Safety monitoring system for IPFS (Internet Protocol File System) server room

Abstract

The invention provides an IPFS server machine room safety monitoring system, which relates to the technical field of machine room monitoring and comprises a data acquisition module for acquiring machine room state parameters and a fire judgment module for obtaining a fire judgment image and giving an alarm when fire electric equipment is marked on the fire judgment image. The feature extraction module is used for acquiring the state parameter features, and the fire prediction module is used for combining the state parameter features and the state parameters to generate a fire risk assessment index used for assessing the fire risk of the corresponding electric equipment and giving an alarm when the fire risk assessment index is larger than a fire risk threshold value. According to the invention, when the fire risk assessment index is calculated, multiple influence factors of fire hazards of the electric equipment are comprehensively considered, the accuracy of fire hazard assessment of the electric equipment is improved, and the alarm accuracy of a safety monitoring system is ensured.

Description

Technical Field

[0001] The present invention relates to the technical field of computer room monitoring, and in particular to an IPFS server computer room security monitoring system. Background Art

[0002] Currently, we are in an era of high-speed information exchange and dissemination, and information networks have been closely integrated with our daily office, life and study. As an important place for information processing and exchange, the IPFS server room is used to house a variety of important equipment, such as network switches, server clusters, storage, data input / output wiring, network monitoring terminals, etc. In order to ensure the stable operation of the equipment in the room, it is necessary to monitor the electrical equipment in the room to prevent abnormal temperature rise of the electrical equipment, which may easily cause fire.

[0003] In the prior art, a computer room temperature monitoring method, device, equipment and storage medium (classification number G06F) with publication number "CN111240943B" discloses the following contents: obtaining temperature values ​​collected by multiple temperature sensors distributed in the computer room; comparing the multiple temperature values ​​with sensitive temperature reference values ​​respectively, and determining the temperature value exceeding the sensitive temperature reference value as the sensitive temperature, wherein the sensitive temperature reference value is lower than the upper limit of the standard temperature range of the computer room; clustering the multiple sensitive temperatures according to their respective positions in the computer room to obtain at least one heating area, wherein each heating area includes at least one sensitive temperature; determining the temperature change trend according to the at least one heating area and the historical heating areas in the computer room, and determining whether to issue a temperature alarm according to the temperature change trend.

[0004] However, the above-mentioned prior art still has major defects. For example, the above-mentioned prior art only considers the single factor of temperature in the temperature alarm judgment. When conducting safety monitoring of electrical equipment in the computer room to prevent fire, not only the temperature factor of the electrical equipment should be considered, but also the current, operating time, humidity and other factors that are likely to cause safety hazards should be considered to ensure the accuracy of safety monitoring and fire prevention. Therefore, directly using the above-mentioned temperature monitoring method for safety monitoring has a certain one-sidedness, and cannot accurately evaluate the possibility of fire risk of electrical equipment in the computer room, and is prone to false alarms or no alarms. That is, the safety monitoring system arranged by the above-mentioned prior art solution will have the defect of inaccurate alarms.

[0005] The above information disclosed in this Background section is only for enhancement of understanding of the background of the present disclosure and therefore it may contain information that does not constitute the prior art that is already known to one of ordinary skill in the art. Summary of the invention

[0006] The purpose of the present invention is to provide an IPFS server room security monitoring system to solve the problems raised in the above background technology.

[0007] To achieve the above object, the present invention provides the following technical solutions: An IPFS server room security monitoring system includes the following modules: A data acquisition module, which is used to obtain the state parameters of the IPFS server room at each collection time within the monitoring period, including the infrared image, color image and humidity in the room, as well as the current and operating time of the electrical equipment in the room; An image processing module, which is used to process the color image and the infrared image to obtain a color-annotated image and a sub-infrared image of each electrical device in the computer room; A fire judgment module, which is used to process the color-annotated image and the sub-infrared image to obtain a fire judgment image, and to generate an alarm when a fire-caused electrical device is marked on the fire judgment image; A feature extraction module, which is used to extract features of the sub-infrared image, humidity, and current within the monitoring time period to obtain state parameter features, including the average temperature of the entire area, the average temperature of the over-temperature area, the average temperature of the risk area, the proportion of the risk area and the proportion of the over-temperature area at each acquisition time, as well as the average humidity, the average current, and the current standard deviation within the monitoring time period; The fire prediction module is used to combine the state parameter characteristics and the state parameters to generate a fire risk assessment index for evaluating the fire risk of the applied electrical equipment, and to issue an alarm when the fire risk assessment index is greater than the fire risk threshold.

[0008] Furthermore, the data acquisition module includes a shooting component for photographing electrical equipment in the computer room, and the shooting component includes a color camera for photographing color images and an infrared camera for photographing infrared images. The data acquisition module also includes a humidity sensor installed inside the computer room for humidity collection, and a UPS system installed in the computer room, and the UPS system is used to obtain current and operating time data of the electrical equipment.

[0009] Furthermore, the image processing module is composed of a labeling model and a segmentation model, wherein the labeling model is constructed based on the YOLO network, and the segmentation model is constructed based on the Mask R-CNN model. The labeling model is used to process the color image to obtain a color labeled image with electrical equipment labeled, and the segmentation model segments the infrared image based on the color labeled image to obtain a sub-infrared image.

[0010] Furthermore, the fire judgment module includes a fire judgment model constructed based on a convolutional neural network, a single-chip microcomputer, and an alarm module electrically connected to the single-chip microcomputer and configured to perform an alarm. The fire judgment model is configured to process color-annotated images and sub-infrared images to obtain fire judgment images. The alarm module includes a sonic alarm that generates an alarm, and a signal transmitter that transmits image data to a central control room. The image data is a fire judgment image that is marked with electrical equipment on fire. The single-chip microcomputer is configured to perform an alarm judgment based on the fire judgment image processed by the fire judgment model, specifically, to retrieve the annotation information of the fire judgment image to determine whether electrical equipment on fire is marked in the fire judgment image, and when electrical equipment on fire is marked on the fire judgment image, to control the alarm module to perform an alarm.

[0011] Furthermore, the fire prediction module includes a single chip microcomputer and an alarm electrically connected thereto, and the single chip microcomputer is used to generate a fire risk assessment index, and to control the alarm to sound an alarm when the fire risk assessment index is greater than a fire risk threshold, and the alarm is a sonic alarm that sounds an alarm.

[0012] Furthermore, the fire prediction module also includes an image annotation device, which is used to mark the electrical equipment whose fire risk assessment index is greater than the fire risk threshold in the color annotated image, use the annotated color annotated image as the fire prediction image, and transmit the fire prediction image to the central control room via a wired medium or a wireless medium.

[0013] Furthermore, the calculation logic of the fire risk assessment index is: based on the state parameter characteristics and the state parameters, a temperature risk assessment index for assessing the temperature risk status of electrical equipment and an operation risk assessment index for assessing the operation risk status of electrical equipment are generated; and then the temperature risk assessment index and the operation risk assessment index are combined to generate a fire risk assessment index for assessing the possibility of fire risk in electrical equipment.

