Electrical cabinet fire extinguishing method based on multi-sensor collaborative analysis

Abstract

The invention provides an electrical cabinet fire extinguishing method based on multi-sensor collaborative analysis, and relates to the technical field of electrical cabinet fire extinguishment. Infrared data of electrical equipment are identified by establishing an infrared temperature identification model, and the electrical equipment is identified according to the identified temperature, position relation, electrical connection relation and average temperature in a cabinet of the electrical equipment; the equipment spontaneous combustion risk of the electrical equipment is assessed by calculating a spontaneous combustion assessment value, a fire assessment value is calculated according to the spontaneous combustion assessment value, position distribution, combustible gas concentration, oxygen concentration and arc probability of the electrical equipment, and the loss estimation coefficient of the electrical room is judged according to the emergency coefficient, the fire assessment value and the fire grade of the electrical cabinet. And constructing a fire extinguishing relation model among the fire extinguishing agent, the fire assessment value and the fire behavior grade, and calculating the use dosage of the fire extinguishing agent of each electrical cabinet by taking the minimum loss estimation coefficient of the electrical room as a target and the residual dosage and the maximum safety concentration of the fire extinguishing agent as constraint conditions.

Description

Technical Field

[0001] The present invention relates to the technical field of electrical cabinet fire extinguishing, and in particular to an electrical cabinet fire extinguishing method based on multi-sensor collaborative analysis. Background Art

[0002] In the operation of modern electrical equipment, especially in critical facilities such as electrical cabinets, electrical fires have become an increasingly serious safety hazard. Due to the large number of devices and complex electrical connections within electrical cabinets, and the difficulty in timely detecting and accurately judging fire risks when a fire occurs, traditional fire detection methods often suffer from problems such as slow response, low accuracy, and limitations. Existing fire detection technologies mainly rely on single sensors such as temperature sensors and smoke detectors, which are often unable to comprehensively and accurately monitor fire risks within electrical cabinets. At the same time, existing fire detection technologies do not fully consider the impact of abnormal temperatures in electrical equipment on related equipment, nor the impact of flammable gases volatilized in high-temperature environments on closed electrical cabinets. In addition, traditional fire extinguishing agent distribution methods are mostly set based on experience and rules, and are unable to dynamically adjust the amount of fire extinguishing agent used based on real-time fire assessment results, making it difficult to achieve the optimal fire extinguishing effect.

[0003] 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 form the prior art that is already known to a person of ordinary skill in the art. Summary of the Invention

[0004] The object of the present invention is to provide an electrical cabinet fire extinguishing method with multi-sensor collaborative analysis to solve the problems raised in the above background technology.

[0005] To achieve the above object, the present invention provides the following technical solutions: A multi-sensor collaborative analysis method for electrical cabinet fire extinguishing, comprising the following steps: Step 1: Obtain the schematic diagram of the electrical cabinet, determine the number of the electrical equipment based on the schematic diagram, and map the electrical equipment number to its location within the electrical cabinet. Using the acquired infrared image of the electrical cabinet and the location of the electrical equipment in the image, establish an infrared temperature recognition model to identify the infrared data of the electrical equipment. Step 2: Calculate the temperature impact value of the electrical equipment based on the identified temperature of the electrical equipment, its position in the electrical cabinet, and its electrical connection relationship in the schematic diagram. Assess the spontaneous combustion risk of the electrical equipment by calculating the spontaneous combustion assessment value based on the temperature of the electrical equipment, the temperature impact value, and the average temperature in the cabinet. Step 3: Obtain environmental data within the electrical cabinet and determine the fire risk of the electrical cabinet by calculating the fire risk coefficient based on the spontaneous combustion assessment value and location distribution of the electrical equipment. Determine the deflagration probability of the electrical cabinet based on the combustible gas concentration, oxygen concentration, and arc probability. Calculate the fire assessment value based on the deflagration probability and fire risk coefficient. Step 4: Assign an emergency coefficient to each electrical cabinet. Based on the fire assessment value and fire severity, determine the loss estimation coefficient of the electrical room. Based on the historical fire extinguishing effect of fire extinguishing agents, construct a fire extinguishing relationship model between the fire extinguishing agent, the fire assessment value, and the fire severity. Step 5: Obtain the remaining dosage of fire extinguishing agent. Using the linear programming method, with the dosage of fire extinguishing agent used in each electrical cabinet as the decision target, the minimum loss estimation coefficient of the power room as the goal, and the remaining dosage of fire extinguishing agent and the maximum safe concentration as constraints, calculate the dosage used in each electrical cabinet.

