeds the historical standard deviation range, the predicted temperature data is determined to be abnormal.

[0089] II. Trend scoring: By defining a scoring function to evaluate the rate of temperature rise or fall, it is judged whether the trend of temperature change is abnormal.

[0090] The specific expression is as follows:

[0091]

[0092] In the formula, T tRepresents the temperature value of the battery pack at time t, T t+1 It represents the temperature value of the battery pack at time t+1, Δt represents the time interval; S represents the trend score, which is used to represent the rate of change of temperature over time, that is, the speed at which the temperature rises or falls.

[0093] If the trend score S is greater than the preset threshold, it may indicate abnormal temperature rise. This means that the temperature rise rate exceeds the normal range, which may indicate some abnormal situation, such as equipment failure or environmental changes.

[0094] In the embodiment of the present application, an efficient abnormal temperature warning mechanism is demonstrated by combining the statistical analysis of historical temperature data with real-time temperature monitoring. For example, when the system monitors the current temperature of 30°C in real time based on the temperature mean of 28°C and standard deviation of 2°C of the battery pack in the past hour, it is found through residual analysis that although this temperature value has not yet reached the abnormal standard, combined with the analysis of the prediction model, it is expected that the temperature will rise to 32°C in the next 5 minutes, close to the set safety threshold. The system will warn when the real-time data deviates from the historical mean by 3σ h The first-level warning operation is triggered when the temperature exceeds the safety threshold of 35°C. The setting of this warning threshold is based on the principle of 3 times the standard deviation, which ensures the accuracy of the warning. At the same time, the system predicts the future temperature trend. When the predicted temperature exceeds the safety threshold of 35°C, the second-level warning operation is triggered, that is, emergency alarm measures are taken immediately to notify the operation and maintenance personnel to take corresponding actions.

[0095] In an embodiment of the present application, the prediction window and alarm threshold can also be dynamically adjusted according to the real-time charge and discharge status of the battery pack to ensure the accuracy and timeliness of the early warning. For example, through intelligent analysis of the current temperature trend, if the prediction algorithm shows that the battery temperature will approach or exceed the preset safety upper limit within the next 5 minutes, the system will immediately trigger an alarm and notify relevant personnel to take emergency measures, such as starting the cooling system or adjusting the load to prevent the risk of overheating. On the contrary, if the prediction results show that the temperature trend is stable or in a declining stage, the system will not trigger an alarm to avoid unnecessary interference and waste of resources. This mechanism not only improves the sensitivity of temperature anomaly detection through the combination of real-time monitoring and intelligent prediction, but also optimizes the alarm strategy to ensure the safety and stability of the operation of the battery pack in the power room.

[0096] Figure 3 Schematic diagram of a system architecture according to an embodiment of the present application. Figure 3As shown in the figure, the system includes five modules: an infrared camera, an NVR recorder, AI graphic analysis, big data analysis, and an early warning system. Among them, the infrared camera module, as the "eyes" of the system, is responsible for continuously monitoring the surface temperature data of the battery pack. Its highly sensitive sensor can capture subtle temperature differences to ensure the acquisition of real-time temperature data. The NVR recorder module undertakes the task of recording and storing the temperature video data collected by the infrared camera, providing the original materials for subsequent analysis. The AI graphic analysis module is one of the brains of the system. Using deep learning technology, especially convolutional neural networks (CNNs), it processes the video data to identify the surface characteristics and abnormal temperature distributions of the battery under different charge and discharge states, enhancing the intelligent analysis ability of the monitoring. The big data analysis module integrates real-time and historical data and uses time series analysis algorithms such as long short-term memory networks (LSTMs) for temperature trend prediction and pattern recognition, dynamically adjusting the early warning threshold to improve the accuracy and timeliness of early warnings. The early warning system module, as the last link of the system, intelligently judges the change trend of the battery temperature based on the results of big data analysis. When detecting potential temperature anomalies, it can immediately trigger a multi-level early warning mechanism, from mild warnings to emergency alarms, ensuring a quick response, thus effectively preventing and dealing with overheating risks and ensuring the operation safety of the power machine room.

[0097] Overall, in the embodiment of the present application, the real-time temperature monitoring of infrared imaging technology is combined with the intelligent analysis of AI deep learning algorithms. Through the collaborative work of convolutional neural networks (CNNs) and long short-term memory networks (LSTMs), accurate identification and predictive early warning of battery pack temperature anomalies are achieved. The system can not only capture subtle changes in the surface temperature of the battery in real time but also perform trend analysis based on historical data, dynamically adjusting the temperature threshold to effectively avoid false alarms. At the same time, through a multi-level early warning mechanism, timely notifications and response strategies are provided when the temperature is abnormal, including starting an automatic cooling mechanism, significantly improving the safety monitoring level of the battery pack in the power machine room and the intelligent degree of equipment maintenance. This technology combines image recognition, big data analysis, and machine learning prediction, providing an efficient, accurate, and forward-looking solution for the field of battery temperature monitoring.