[0014] Further, the calculation logic of the temperature risk assessment index is: based on the state parameter characteristics, an overtemperature assessment index, a risk assessment index and a risk rise assessment index are generated, and the temperature risk assessment index for assessing the temperature risk status of the electrical equipment is generated by combining the above three indexes; The calculation formula of the over-temperature assessment index is as follows: In the formula, is the proportion of over-temperature area of ​​the electrical equipment at the i-th collection time, i is the index of the collection time in the monitoring time period, and , is the total number of collection times during the monitoring period, and is a positive integer greater than 10, is the growth rate of the over-temperature area of ​​the electrical equipment at the i-th collection moment; In the formula, is the kth positive over-temperature region growth rate of the electrical equipment within the monitoring time period, that is, the kth over-temperature region growth rate greater than 0 among the I-1 over-temperature region growth rates, k is the index of the positive over-temperature region growth rate, and , K is the number of positive overtemperature area growth rates during the monitoring period; In the formula, , , are the proportion of over-temperature area, the average temperature of the whole area, and the average temperature of the over-temperature area of ​​the electrical equipment at the I collection time, is the allowable temperature threshold corresponding to the electrical equipment; In the formula, It is the over-temperature assessment index of the electrical equipment during the monitoring period. , are preset weights, and = ; The calculation formula of the risk assessment index is: In the formula, is the risk area proportion of electrical equipment at the i-th collection moment, is the risk area growth rate of the electrical equipment at the i-th collection moment; In the formula, is the growth rate of the jth positive risk area of ​​the electrical equipment during the monitoring period, that is, the growth rate of the jth risk area greater than 0 among the I-1 risk area growth rates, j is the index of the positive risk area growth rate, and , J is the number of growth rates in positive risk areas during the monitoring period, It is the average growth rate of the positive risk area of ​​the electrical equipment during the monitoring period; In the formula, , are the risk area proportion and the average temperature of the risk area of ​​the electrical equipment at the Ith collection time, respectively. is the risk temperature threshold corresponding to the electrical equipment; In the formula, is the risk assessment index of electrical equipment during the monitoring period. , are preset weights, and = ; The calculation formula of the risk assessment index is as follows: In the formula, is the growth rate of the risk area proportion in the over-temperature area of ​​the electrical equipment at the i-th collection moment, is the growth rate of the pth positive risk area proportion of electrical equipment in the monitoring period, that is, the growth rate of the pth risk area proportion greater than 0 among the I-1 risk area proportion growth rates, p is the index of the positive risk area growth rate, and , P is the number of positive risk areas with increasing proportion during the monitoring period; In the formula, It is the risk assessment index of electrical equipment during the monitoring period. , are preset weights, and = ; The calculation formula of the temperature risk assessment index is as follows: In the formula, It is the temperature risk assessment index of the electrical equipment during the monitoring period. , , are the weights of over-temperature assessment index, risk assessment index and risk rise assessment index in the calculation of temperature risk assessment index, respectively, and , , The specific value of is determined by the hierarchical analysis method.

[0015] Furthermore, the calculation formula of the operation risk assessment index is as follows: In the formula, is the threshold correction factor, is the operating time of the electric device at the Ith collection time, Ty' is the operating time threshold of the electric device, , are the current mean and current standard deviation during the monitoring period, is the average humidity during the monitoring period, is the lower limit of the suitable humidity range. is the upper limit of the suitable humidity range. is the humidity deviation value; In the formula, is the operational risk assessment index of the electrical equipment during the monitoring period. , , are the weights of humidity factor, current factor and operation time factor in the calculation of operation risk assessment index, and , , The specific value of is determined by the hierarchical analysis method.

[0016] Furthermore, the calculation formula of the fire risk assessment index is as follows: In the formula, is the fire risk assessment index of electrical equipment during the monitoring period. , , are preset weights, and + + =1, and > > .

[0017] Compared with the prior art, the present invention has the following beneficial effects: The IPFS server room security monitoring system of the present invention first makes a preliminary fire judgment on the electrical equipment in the IPFS server room by including color images and infrared images of the electrical equipment, and then combines and analyzes the temperature, current, operating time data of the electrical equipment and the humidity data in the room to obtain a fire risk assessment index for evaluating the possibility of fire risk of the electrical equipment. Compared with the prior art that only performs temperature judgment, the system comprehensively considers various influencing factors of fire hazards of electrical equipment, improves the accuracy of fire hazard assessment of electrical equipment, and thus ensures the alarm accuracy of the security monitoring system. BRIEF DESCRIPTION OF THE DRAWINGS

[0018] Figure 1 This is a module diagram of the IPFS server room security monitoring system in the present invention; Figure 2 is a schematic diagram of an image processing module in the present invention; Figure 3 is a schematic diagram of a fire judgment module in the present invention; Figure 4 It is a schematic diagram of the fire prediction module in the present invention. DETAILED DESCRIPTION

[0019] In order to make the objectives, technical solutions and advantages of the present invention more clearly understood, the present invention is further described in detail below in conjunction with specific embodiments.

[0020] It should be noted that, unless otherwise defined, the technical terms or scientific terms used in the present invention should be understood by people with ordinary skills in the field to which the present invention belongs. The words "first", "second" and similar words used in the present invention do not indicate any order, quantity or importance, but are only used to distinguish different components. "Include" or "comprise" and similar words mean that the elements or objects appearing before the word include the elements or objects listed after the word and their equivalents, without excluding other elements or objects. "Connect" or "connected" and similar words are not limited to physical or mechanical connections, but may include electrical connections, whether direct or indirect. "Up", "down", "left", "right" and the like are only used to indicate relative positional relationships. When the absolute position of the described object changes, the relative positional relationship may also change accordingly.

[0021] Example: See also Figure 1-Figure 4 The present invention provides an IPFS server room security monitoring system, including the following modules: A data acquisition module, which is used to obtain the state parameters of the IPFS server room at each collection time within the monitoring period, including the infrared image, color image and humidity in the room, as well as the current and operating time of the electrical equipment in the room; It should be noted that the monitoring time period is the time period before the current moment, and the monitoring time period can be three minutes, five minutes, 10 minutes, etc., without limitation here. The acquisition frequency of various status parameters is aligned on the timestamp, and the specific acquisition frequency can be set to once per second, once per three seconds, once per five seconds, etc., without limitation here. Among them, the data acquisition module includes a shooting component for shooting the electrical equipment in the computer room. The number and placement of the shooting components are determined according to the number and distribution of electrical equipment in the IPFS server computer room to ensure that the shot images can cover all electrical equipment in the IPFS server computer room. There is no restriction here. Each shooting component includes a color camera for shooting color images and an infrared camera for shooting infrared images. The color camera and the infrared camera have the same shooting angle and shooting position to ensure that the color camera and the infrared camera shoot the same electrical equipment to obtain the color image and infrared image of the electrical equipment. The setting of the color camera and the infrared camera with the same shooting angle and shooting position also facilitates the registration of the color image and the infrared image, thereby improving the accuracy of the subsequent image segmentation model in segmenting the infrared image. Among them, the data acquisition module includes a humidity sensor installed inside the computer room for humidity collection, and a UPS (uninterruptible power supply) system installed in the computer room. The UPS system is used to obtain the current and operating time data of the electrical equipment. The operating time is the cumulative working time of the electrical equipment from the first start of work to the collection time. In addition, the humidity sensor can be set to multiple and evenly arranged inside the computer room, and the average of the humidity data detected by all humidity sensors is used as the final humidity.