[0006] Furthermore, the infrared data of the electrical equipment includes the temperature of the electrical equipment, the arc generated by the electrical equipment, the fire level and the average temperature inside the cabinet; The specific method of establishing an infrared temperature recognition model based on the infrared image of the interior of the electrical cabinet and the position of the electrical equipment in the image is as follows: Obtain historical infrared images of electrical cabinets, mark the electrical equipment with the equipment number and infrared data according to their position in the image, and establish an infrared temperature recognition model based on a convolutional neural network. Use the historical infrared images as model input and the marked equipment number and infrared data of the electrical equipment as model output. Train the infrared temperature recognition model, and use real-time infrared images as model input to identify the equipment number and infrared data of the electrical equipment in the infrared image. The infrared temperature recognition model is based on a convolutional neural network, which specifically includes a convolution layer, an activation layer, a pooling layer and an output layer.

[0007] Furthermore, the electrical connection relationship includes upper end connection equipment and lower end connection equipment, and the position relationship includes upper, lower, left, right, front, and back. and the distance up, down, left, right, front, and back; The specific method for calculating the temperature impact value of electrical equipment is: in, is the temperature impact value of the electrical equipment, For the The temperature of the electrical equipment with the electrical connection relationship, is the temperature of the electrical equipment, For the The influence coefficient of the electrical connection relationship, is the number of electrical connection relationships, For the The temperature of electrical equipment in different positions, For the The distance of the position relationship, For the The influence coefficient of the position relationship, is the number of positional relationships, , , , are all positive integers, .

[0008] Furthermore, the spontaneous combustion evaluation value is calculated based on the temperature of the electrical equipment, the temperature impact value, and the average temperature in the cabinet. The method for assessing the risk of spontaneous combustion of electrical equipment is as follows: in, is the spontaneous combustion assessment value, is the temperature of the electrical equipment, is the temperature impact value of electrical equipment, is the average temperature inside the electrical cabinet, For the suitable working temperature of electrical equipment, is the upper limit of operating temperature; current When the device is in use, there is a risk of spontaneous combustion of electrical equipment; in, Evaluate thresholds for spontaneous combustion of electrical equipment.

[0009] Furthermore, the environmental data includes smoke concentration, combustible gas concentration, ambient temperature, and oxygen concentration; The specific method for determining the fire risk of an electrical cabinet by calculating the fire risk coefficient of the electrical cabinet is as follows: The process of obtaining the location distribution of electrical equipment is as follows: Obtain the dimensional data of the electrical cabinet, use the panel of the electrical cabinet as the coordinate plane, the bottom edge as the horizontal coordinate, the vertical edge as the vertical coordinate, the lower left corner as the coordinate origin, and the center point of the electrical equipment as the coordinate point. Calculate the spatial risk density of each electrical equipment based on the coordinate point and the spontaneous combustion assessment value using the Gaussian kernel density: in, For the The risk density of each electrical equipment location, For the The spontaneous combustion assessment value of each electrical equipment, For the The coordinate position of each electrical device, is the risk diffusion range, is the number of electrical equipment in the electrical cabinet, is the width of the electrical cabinet, is the length of the electrical cabinet; Calculate the fire risk coefficient based on the spatial risk density: in, is the fire risk factor.

[0010] Furthermore, the method for determining the probability of deflagration is: in, is the probability of deflagration, is the calibration factor, is the combustible gas concentration, , are the lower and upper limits of combustible gas explosion concentrations, is the oxygen concentration, is the probability of arcing; The method for calculating the fire assessment value based on the probability of deflagration and the abnormal temperature rise coefficient is: in, is the fire assessment value, , are the weights of deflagration probability and abnormal temperature rise coefficient, + .