[0098] According to the embodiment of the present application, a temperature monitoring device is provided. It should be noted that the temperature monitoring device in the embodiment of the present application can be used to execute the temperature monitoring method provided in the embodiment of the present application. The following introduces the temperature monitoring device provided in the embodiment of the present application.

[0099] Figure 4 is a structural diagram of a temperature monitoring device provided according to the embodiment of the present application. As Figure 4 shown, the device includes:

[0100] An acquisition module 40 is configured to acquire the temperature data and status video of the battery pack. The temperature data includes the temperature values of the battery pack at different times, and the status video includes the video data of the battery pack during the charging process and the discharging process;

[0101] An analysis module 42 is configured to analyze the temperature data and the status video to obtain an analysis result of the data. The analysis result of the data at least includes the real-time temperature data and predicted temperature data of the battery pack. The real-time temperature data is used to represent the real-time temperature change of the battery pack, and the predicted temperature data is used to represent the temperature change trend of the battery pack within a future time period;

[0102] An early warning module 44 is configured to determine the abnormal degree of the battery pack according to the analysis result of the data, and perform an early warning operation when the abnormal degree exceeds a preset threshold.

[0103] Through the acquisition module 40, the analysis module 42 and the early warning module 44 in the above temperature monitoring device, the purpose of accurately monitoring the temperature change of the battery pack, predicting and warning potential high-temperature risks in advance is achieved, thereby realizing the technical effects of improving the safety level of the power machine room, ensuring the stable operation of power equipment, and reducing equipment failures and maintenance costs caused by abnormal temperatures. Furthermore, the technical problem that the temperature monitoring methods in the related technologies mostly rely on periodic detection or fixed temperature threshold alarms and cannot fully identify abnormal temperature changes, especially when the battery pack is in different charge and discharge states and false alarms occur due to different temperature fluctuation ranges, is solved.

[0104] In the temperature monitoring device provided in the embodiment of the present application, the acquisition module is further configured to collect the temperature data of the battery pack in the operating state at preset time intervals through the sensor in the infrared camera.

[0105] In the temperature monitoring device provided in the embodiment of the present application, the analysis module is further configured to analyze the temperature data and the status video through a convolutional neural network to obtain real-time temperature data; analyze the temperature data and the status video through a long short-term memory network to obtain predicted temperature data.

[0106] In the temperature monitoring device provided in the embodiment of the present application, the analysis module is further configured to determine a first feature map corresponding to the temperature data and a second feature map corresponding to the status video, where the first feature map is used to represent the thermal imaging feature corresponding to the temperature data, and the second feature map is used to represent the status change feature corresponding to each frame of the status video; perform feature extraction on the first feature map through a convolutional neural network to obtain the temperature distribution information of the battery pack, where the temperature distribution information includes the temperature hot spot area and the temperature gradient of the battery pack; perform feature extraction on the second feature map through a convolutional neural network to obtain the status change information of the battery pack, where the status change information includes the color change, the surface status change of the battery pack, and the visual feature change of the battery pack in the charging state and the discharging state; determine the real-time temperature data according to the temperature distribution information and the status change information.

[0107] In the temperature monitoring device provided in the embodiment of the present application, the analysis module is further configured to convert the temperature data into time series data and determine a feature vector corresponding to each frame of the status video, where the feature vector is used to represent the temperature change feature of the battery pack and the status change feature of the battery pack in the charging state and the discharging state; fuse the time series data and the feature vector to obtain the real-time temperature sequence of the battery pack; obtain the historical temperature data of the battery pack, and process the historical temperature data and the real-time temperature sequence through a long short-term memory network to obtain the predicted temperature data.

[0108] In the temperature monitoring device provided in the embodiment of the present application, the warning module is further configured to compare the real-time temperature data with the historical temperature data to obtain a first comparison result, where the first comparison result is used to reflect the abnormal degree of the real-time temperature data; in the case where the first comparison result indicates that the deviation value between the real-time temperature data and the historical temperature data exceeds the first preset threshold in the preset threshold, determine that the real-time temperature data is abnormal and execute a first-level warning operation in the warning operation, where the first-level warning operation is used to generate a warning message and send it to the target object.