[0022] An image processing module, which is used to process the color image and the infrared image to obtain a color-annotated image and a sub-infrared image of each electrical device in the computer room; As an implementation mode, the image processing module is composed of a labeling model and a segmentation model, wherein the labeling model is constructed based on a YOLO network, and the segmentation model is constructed based on a Mask R-CNN model, the labeling model is used to process a color image to obtain a color labeled image with electrical equipment labeled, and the segmentation model segments the infrared image based on the color labeled image to obtain a sub-infrared image; Before using the annotation model and the segmentation model to process the color images and infrared images within the monitoring period, the annotation model and the segmentation model need to be trained first, as follows: Training the annotation model includes: photographing each electrical equipment in the computer room under different shooting angles and lighting conditions, obtaining a large number of color images containing electrical equipment to form a data set, using image annotation software to annotate the color images in the data set. Specifically, the color images can be annotated by using image annotation software such as LabelMe, VGG Image Annotator, CVAT, etc. The annotation method is bounding box annotation, specifically drawing a rectangular box including the electrical equipment, dividing the annotated data set into a training set and a test set at a ratio of 70 to 30, using the training set to train the annotation model, and selecting a framework such as TensorFlow / Keras or PyTorch as a deep learning framework to configure the training environment, and selecting a suitable learning rate (such as between 0.01-0.001) and batch size (such as 32 or 64) to constrain the training process, obtaining the prediction result through forward propagation, calculating the loss value through a loss function (such as a cross entropy loss function), and performing a reverse propagation. The model parameters are updated through propagation. When the predetermined number of training rounds is reached (determined according to the number of electrical equipment in the computer room, such as setting the number of training rounds to 30-80), the training is stopped. The performance of the labeling model is evaluated through the test set. Specifically, the model performance can be evaluated through indicators such as accuracy, precision, and recall. When the labeling model reaches the predetermined standard (such as accuracy greater than 95%), it means that the labeling model training is completed. Then, the color images within the monitoring time period are input into the labeling model after training to obtain the corresponding color labeling images. If the labeling model does not reach the predetermined standard, the model-related hyperparameters (such as learning rate, batch size, etc.) are changed and retrained until the labeling model reaches the predetermined standard. Training the segmentation model includes: photographing each electrical device in the computer room under different shooting angles and lighting conditions, obtaining a large number of color images and infrared images containing electrical devices, inputting the color images into the trained annotation model to obtain color annotated images, segmenting the infrared images based on the color annotated images, to obtain sub-infrared images of each electrical device in the infrared image, specifically, using image segmentation tools (such as GIMP, ImageJ, RectLabel, etc.) for manual segmentation, combining the color annotated images, infrared images, and the sub-infrared images obtained after segmentation to form a data set, and dividing the data set into a training set, a validation set, and a test set at a ratio of 70:15:15, using the training set to train the segmentation model, specifically using the color annotated images and the corresponding infrared images as input, and using the sub-infrared images obtained by segmentation as labels to train the segmentation model, and selecting TensorFlow / Keras or PyT during training. Orch and other frameworks are used as deep learning frameworks to configure the training environment, select appropriate optimizers (such as Adam, SGD, etc.), learning rates (such as between 0.01-0.001) and batch sizes (such as 32 or 64) to constrain the training process, use the validation set to monitor the loss and accuracy during the training process to prevent overfitting, which can be monitored through visualization tools (such as TensorBoard), and use the test set to evaluate the performance of the segmentation model. Specifically, the model performance can be evaluated through indicators such as intersection-over-union ratio, pixel accuracy and F1-score. When the segmentation model reaches the predetermined standard (such as pixel accuracy greater than 90%), it means that the segmentation model training is completed, and the color annotated image and infrared image within the monitoring time period are input into the segmentation model after training to obtain the corresponding sub-infrared image. If the segmentation model does not meet the predetermined standard, the model-related hyperparameters (such as learning rate, batch size, etc.) are changed and retrained until the segmentation model meets the predetermined standard; A fire judgment module, which is used to process the color-annotated image and the sub-infrared image to obtain a fire judgment image, and to generate an alarm when a fire-caused electrical device is marked on the fire judgment image; As an implementation mode, the fire judgment module includes a fire judgment model constructed based on a convolutional neural network, a single-chip microcomputer, and an alarm module electrically connected to the single-chip microcomputer and alarming. The fire judgment model is used to process the color-annotated image and the sub-infrared image to obtain the fire judgment image. The specific working principle is: when the fire judgment image corresponding to any acquisition time within the monitoring time period is marked with an electrical equipment on fire, a fire alarm is generated, and the fire judgment image marked with the electrical equipment on fire is sent to the central control room. The alarm module includes a sonic alarm for generating an alarm, and a signal transmitter for transmitting image data to the central control room. The image data is the fire judgment image marked with the electrical equipment on fire. The single-chip microcomputer is used to perform alarm judgment according to the fire judgment image processed by the fire judgment model, specifically, to retrieve the annotation information of the fire judgment image to determine whether the fire judgment image is marked with an electrical equipment on fire, and when the fire judgment image is marked with an electrical equipment on fire, the alarm module is controlled to perform alarm work; It should be noted that the infrared image is first segmented, and then the color-annotated image and the segmented sub-infrared image are input into the setting of the fire judgment model. The segmented sub-infrared image can better identify whether the electrical equipment is on fire at a smaller scale, and it is also easier to correspond one-to-one with the electrical equipment on the color-annotated image. Compared with inputting the entire infrared image into the fire judgment model, the accuracy of the model judgment is improved; Before using the fire judgment model to process the color-annotated images and sub-infrared images within the monitoring time period, the fire judgment model needs to be trained first. The process of training the fire judgment model is as follows: Photograph each electrical equipment in the computer room under different shooting angles and lighting conditions, obtain a large number of color images and infrared images when the electrical equipment is on fire and when the electrical equipment is not on fire, and process the color images and infrared images through an image processing module to obtain color-annotated images and corresponding sub-infrared images to form a data set, and use an image annotation tool to annotate the corresponding color-annotated image area according to whether the electrical equipment is on fire. Specifically, image annotation software such as LabelImg and VGG Image Annotator can be used to annotate the color-annotated image, and the annotation method is bounding box annotation. Specifically, if the electrical equipment is not on fire, "0" is annotated at the bounding box of the electrical equipment in the color-annotated image to indicate that the electrical equipment is a non-fired electrical equipment. Conversely, if the electrical equipment is on fire, "1" is annotated at the bounding box of the electrical equipment in the color-annotated image to indicate that the electrical equipment is a fire-occurring electrical equipment, and the annotated color-annotated image is used as a fire judgment image; The labeled data set is divided into training set, validation set and test set in the ratio of 70:20:10. The sub-infrared image and color labeled image in the training set are used as input, and the corresponding fire judgment image is used as the label to train the fire judgment model. A framework such as TensorFlow / Keras or PyTorch is selected as the deep learning framework to configure the training environment, and an appropriate learning rate (such as between 0.01-0.001) and batch size (such as 32 or 64) are selected to constrain the training process. The prediction results are obtained through forward propagation, and the loss value is calculated through the loss function (such as the binary cross entropy function). The model parameters are updated through back propagation. After reaching the predetermined number of training rounds (determined according to the number of electrical equipment in the computer room, such as setting the number of training rounds to The training is stopped at 30-80 seconds. The loss and accuracy during the training process are monitored through callback functions (such as ModelCheckpoint or EarlyStopping) to prevent overfitting. The performance of the annotation model is evaluated through the test set. Specifically, the model performance can be evaluated through indicators such as accuracy, precision, and recall. When the fire judgment model reaches the predetermined standard (such as accuracy greater than 95%), it means that the training of the fire judgment model is completed. The sub-infrared image and the color annotation image within the monitoring time period are input into the trained fire judgment model to obtain the fire judgment image. If the fire judgment model does not reach the predetermined standard, the model-related hyperparameters (such as learning rate, batch size, etc.) are changed and retrained until the fire judgment model reaches the predetermined standard. It should be noted that before the relevant images are processed by the annotation model, segmentation model and fire judgment model above, the images need to be preprocessed first. The preprocessing includes conventional normalization, resizing and other operations to ensure the consistency of the input data. This is the existing technology and will not be elaborated here.