[0011] Furthermore, the specific method for determining the loss estimation coefficient of the power room is: in, is the loss estimation coefficient of the power room, For fire level, For the The emergency factor allocated to each electrical cabinet.

[0012] Furthermore, the fire extinguishing relationship model between the fire extinguishing agent and the fire assessment value and fire level is based on a BP neural network, which specifically includes an input layer, a hidden layer, and an output layer: The amount of fire extinguishing agent used, fire assessment value, and fire level are used as model input data, and the fire assessment value and fire level after using the fire extinguishing agent are used as model output data to train the fire extinguishing relationship model. The model input data is input into the trained fire extinguishing relationship model to predict the fire assessment value and fire level after using the fire extinguishing agent.

[0013] Furthermore, the specific calculation method for minimizing the loss estimation coefficient of the power room to be the objective function is: in, is the objective function for minimizing the loss estimation coefficient of the power room, The dosage of the fire extinguishing agent output by the fire extinguishing relationship model After the fire level, For the The emergency factor assigned to each electrical cabinet, The dosage of the fire extinguishing agent output by the fire extinguishing relationship model Fire assessment value after The calculation method with the remaining dose of fire extinguishing agent and the maximum safe concentration as constraints is: in, For the The amount of fire extinguishing agent used in each electrical cabinet, is the remaining amount of fire extinguishing agent, is the extinguishing agent volume conversion coefficient, is the background concentration of fire extinguishing agent in the power room when in use, is the upper limit of safe concentration, is the volume of the power room.

[0014] Compared with the prior art, the present invention has the following beneficial effects: the present invention identifies infrared data of electrical equipment by establishing an infrared temperature recognition model, and evaluates the risk of spontaneous combustion of the electrical equipment by calculating a spontaneous combustion assessment value based on the identified temperature, position relationship, electrical connection relationship, and average temperature inside the cabinet of the electrical equipment; calculates a fire assessment value based on the spontaneous combustion assessment value, position distribution, combustible gas concentration, oxygen concentration, and arc probability of the electrical equipment; determines the loss estimation coefficient of the electrical room based on the emergency coefficient, fire assessment value, and fire level of the electrical cabinet; and constructs a fire extinguishing relationship model between the fire extinguishing agent, the fire assessment value, and the fire level; calculates the dosage of the fire extinguishing agent used in each electrical cabinet with the goal of minimizing the loss estimation coefficient of the electrical room and the remaining dosage and maximum safe concentration of the fire extinguishing agent as constraints; Based on multi-sensor collaborative analysis technology, the present invention can achieve real-time monitoring and evaluation of key factors such as temperature, fire risk, and spontaneous combustion risk of equipment in electrical cabinets, which has significant advantages over traditional fire detection methods. Through accurate temperature impact value calculation and environmental data analysis, the fire risk of the electrical cabinet can be evaluated more accurately, avoiding the problem of a single sensor being unable to cover the entire situation or incomplete information, thereby improving the accuracy and timeliness of fire warnings. In addition, combined with the linear programming method to optimize the distribution of fire extinguishing agents, personalized fire extinguishing strategies can be formulated according to the actual fire risk of each electrical cabinet, which not only improves the fire extinguishing efficiency, but also minimizes the waste of fire extinguishing agents and equipment damage, ensuring that the electrical cabinet can be most effectively protected in the event of a fire. BRIEF DESCRIPTION OF THE DRAWINGS

[0015] Figure 1 Schematic diagram of the overall method of the present invention. DETAILED DESCRIPTION

[0016] 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 with reference to specific embodiments.

[0017] It should be noted that, unless otherwise defined, the technical or scientific terms used in the present invention should have the usual meanings understood by people with ordinary skills in the field to which the present invention belongs. The "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 position relationships. When the absolute position of the object being described changes, the relative position relationship may also change accordingly.