[0109] In the temperature monitoring device provided in the embodiment of the present application, the warning module is further configured to compare the predicted temperature data with the real-time temperature data to obtain a second comparison result, where the second comparison result reflects the abnormal degree of the predicted temperature data; in the case where the second comparison result indicates that the deviation value between the predicted temperature data and the real-time temperature data exceeds the second preset threshold in the preset threshold, determine that the predicted temperature data is abnormal and execute a second-level warning operation in the warning operation, where the second-level warning operation is used to trigger an automatic cooling mechanism.

[0110] The embodiment of the present application further provides an electronic device, including: a memory and a processor, where the memory is used to store program instructions; the processor is connected to the memory and is used to execute the above temperature monitoring method.

[0111] It should be noted that the above electronic device is used to execute Figure 2 the temperature monitoring method shown, so the relevant explanations in the above temperature monitoring method also apply to this electronic device, and will not be elaborated here.

[0112] The embodiment of the present application also provides a non-volatile storage medium, which includes a stored computer program. Wherein, the device where the non-volatile storage medium is located executes the above temperature monitoring method by running the computer program.

[0113] It should be noted that the above non-volatile storage medium is used to execute Figure 2 the temperature monitoring method shown, so the relevant explanations in the above temperature monitoring method also apply to this non-volatile storage medium, and will not be elaborated here.

[0114] The embodiment of the present application also provides a computer program product, including computer instructions, which implement the above temperature monitoring method when executed by a processor.

[0115] It should be noted that the above computer program product is used to execute Figure 2 the temperature monitoring method shown, so the relevant explanations in the above temperature monitoring method also apply to this computer program product, and will not be elaborated here.

[0116] The serial numbers of the above embodiments of the present application are only for description and do not represent the advantages and disadvantages of the embodiments.

[0117] In the above embodiments of the present application, the descriptions of each embodiment have their own emphases. For the parts not detailed in a certain embodiment, reference can be made to the relevant descriptions of other embodiments.

[0118] In several embodiments provided by the present application, it should be understood that the disclosed technical content can be implemented in other ways. Among them, the device embodiments described above are only illustrative. For example, the division of the units can be a logical function division. In actual implementation, there can be other division methods. For example, multiple units or components can be combined or integrated into another system, or some features can be ignored or not executed. Another point, the displayed or discussed coupling or direct coupling or communication connection to each other can be through some interfaces. The indirect coupling or communication connection of units or modules can be in an electrical or other form.

[0119] The unit described as a separation component may or may not be physically separated. The component shown as a unit may or may not be a physical unit, that is, it may be located in one place or distributed over multiple units. Some or all of the units can be selected according to actual needs to achieve the purpose of the solution of this embodiment.

[0120] In addition, each functional unit in various embodiments of the present application can be integrated in a processing unit, or each unit can exist physically alone, or two or more units can be integrated in one unit. The above-mentioned integrated unit can be implemented in the form of hardware or in the form of a software functional unit.

[0121] If the above-mentioned integrated unit is implemented in the form of a software functional unit and sold or used as an independent product, it can be stored in a computer-readable storage medium. Based on this understanding, the technical solution of the present application, in essence, or the part that contributes to the prior art, or all or part of this technical solution, can be embodied in the form of a software product. This computer software product is stored in a storage medium and includes several instructions to enable a computer device (which can be a personal computer, a server or a network device, etc.) to execute all or part of the steps of the methods described in various embodiments of the present application. The foregoing storage medium includes: USB flash drive, read-only memory (ROM, Read-Only Memory), random access memory (RAM, Random Access Memory), mobile hard disk, magnetic disk or optical disc and other various media that can store program codes.

[0122] The above are only the preferred embodiments of the present application. It should be noted that for those of ordinary skill in the art, without departing from the principle of the present application, several improvements and modifications can be made, and these improvements and modifications should also be regarded as the protection scope of the present application.

Claims

1. A temperature monitoring method, characterized in that: include: Acquire temperature data and status video of the battery pack, wherein the temperature data includes temperature values ​​of the battery pack at different times, and the status video includes video data of the battery pack during charging and discharging; Analyze the temperature data and the status video to obtain a data analysis result, wherein the data analysis result at least includes real-time temperature data and predicted temperature data of the battery pack, the real-time temperature data is used to represent the real-time temperature change of the battery pack, and the predicted temperature data is used to represent the temperature change trend of the battery pack in a future time period; The abnormality level of the battery pack is determined according to the data analysis result, and an early warning operation is performed when the abnormality level exceeds a preset threshold.