[0023] A feature extraction module, which is used to extract features of the sub-infrared image, humidity, and current within the monitoring time period to obtain state parameter features, including the average temperature of the entire area, the average temperature of the over-temperature area, the average temperature of the risk area, the proportion of the risk area and the proportion of the over-temperature area at each acquisition time, as well as the average humidity, the average current, and the current standard deviation within the monitoring time period; Among them, the whole area refers to the entire area of ​​a single sub-infrared image, the overtemperature area is the area composed of pixels in a single sub-infrared image whose temperature exceeds the allowable temperature threshold, the risk area is the area composed of pixels in a single sub-infrared image whose temperature exceeds the risk temperature threshold, the risk area ratio is the ratio of the number of pixels in the sub-infrared image whose temperature is higher than the risk temperature threshold to the total number of pixels in the sub-infrared image, and the overtemperature area ratio is the ratio of the number of pixels in the sub-infrared image whose temperature is higher than the allowable temperature threshold to the total number of pixels in the sub-infrared image; It should be noted that the risk area ratio and over-temperature area ratio can be obtained through image processing software such as MATLAB or Python, and the relevant temperature can be obtained through infrared image processing software, which will not be elaborated here; It should be noted that the allowable temperature threshold is the maximum allowable operating temperature of the electrical equipment, and its value can be obtained according to the relevant operation manual of the electrical equipment. The risk temperature threshold is the temperature at which the electrical equipment is prone to fire risk, and its value can be determined by relevant experts, or the lowest ignition point of the electrical equipment is used as the risk temperature threshold, or the risk temperature threshold is selected within the range of 2-5 times the allowable temperature threshold, or the lowest temperature at which the electrical equipment starts to smoke during previous operation is used as the risk temperature threshold. There is no restriction here, but it should be noted that the lower the value of the risk temperature threshold, the higher the sensitivity of the fire prediction below, and the more likely it is to have a misjudgment problem, and the higher the value of the risk temperature threshold, the lower the sensitivity of the fire prediction below, and the more likely it is to have an alarm delay problem. The specific value is selected and adjusted by the staff according to the actual situation; It should be noted that the larger the proportion of risk areas, the larger the area of ​​electrical equipment prone to fire, and the greater the hidden danger of fire in electrical equipment; the larger the proportion of over-temperature areas, the larger the area of ​​electrical equipment exceeding the maximum allowable operating temperature, and the worse the safety performance of the electrical equipment.

[0024] A fire prediction module, wherein the fire prediction module is used to generate a fire risk assessment index for assessing the fire risk of the applied electrical equipment in combination with the state parameter characteristics and the state parameters, and to give an alarm when the fire risk assessment index is greater than a fire risk threshold, and the fire prediction module includes a single-chip microcomputer and an alarm electrically connected thereto, and the single-chip microcomputer is used to generate the fire risk assessment index, and to control the alarm to give an alarm when the fire risk assessment index is greater than the fire risk threshold, and the alarm is a sonic alarm that gives an alarm; Furthermore, the fire prediction module also includes an image annotation device, which has built-in image annotation software such as LabelMe, VGG Image Annotator or CVAT, and is used to annotate electrical equipment with a fire risk assessment index greater than a fire risk threshold in the color annotation image. For example, if the fire risk assessment index of the electrical equipment is not greater than the fire risk threshold, "0" is annotated at the boundary box of the electrical equipment in the color annotation image to indicate that the electrical equipment is an electrical equipment without a fire risk. Conversely, if the fire risk assessment index of the electrical equipment is greater than the fire risk threshold, "2" is annotated at the boundary box of the electrical equipment in the color annotation image to indicate that the electrical equipment is an electrical equipment with a fire risk. The annotated color annotation image is used as a fire prediction image, and the fire prediction image is transmitted to the central control room through a wired medium (such as optical fiber) or a wireless medium (such as Bluetooth), so that the staff can promptly inspect the electrical equipment with a fire risk. The calculation logic of the fire risk assessment index is as follows: based on the state parameter characteristics and the state parameters, a temperature risk assessment index for assessing the temperature risk status of electrical equipment and an operation risk assessment index for assessing the operation risk status of electrical equipment are generated; then, the temperature risk assessment index and the operation risk assessment index are combined to generate a fire risk assessment index for assessing the possibility of fire risk of electrical equipment; The calculation logic of the temperature risk assessment index is as follows: based on the state parameter characteristics, an over-temperature assessment index, a risk assessment index and a risk rise assessment index are generated, and the temperature risk assessment index used to assess the temperature risk status of electrical equipment is generated by combining the above three indexes; The calculation formula of the over-temperature assessment index is as follows: In the formula, is the proportion of over-temperature area of ​​the electrical equipment at the i-th collection time, i is the index of the collection time in the monitoring time period, and , is the total number of collection times during the monitoring period, and is a positive integer greater than 10, is the growth rate of the over-temperature area of ​​the electrical equipment at the i-th collection moment, which is used to measure the spread speed of the over-temperature area of ​​the electrical equipment from the i-1th collection moment to the i-th collection moment, and The larger it is, the faster the over-temperature area of ​​the electrical equipment spreads from the i-1th collection moment to the i-th collection moment, and the higher the temperature risk of the electrical equipment. In the formula, is the kth positive over-temperature region growth rate of the electrical equipment within the monitoring time period, that is, the kth over-temperature region growth rate greater than 0 among the I-1 over-temperature region growth rates, k is the index of the positive over-temperature region growth rate, and , K is the number of positive overtemperature area growth rates during the monitoring period, is the average growth rate of the positive overtemperature area of ​​the electrical equipment during the monitoring period, and The larger it is, the faster the over-temperature area of ​​the electrical equipment spreads, and the higher the temperature risk of the electrical equipment. Compared with the growth rate of the over-temperature area during the entire monitoring period, The setting only focuses on the growth rate of the positive over-temperature area, which can better reflect the spread speed of the over-temperature area of ​​the electrical equipment at adjacent collection times; In the formula, , , They are the proportion of over-temperature areas, the average temperature of the entire area, and the average temperature of the over-temperature area of ​​the electrical equipment at the first collection time. The collection times within the monitoring period are sorted in order, and the first collection time is the collection time closest to the current time. is the allowable temperature threshold corresponding to the electrical equipment; In the formula, The over-temperature assessment index of the electrical equipment during the monitoring period is used to comprehensively evaluate the temperature risk of the electrical equipment from three levels: the proportion of over-temperature areas, the temperature of over-temperature areas, and the temperature of the entire area. The larger the over-temperature assessment index, the higher the temperature risk of the electrical equipment. It should be noted that The larger the value, the more over-temperature areas the electrical equipment currently has, which means the higher the temperature risk of the electrical equipment. The larger the value is, the faster the over-temperature area of ​​the electrical equipment spreads at adjacent collection times, further increasing the temperature risk of the electrical equipment. Multiply The future over-temperature area ratio of electrical equipment is characterized in the form of. This product evaluates the temperature risk of electrical equipment from a future perspective, which is convenient for timely discovery of electrical equipment with hidden dangers. The larger the product, the higher the temperature risk of the electrical equipment and the larger the over-temperature assessment index. However, since it is uncertain whether the electrical equipment will heat up during the process from the first collection time to the future time, the To The reason for the correction is that the more times the proportion of the over-temperature area at the later acquisition time is greater than the proportion of the over-temperature area at the previous acquisition time during the monitoring period, The larger the value, the higher the probability of the increase of over-temperature area of ​​electrical equipment (for example, when When the value is equal to 1, it means that the proportion of over-temperature area at each subsequent acquisition moment in the monitoring period is greater than the proportion of over-temperature area at the previous acquisition moment, which means that the over-temperature area increases over time. When the value is equal to 0, it means that the proportion of over-temperature area at each subsequent collection time in the monitoring period is not greater than the proportion of over-temperature area at the previous collection time, which means that the over-temperature area has not increased over time). The setting converts the probability of the over-temperature area increasing into the rate of the over-temperature area increasing. That is, the over-temperature area will increase by default, but the rate of increase will increase with the increase of probability, so as to ensure the sensitivity of the temperature risk assessment of electrical equipment. The square setting is used for The correction indicates that as the probability increases, the rate of increase of the over-temperature area is also faster. This nonlinear setting further ensures the sensitivity of the temperature risk assessment of electrical equipment, and , and The larger the value, the more serious the over-temperature situation is, the higher the temperature risk of the electrical equipment is, and the larger the over-temperature assessment index is. It should be noted that the ratio The larger the ratio is, the closer the overall temperature of the electrical equipment at the first collection time is to the average temperature of the over-temperature area, that is, the more serious the overall temperature of the electrical equipment at the first collection time is, the greater the temperature risk is. On this basis, the ratio The larger the value is, the greater the degree to which the average temperature of the over-temperature area of ​​the electrical equipment at the first collection time exceeds the allowable temperature threshold, that is, the more serious the over-temperature situation of the electrical equipment at the first collection time is, and the overall temperature situation can more fundamentally reflect the temperature risk of the electrical equipment (for example, even if the ratio is No matter how big it is, if the ratio The risk of the overheating of the electrical equipment is relatively small, which only means that the overheating of the electrical equipment is relatively serious, far less than the risk caused by the overall temperature rise of the electrical equipment to close to the allowable temperature threshold). As the base number, As an index, The temperature risk severity of electrical equipment is characterized by the form of temperature. The setting of 1 avoids the problem of the base number being less than 1, and the ratio and ratio The larger it is, the higher the temperature risk of the electrical equipment is, and the greater the over-temperature assessment index is; In the formula, , are all preset weights. is the weight of the over-temperature area factor in the calculation of the over-temperature assessment index, is the weight of the temperature factor in the calculation of the over-temperature assessment index. Since the over-temperature area and temperature complement each other and influence each other in the temperature risk assessment, the same weight is given to them, that is, = =0.5; The calculation formula of the risk assessment index is: In the formula, is the risk area proportion of electrical equipment at the i-th collection moment, is the risk area growth rate of the electrical equipment at the i-th collection time, which is used to measure the risk area spread speed of the electrical equipment from the i-1th collection time to the i-th collection time, and The larger it is, the faster the risk area of ​​the electrical equipment spreads from the i-1th collection moment to the i-th collection moment, and the higher the temperature risk of the electrical equipment. In the formula, is the growth rate of the jth positive risk area of ​​the electrical equipment during the monitoring period, that is, the growth rate of the jth risk area greater than 0 among the I-1 risk area growth rates, j is the index of the positive risk area growth rate, and , J is the number of growth rates in positive risk areas during the monitoring period, is the average growth rate of the positive risk area of ​​the electrical equipment during the monitoring period, and The larger the value is, the faster the risk area of ​​the electrical equipment spreads, and the higher the temperature risk of the electrical equipment. Compared with the growth rate of the risk area during the entire monitoring period, The setting only focuses on the growth rate of positive risk areas, which can better reflect the spread speed of risk areas of electrical equipment at adjacent collection times; In the formula, , are the risk area proportion and the average temperature of the risk area of ​​the electrical equipment at the Ith collection time, respectively. is the risk temperature threshold corresponding to the electrical equipment; In the formula, The risk assessment index of the electrical equipment during the monitoring period is used to comprehensively evaluate the temperature