2. The method according to claim 1, characterized in that Get the temperature data of the battery pack, including: The temperature data of the battery pack in the operating state is collected at preset time intervals through the sensor in the infrared camera.

3. The method according to claim 1, characterized in that Analyze the temperature data and the status video to obtain data analysis results, including: Analyzing the temperature data and the state video by a convolutional neural network to obtain the real-time temperature data; The temperature data and the state video are analyzed by a long short-term memory network to obtain the predicted temperature data.

4. The method according to claim 3, characterized in that: The temperature data and the state video are analyzed by a convolutional neural network to obtain the real-time temperature data, including: Determine a first characteristic graph corresponding to the temperature data, and determine a second characteristic graph corresponding to the state video, wherein the first characteristic graph is used to represent the thermal imaging feature corresponding to the temperature data, and the second characteristic graph is used to represent the state change feature corresponding to each frame image in the state video; Extracting features from the first feature graph by using the convolutional neural network to obtain temperature distribution information of the battery pack, wherein the temperature distribution information includes a temperature hotspot area and a temperature gradient of the battery pack; Extracting features from the second feature graph by using the convolutional neural network to obtain state change information of the battery pack, wherein the state change information includes color change, surface state change, and visual feature change of the battery pack in a charging state and a discharging state; The real-time temperature data is determined according to the temperature distribution information and the state change information.

5. The method according to claim 3, characterized in that: Analyzing the temperature data and the state video through a long short-term memory network to obtain the predicted temperature data includes: Convert the temperature data into time series data, and determine a feature vector corresponding to each frame of the state video, wherein the feature vector is used to represent the temperature change characteristics of the battery pack and the state change characteristics of the battery pack in the charging state and the discharging state; Fusion of the time series data and the feature vector to obtain a real-time temperature series of the battery pack; The historical temperature data of the battery pack is acquired, and the historical temperature data and the real-time temperature sequence are processed by the long short-term memory network to obtain the predicted temperature data.

6. The method according to claim 5, characterized in that Determining the abnormality of the battery pack according to the data analysis result, and executing a warning operation when the abnormality exceeds a preset threshold, including: Comparing the real-time temperature data with the historical temperature data to obtain a first comparison result, wherein the first comparison result is used to reflect the abnormality of the real-time temperature data; When the first comparison result indicates that the deviation value between the real-time temperature data and the historical temperature data exceeds the first preset threshold among the preset thresholds, it is determined that the real-time temperature data is abnormal, and a first-level warning operation among the warning operations is performed, wherein the first-level warning operation is used to generate warning information and send it to the target object.

7. The method according to claim 1, characterized in that Determining the abnormality of the battery pack according to the data analysis result, and executing a warning operation when the abnormality exceeds a preset threshold, further comprising: Comparing the predicted temperature data with the real-time temperature data to obtain a second comparison result, wherein the second comparison result reflects the abnormality degree of the predicted temperature data; When the second comparison result indicates that the deviation value between the predicted temperature data and the real-time temperature data exceeds the second preset threshold among the preset thresholds, it is determined that the predicted temperature data is abnormal, and a secondary warning operation among the warning operations is executed, wherein the secondary warning operation is used to trigger an automatic cooling mechanism.

8. A temperature monitoring device, characterized in that: include: An acquisition module, used to acquire temperature data and status video of the battery pack, wherein the temperature data includes temperature values ​​of the battery pack at different times, and the status video includes video data of the battery pack during charging and discharging; an analysis module, configured to analyze the temperature data and the status video to obtain a data analysis result, wherein the data analysis result at least includes real-time temperature data and predicted temperature data of the battery pack, the real-time temperature data is used to indicate the real-time temperature change of the battery pack, and the predicted temperature data is used to indicate the temperature change trend of the battery pack in a future time period; The early warning module is used to determine the abnormality degree of the battery pack according to the data analysis result, and perform an early warning operation when the abnormality degree exceeds a preset threshold.

9. An electronic device, characterized in that: include: A memory and a processor, wherein the memory is used to store program instructions; The processor is connected to the memory, and is used to execute the temperature monitoring method described in any one of claims 1 to 7.

10. A non-volatile storage medium, characterized in that: The non-volatile storage medium includes a stored computer program, wherein the device where the non-volatile storage medium is located executes the temperature monitoring method according to any one of claims 1 to 7 by running the computer program.

11. A computer program product comprising computer instructions, characterized in that: When the computer instructions are executed by a processor, the temperature monitoring method according to any one of claims 1 to 7 is implemented.

Images