risk of the electrical equipment from three levels: the proportion of risk areas, the temperature of risk areas, and the temperature of the entire area. The larger the risk assessment index, the higher the temperature risk of the electrical equipment. It should be noted that The larger the value, the more risk areas the electrical equipment currently has, which means the higher the temperature risk of the electrical equipment. The larger the value is, the faster the risk area of ​​the electrical equipment spreads at adjacent collection times, further increasing the temperature risk of the electrical equipment. Multiply The product represents the proportion of future risk areas of electrical equipment in the form of. This product evaluates the temperature risk of electrical equipment from a future perspective, making it easier to find electrical equipment with hidden dangers in a timely manner. The larger the product, the higher the temperature risk of the electrical equipment, and the greater the risk assessment index. However, since it is uncertain whether the electrical equipment will heat up during the process from the first collection time to the future time, the To The reason for the correction is that the more times the proportion of risk areas at the later collection time is greater than the proportion of risk areas at the previous collection time during the monitoring period, The larger the value, the higher the probability of an increase in the risk area of ​​electrical equipment (for example, when When the value is equal to 1, it means that the proportion of risk areas at each subsequent collection time in the monitoring period is greater than the proportion of risk areas at the previous collection time, which means that the risk area has been increasing over time. When the value is equal to 0, it means that the proportion of risk areas at each subsequent collection time during the monitoring period is not greater than the proportion of risk areas at the previous collection time, which means that the risk area has not increased over time). The setting converts the probability of risk area increase into the rate of risk area increase, that is, the default risk area will increase, but the rate of increase will increase with the increase of probability, so as to ensure the sensitivity of the temperature risk assessment of electrical equipment. The square setting is used for The correction indicates that as the probability increases, the risk area increases faster. This nonlinear setting further ensures the sensitivity of the temperature risk assessment of electrical equipment. , and The larger it is, the more serious the situation in the risk area is, the higher the temperature risk of the electrical equipment is, and the greater the risk assessment index is; It should be noted that the ratio The larger the ratio is, the closer the overall temperature of the electrical equipment at the first collection time is to the average temperature of the risk area, that is, the more serious the overall temperature of the electrical equipment at the first collection time is, the greater the temperature risk is. On this basis, the ratio The larger the value is, the greater the degree to which the average temperature of the risk area of ​​the electrical equipment at the I collection time exceeds the risk temperature threshold, that is, the more serious the temperature risk of the electrical equipment at the I collection time is, and the overall temperature situation can more fundamentally reflect the temperature risk of the electrical equipment (for example, even if the ratio is No matter how big it is, if the ratio It is small, which only means that the temperature risk of the local electrical equipment is more serious, which is far less than the risk caused by the overall temperature rise of the electrical equipment close to the risk temperature threshold). As the base number, As an index, The temperature risk severity of electrical equipment is characterized by the form of temperature. The setting of 1 avoids the problem of the base number being less than 1, and the ratio and ratio The larger it is, the higher the temperature risk of the electrical equipment is, and the greater the risk assessment index is; In the formula, , are all preset weights. is the weight of the risk area factor in the calculation of the risk assessment index, is the weight of the temperature factor in the calculation of the risk assessment index. Since the risk area and temperature complement each other and influence each other in the temperature risk assessment, the same weight is given to them, that is, = =0.5; The calculation formula of the risk assessment index is as follows: In the formula, is the growth rate of the risk area proportion in the over-temperature area of ​​the electrical equipment at the i-th collection moment, which is used to measure the spread speed of the risk area in the over-temperature area of ​​the electrical equipment from the i-1th collection moment to the i-th collection moment, and The larger it is, the faster the risk area in the over-temperature area spreads from the i-1th collection moment to the i-th collection moment. That is, the proportion of risk areas in the over-temperature area increases from the i-1th collection moment to the i-th collection moment, and the higher the temperature risk of the electrical equipment. In the formula, is the growth rate of the pth positive risk area proportion of electrical equipment in the monitoring period, that is, the growth rate of the pth risk area proportion greater than 0 among the I-1 risk area proportion growth rates, p is the index of the positive risk area growth rate, and , P is the number of positive risk areas with increasing proportion during the monitoring period, is the average growth rate of the proportion of positive risk areas of electrical equipment during the monitoring period, and The larger the value is, the faster the risk area in the over-temperature area of ​​the electrical equipment spreads, and the higher the temperature risk of the electrical equipment. Compared with the growth rate of the risk area proportion during the entire monitoring period, The setting only focuses on the growth rate of the proportion of positive risk areas, which can better reflect the spread speed of the proportion of risk areas in the over-temperature areas of electrical equipment at adjacent collection times; In the formula, The risk rise assessment index of the electrical equipment during the monitoring period is used to comprehensively evaluate the temperature risk of the electrical equipment from four levels: the proportion of over-temperature areas, the proportion of risk areas, the temperature of risk areas, and the temperature of over-temperature areas. The larger the risk rise assessment index, the higher the temperature risk of the electrical equipment. It should be noted that The larger the value is, the faster the risk area in the over-temperature area of ​​the electrical equipment spreads, that is, the greater the degree to which the over-temperature area of ​​the electrical equipment is transformed into the risk area, the higher the temperature risk of the electrical equipment. On this basis, the larger the value is, the higher the probability that the over-temperature area of ​​the electrical equipment will be transformed into the risk area in the future. Therefore, the use and The product of is used to characterize the impact of the increase in the proportion of risk areas in the over-temperature area on the temperature risk of electrical equipment. The reasons are as follows: The setting converts the probability of the increase of the risk area ratio in the over-temperature area into a rate, that is, the default risk area ratio in the over-temperature area will increase, but the rate of increase will increase with the increase of probability, so as to ensure the sensitivity of the temperature risk assessment of electrical equipment. The square setting is used for The correction indicates that as the probability increases, the rate at which the risk area proportion increases is also faster. This nonlinear setting further ensures the sensitivity of the temperature risk assessment of electrical equipment, and The larger the sum and P are, the more serious the increase in the proportion of risk areas in the over-temperature area is, the higher the temperature risk of the electrical equipment is, and the greater the risk rise assessment index is; It should be noted that the ratio The larger the ratio is, the closer the average temperature of the over-temperature area of ​​the electrical equipment at the first collection time is to the average temperature of the risk area, that is, the more serious the temperature situation of the over-temperature area of ​​the electrical equipment at the first collection time is, the greater the temperature risk is. On this basis, the ratio The larger the value is, the greater the degree to which the average temperature of the risk area of ​​the electrical equipment at the first collection time exceeds the risk temperature threshold, that is, the more serious the risk situation of the electrical equipment at the first collection time is. In addition, compared with the risk temperature, the temperature situation of the overtemperature area can more fundamentally reflect the temperature risk of the electrical equipment (for example, even if the ratio is No matter how big it is, if the ratio It is small, which only means that the risk of local electrical equipment is more serious, far less than the risk caused by the overall temperature rise of the over-temperature area of ​​the electrical equipment close to the risk temperature threshold). As the base number, As an index, The temperature risk severity of electrical equipment is characterized by the form of temperature. The setting of 1 avoids the problem of the base number being less than 1, and the ratio and ratio The larger it is, the higher the temperature risk of the electrical equipment is, and the greater the risk assessment index is; In the formula, , are all preset weights. is the weight of the risk area ratio in the overtemperature area in the risk assessment index calculation, is the weight of the temperature factor in the calculation of the risk assessment index. Since the proportion of risk areas in the overtemperature area and the temperature are complementary and mutually influential in the temperature risk assessment, the same weight is given to them, that is, = =0.5; The calculation formula of the temperature risk assessment index is as follows: In the formula, The temperature risk assessment index of the electrical equipment during the monitoring period is used to combine the over-temperature assessment index, risk assessment index and risk increase assessment index to comprehensively evaluate the temperature risk status of the electrical equipment from three levels: the whole area-over-temperature area, the whole area-risk area, and the over-temperature area-risk area. The larger the temperature risk assessment index, the higher the temperature risk of fire in the electrical equipment. In the formula, , , are the weights of over-temperature assessment index, risk assessment index and risk rise assessment index in the calculation of temperature risk assessment index, respectively, and , , The specific value of is determined by the hierarchical analysis method, and the specific logic is as follows: The three indicators of overheating assessment index, risk assessment index and risk increase assessment index are marked, and the relative importance values ​​between them are determined by the nine-scale method to construct a judgment matrix, in which the index of overheating assessment index is marked as 1, the index of risk assessment index is marked as 2, and the index of risk increase assessment index is marked as 3. The constructed judgment matrix is: Among them, f and v both represent the index of the index, and , , Indicates the importance of the index f relative to the index v for the temperature risk assessment index, The specific value is determined by relevant experts using a 1-9 scoring method. Indicates that the index f is more important for the temperature risk assessment index than the index v. It means that the index with index f is extremely unimportant to the temperature risk assessment index compared with the index with index v; Divide each element value in the judgment matrix by the sum of its columns to obtain a normalized judgment matrix, calculate the mean of the element values ​​in each row of the normalized judgment matrix, and use the mean of the element values ​​in the first row as the proportional coefficient of the overtemperature assessment index, the mean of the element values ​​in the second row as the proportional coefficient of the risk assessment index, and the mean of the element values ​​in the third row as the proportional coefficient of the risk increase assessment index. With the constraint that the sum of the scaled values ​​is equal to 1, scale the three proportional coefficients in equal proportions, and use the scaled values ​​as the weights of the corresponding indexes; The calculation formula of the operation risk assessment index is as follows: In the formula, is the operating time of the electrical equipment at the Ith collection moment, Ty' is the operating time threshold of the electrical equipment, and the reference service life of the electrical equipment can be used as the operating time threshold, which can be obtained from its product performance table or the manufacturer. , are the current mean and current standard deviation of the electrical equipment during the monitoring period, The humidity average value during the monitoring period is used to reflect the average humidity condition of the computer room during the monitoring period. is the lower limit of the suitable humidity range. is the upper limit of the suitable humidity range. It is the humidity deviation value, which is used to measure the degree to which the humidity in the computer room deviates from the appropriate humidity range during the monitoring period. The larger the humidity deviation value, the worse the humidity in the computer room during the monitoring period. The reasons are as follows: Under too low humidity, electrical equipment will be interfered by static electricity, and it is more likely to short-circuit and heat up due to static electricity interference. Under higher humidity, electrical equipment is more likely to corrode and short-circuit and heat up due to radiation. Therefore, higher or lower humidity will have an adverse effect on electrical equipment and increase the operating risk of electrical equipment. Therefore, the humidity deviation value is used as an influencing factor for calculating the operating risk assessment index. It should be noted that the appropriate humidity range of each electrical equipment can be obtained through the relevant operation manual or by asking the merchant, which will not be elaborated here. In the formula, It is the operation risk assessment index of the electrical equipment during the monitoring period. The operation risk assessment index is used to comprehensively assess the operation risk of the electrical equipment from three aspects: humidity, current of the electrical equipment and operation time. The larger the operation risk assessment index, the higher the operation risk of the electrical equipment and the more likely it is to have overheating and fire problems. It should be noted that the larger the humidity deviation value, the worse the humidity in the room during the monitoring period, and the higher the operating risk of the electrical equipment. As the temperature deviation value continues to increase, the operating risk caused by humidity to the electrical equipment becomes more significant. Therefore, the use of exponentially increasing To characterize the significant impact of humidity on the operating risk of electrical equipment, the larger the humidity deviation value, the higher the operating risk of the electrical equipment, and the larger the operating risk assessment index; It should be noted that the larger the current mean, the larger the current of the electrical equipment during the monitoring period, that is, the larger the load of the electrical equipment, the easier it is to overheat and increase the operating risk. On this basis, the larger the current standard deviation, the more unstable the current of the electrical equipment during the monitoring period. Unstable current fluctuations will cause the thermal stress and mechanical stress of the electrical equipment to increase, thereby increasing the possibility of failure of the electrical equipment, that is, increasing the possibility of overheating due to failure, and increasing the operating risk of the electrical equipment. Especially under the premise of high current mean, high current standard deviation will make the electrical equipment withstand a large current shock in an instant, thereby significantly increasing the possibility of rapid heating of the electrical equipment. As the base number, As an index, The nonlinear influence of current factors on the operation risk of electrical equipment is comprehensively characterized in the form of, and the larger the current mean and current standard deviation, the worse the power consumption of the electrical equipment, the more likely it is to have operation risks, and the larger the operation risk assessment index; It should be noted that The larger the value is, the longer the working time of the electrical equipment is at the current moment, the more serious the temperature rise of the electrical equipment due to aging is, and the higher the operating risk of the electrical equipment is. Characterize the nonlinear effect of the operating time factor on the operating risk of electrical equipment. The reasons are as follows: before the operating time of the electrical equipment is lower than the operating time threshold, the electrical equipment will have better operating performance and lower aging degree. At this time, the operating risk of the electrical equipment is low. However, when the operating time of the electrical equipment exceeds the operating time threshold, the aging degree of the electrical equipment will increase significantly over time, greatly increasing the operating risk of the electrical equipment. Therefore, the operating risk of the electrical equipment will first increase slowly with the operating time. After reaching a critical point (i.e., the operating time threshold), the operating risk of the electrical equipment will increase sharply with the operating time. Therefore, the operating risk of the electrical equipment is adopted. Characterize the impact trend of the operating time factor on the operating risk of electrical equipment, and the operating time The larger it is, the greater the aging degree of the electrical equipment is, the more likely it is to have operating risks, and the greater the operating risk assessment index is; It should be noted that It is the threshold correction coefficient, which is used to correct the operation time threshold. Since the operation time threshold is a theoretical value, the actual critical point will fluctuate due to the influence of the operation environment of the electrical equipment. For example, when the operation environment of the electrical equipment is poor (such as an unfavorable operation environment of overload, overheating and overhumidity), the actual service life of the electrical equipment will be shortened, and the actual critical point should be less than the operation time threshold. If the electrical equipment is operated in a suitable operation environment, the actual service life of the electrical equipment will increase, and the actual critical point should be greater than the operation time threshold. Therefore, the threshold correction coefficient is used for correction. The value range of the threshold correction coefficient is within the range of 0.8-1.2. The specific value is determined by relevant experts according to the actual operation environment of the electrical equipment. For example, when the operation environment of the electrical equipment is poor, a smaller threshold correction coefficient (such as 0.8) is used, and when the operation environment of the electrical equipment is good, a larger threshold correction coefficient (such as 1.2) is used. In the formula, , , are the weights of humidity factor, current factor and operation time factor in the calculation of operation risk assessment index, and , , The specific value of is determined by the hierarchical analysis method. The specific logic is consistent with the above and will not be repeated here; The calculation formula of the fire risk assessment index is as follows: In the formula, It is the fire risk assessment index of the electrical equipment during the monitoring period. The fire risk assessment index is used to combine the temperature risk assessment index, the operation risk assessment index and the fire risk parameter interaction item to comprehensively assess the possibility of fire risk of the electrical equipment. The larger the fire risk assessment index, the higher the risk of fire of the electrical equipment. It should be noted that the temperature risk assessment index conducts a comprehensive analysis of temperature factors to assess the possibility of fire risks in electrical equipment, and the larger the temperature risk assessment index, the higher the risk of fire in electrical equipment. Similarly, the operation risk assessment index combines humidity, current and operating time, three parameters that affect the performance of electrical equipment, to assess the possibility of fire risks in electrical equipment, and the larger the operation risk assessment index, the higher the risk of fire in electrical equipment. In addition, when assessing the possibility of fire risks in electrical equipment, the overall temperature mean, humidity and current of the electrical equipment are important considerations and need to be focused on. Therefore, the product of humidity deviation value, current mean and the temperature mean of the entire area at the first collection time is introduced as an interaction term, so as to focus on these three important considerations when assessing the possibility of fire risks in electrical equipment. Because the product of the three is prone to cause extreme value problems, through In order to smooth the nonlinear effect of the interaction item on the possibility of fire in electrical equipment, the larger the humidity deviation value, the current mean value and the average temperature of the whole area at the Ith acquisition time, the larger the interaction item, which means that the state of the electrical equipment is worse, and the possibility of fire risk of the electrical equipment is higher; In the formula, , , are all preset weights. Specifically, it is the weight of the temperature risk assessment index in the calculation of the fire risk assessment index. Specifically, it is the weight of the operation risk assessment index in the calculation of the fire risk assessment index. Specifically, it is the weight of the interaction item in the calculation of the fire risk assessment index. Since the temperature factor reflects the possibility of fire in electrical equipment from an intuitive level, the temperature risk assessment index is given the largest weight. Since the factors considered by the interaction item have been mentioned in the calculation of the temperature risk assessment index and the operation risk assessment index, in order to avoid excessively increasing the weight of the interaction item in the calculation of the fire risk assessment index, the interaction item is given the smallest weight, and the operation risk assessment index is given the second largest weight. + + = 1, let > > ; As an implementation method, The value range is 0.5-0.7. The value range is 0.2-0.3, The value range is 0.1-0.2. The specific value is set by the staff according to the actual situation and is not limited here; It should be noted that the value of the fire risk threshold is determined by relevant experts based on previous data. For example, multiple sets of status parameters of the same type of electrical equipment within one hour before a fire occurred in the past are first obtained, and then the fire risk assessment index is obtained in the same way. The minimum value or average value of the fire risk assessment index is used as the fire risk threshold.

[0025] The above formulas are all dimensionless and numerical calculations. The formula is a formula for the most recent real situation obtained by collecting a large amount of data and performing software simulation. The preset parameters in the formula are set by technicians in this field according to actual conditions.

[0026] The above embodiments may be implemented in whole or in part by software, hardware, firmware or any other combination thereof. When implemented by software, the above embodiments may be implemented in whole or in part in the form of a computer program product. Those skilled in the art may appreciate that the units and algorithm steps of each example described in conjunction with the embodiments disclosed herein may be implemented by electronic hardware, or a combination of computer software and electronic hardware. Whether these functions are performed by hardware or software methods depends on the specific application and design constraints of the technical solution.

[0027] The units described as separate components may or may not be physically separated, and the components shown as units may or may not be physical units, and may be located in one place or distributed on multiple network units. Some or all of the units may be selected according to actual needs to achieve the purpose of the solution of this embodiment.

[0028] The above description is only a specific implementation manner of the present application, but the protection scope of the present application is not limited thereto. Any technician familiar with the technical field can easily think of changes or substitutions within the technical scope disclosed in the present application, which should be included in the protection scope of the present application.

Claims

1. An IPFS server room security monitoring system, characterized in that: Includes the following modules: A data acquisition module, which is used to obtain the state parameters of the IPFS server room at each collection time within the monitoring period, including the infrared image, color image and humidity in the room, as well as the current and operating time of the electrical equipment in the room; An image processing module, which is used to process the color image and the infrared image to obtain a color-annotated image and a sub-infrared image of each electrical device in the computer room; A fire judgment module, which is used to process the color-annotated image and the sub-infrared image to obtain a fire judgment image, and to generate an alarm when a fire-caused electrical device is marked on the fire judgment image; A feature extraction module, which is used to extract features of the sub-infrared image, humidity, and current within the monitoring time period to obtain state parameter features, including the average temperature of the entire area, the average temperature of the over-temperature area, the average temperature of the risk area, the proportion of the risk area and the proportion of the over-temperature area at each acquisition time, as well as the average humidity, the average current, and the current standard deviation within the monitoring time period; The fire prediction module is used to combine the state parameter characteristics and the state parameters to generate a fire risk assessment index for evaluating the fire risk of the applied electrical equipment, and to issue an alarm when the fire risk assessment index is greater than the fire risk threshold.

2. The IPFS server room security monitoring system according to claim 1 is characterized in that: The data acquisition module includes a shooting component for photographing electrical equipment in the computer room, the shooting component includes a color camera for photographing color images, and an infrared camera for photographing infrared images. The data acquisition module also includes a humidity sensor installed inside the computer room for humidity collection, and a UPS system installed in the computer room, the UPS system is used to obtain current and operating time data of the electrical equipment.

3. The IPFS server room security monitoring system according to claim 1 is characterized in that: The image processing module is composed of a labeling model and a segmentation model, wherein the labeling model is constructed based on the YOLO network, and the segmentation model is constructed based on the Mask R-CNN model. The labeling model is used to process the color image to obtain a color labeled image with electrical equipment labeled, and the segmentation model segments the infrared image based on the color labeled image to obtain a sub-infrared image.

4. The IPFS server room security monitoring system according to claim 1, characterized in that: The fire judgment module includes a fire judgment model constructed based on a convolutional neural network, a single-chip microcomputer, and an alarm module electrically connected to the single-chip microcomputer and alarming. The fire judgment model is used to process a color-annotated image and a sub-infrared image to obtain a fire judgment image. The alarm module includes a sonic alarm that generates an alarm, and a signal transmitter that transmits image data to a central control room. The image data is a fire judgment image marked with electrical equipment on fire. The single-chip microcomputer is used to retrieve the annotation information of the fire judgment image to determine whether the fire judgment image is marked with electrical equipment on fire, and when the fire judgment image is marked with electrical equipment on fire, the alarm module is controlled to alarm.

5. The IPFS server room security monitoring system according to claim 1 is characterized in that: The fire prediction module includes a single chip microcomputer and an alarm electrically connected thereto, and the single chip microcomputer is used to generate a fire risk assessment index, and when the fire risk assessment index is greater than a fire risk threshold, the alarm is controlled to sound an alarm, and the alarm is a sonic alarm that sounds an alarm.

6. The IPFS server room security monitoring system according to claim 5 is characterized in that: The fire prediction module also includes an image annotation device, which is used to annotate electrical equipment with a fire risk assessment index greater than a fire risk threshold in a color annotated image, use the annotated color annotated image as a fire prediction image, and transmit the fire prediction image to a central control room via a wired medium or a wireless medium.

7. The IPFS server room security monitoring system according to claim 1, characterized in that: The calculation logic of the fire risk assessment index is: based on the state parameter characteristics and the state parameters, a temperature risk assessment index for assessing the temperature risk status of electrical equipment and an operation risk assessment index for assessing the operation risk status of electrical equipment are generated; and then the temperature risk assessment index and the operation risk assessment index are combined to generate a fire risk assessment index for assessing the possibility of fire risk in electrical equipment.

8. The IPFS server room security monitoring system according to claim 7 is characterized in that: The calculation logic of the temperature risk assessment index is as follows: based on the state parameter characteristics, an overtemperature assessment index, a risk assessment index and a risk rise assessment index are generated, and the temperature risk assessment index for assessing the temperature risk status of the electrical equipment is generated by combining the above three indexes; The calculation formula of the over-temperature assessment index is as follows: In the formula, is the proportion of over-temperature area of ​​the electrical equipment at the i-th collection time, i is the index of the collection time in the monitoring time period, and , is the total number of collection times during the monitoring period, and is a positive integer greater than 10, is the growth rate of the over-temperature area of ​​the electrical equipment at the i-th collection moment; In the formula, is the kth positive over-temperature region growth rate of the electrical equipment within the monitoring time period, that is, the kth over-temperature region growth rate greater than 0 among the I-1 over-temperature region growth rates, k is the index of the positive over-temperature region growth rate, and , K is the number of positive overtemperature area growth rates during the monitoring period, It is the average growth rate of the positive over-temperature area of ​​the electrical equipment during the monitoring period; In the formula, , , are the proportion of over-temperature area, the average temperature of the whole area, and the average temperature of the over-temperature area of ​​the electrical equipment at the I collection time, is the allowable temperature threshold corresponding to the electrical equipment; In the formula, It is the over-temperature assessment index of the electrical equipment during the monitoring period. , are preset weights, and = ; The calculation formula of the risk assessment index is: In the formula, is the risk area proportion of electrical equipment at the i-th collection moment, is the risk area growth rate of the electrical equipment at the i-th collection moment; In the formula, is the growth rate of the jth positive risk area of ​​the electrical equipment during the monitoring period, that is, the growth rate of the jth risk area greater than 0 among the I-1 risk area growth rates, j is the index of the positive risk area growth rate, and , J is the number of growth rates in positive risk areas during the monitoring period, It is the average growth rate of the positive risk area of ​​the electrical equipment during the monitoring period; In the formula, , are the risk area proportion and the average temperature of the risk area of ​​the electrical equipment at the Ith collection time, respectively. is the risk temperature threshold corresponding to the electrical equipment; In the formula, is the risk assessment index of electrical equipment during the monitoring period. , are preset weights, and = ; The calculation formula of the risk assessment index is as follows: In the formula, is the growth rate of the risk area proportion in the over-temperature area of ​​the electrical equipment at the i-th collection moment, is the growth rate of the pth positive risk area proportion of electrical equipment in the monitoring period, that is, the growth rate of the pth risk area proportion greater than 0 among the I-1 risk area proportion growth rates, p is the index of the positive risk area growth rate, and , P is the number of positive risk areas with increasing proportion during the monitoring period, The average growth rate of the proportion of positive risk areas of electrical equipment during the monitoring period; In the formula, It is the risk assessment index of electrical equipment during the monitoring period. , are preset weights, and = ; The calculation formula of the temperature risk assessment index is as follows: In the formula, It is the temperature risk assessment index of the electrical equipment during the monitoring period. , , are the weights of over-temperature assessment index, risk assessment index and risk rise assessment index in the calculation of temperature risk assessment index, respectively, and , , The specific value of is determined by the hierarchical analysis method.

9. The IPFS server room security monitoring system according to claim 8, characterized in that: The calculation formula of the operation risk assessment index is as follows: In the formula, is the threshold correction coefficient, is the operating time of the electric device at the Ith collection time, Ty' is the operating time threshold of the electric device, , are the current mean and current standard deviation of the electrical equipment during the monitoring period, is the average humidity during the monitoring period, is the lower limit of the suitable humidity range. is the upper limit of the suitable humidity range. is the humidity deviation value; In the formula, is the operational risk assessment index of the electrical equipment during the monitoring period. , , are the weights of humidity factor, current factor and operation time factor in the calculation of operation risk assessment index, and , , The specific value of is determined by the hierarchical analysis method.

10. The IPFS server room security monitoring system according to claim 9, characterized in that: The calculation formula of the fire risk assessment index is as follows: In the formula, is the fire risk assessment index of electrical equipment during the monitoring period. , , are preset weights, and + + =1, and > > .

Images