[0018] Example: See also Figure 1 , the present invention provides a technical solution: A multi-sensor collaborative analysis method for electrical cabinet fire extinguishing, comprising the following steps: Step 1: Obtain the schematic diagram of the electrical cabinet, determine the number of the electrical equipment based on the schematic diagram, and map the number of the electrical equipment to the position in the electrical cabinet. By obtaining the infrared image inside the electrical cabinet and the position of the electrical equipment in the image, an infrared temperature recognition model is established to identify the infrared data of the electrical equipment.

[0019] Electrical equipment, including contactors, relays, and circuit breakers, are installed within electrical cabinets. When electrical equipment (such as circuit breakers and contactors) carries excessive current for extended periods, experiences a short circuit, or experiences a fault, the resistance of the equipment can cause it to heat up. For example, when the current is too high, the wires and contacts within the equipment can heat up dramatically due to resistance heat. If this overload condition persists, it can cause a fire. Furthermore, when a short circuit occurs, the current rapidly increases, causing the electrical equipment and cables to heat up rapidly, potentially leading to combustion or fire. Monitoring the temperature of electrical equipment can help identify potential fire hazards in advance. Abnormal temperatures are often an early sign of equipment failure. If overheating or fire risks are not addressed promptly, these equipment risks can spread and cause significant damage. Therefore, temperature monitoring is a crucial component of electrical cabinet fire prevention systems.

[0020] By obtaining the electrical cabinet schematic and mapping it to the electrical equipment numbers and locations, each device can be accurately located within the cabinet. This mapping provides a clear physical and logical location relationship for subsequent temperature monitoring and fire risk assessment, ensuring that the temperature of each device is accurately mapped to the actual device.

[0021] Convolutional neural networks (CNNs) are particularly adept at processing and analyzing image data. Infrared images contain a wealth of spatial information (such as temperature distribution and equipment heat sources). Through its multi-layered convolution operations, CNNs can efficiently extract image features, such as the distribution of heat sources and equipment overheating. Compared to traditional image processing methods, CNNs can more accurately and automatically identify and classify key elements in images (such as equipment temperature and arc sparks).

[0022] CNN can automatically learn and identify high-temperature areas based on the temperature distribution of electrical equipment in infrared images, and combine it with the positional relationship of the equipment to accurately assess the temperature conditions of the equipment.

[0023] Arcing is a common precursor to fire when electrical equipment fails. CNN can identify the presence of arcs through their unique brightness and patterns in infrared images and assess their possible impact on fire development.

[0024] By learning the temperature distribution inside the electrical cabinet, CNN can estimate the scale of the fire, identify the extent of the fire spread, and help determine the severity of the fire.

[0025] In this embodiment, the infrared data of the electrical equipment includes the temperature of the electrical equipment, the arc generated by the electrical equipment, the fire level and the average temperature inside the cabinet; The specific method of establishing an infrared temperature recognition model based on the infrared image of the interior of the electrical cabinet and the position of the electrical equipment in the image is as follows: Obtain historical infrared images of electrical cabinets, mark the electrical equipment with the equipment number and infrared data according to their position in the image, and establish an infrared temperature recognition model based on a convolutional neural network. Use the historical infrared images as model input and the marked equipment number and infrared data of the electrical equipment as model output. Train the infrared temperature recognition model, and use real-time infrared images as model input to identify the equipment number and infrared data of the electrical equipment in the infrared image. The infrared temperature recognition model is based on a convolutional neural network, which specifically includes a convolution layer, an activation layer, a pooling layer, and an output layer; The convolution layer extracts features from the input infrared image through convolution operation. The calculation formula is: in, is the input infrared image, is the convolution kernel, is the coordinate of the output pixel matrix, The convolution kernel is Row and The value of the column; The activation layer is calculated as: in, The coordinates of the output matrix of the convolutional layer output; The calculation formula of the pooling layer is: in, is the maximum pooling output; The calculation formula of the output layer is: in, is the output of the pooling layer, is the result of image recognition, is the weight, is the bias parameter.

[0026] Step 2: Calculate the temperature impact value of the electrical equipment based on the identified temperature of the electrical equipment, its position in the electrical cabinet, and its electrical connection relationship in the schematic diagram. Assess the spontaneous combustion risk of the electrical equipment by calculating the spontaneous combustion assessment value based on the temperature of the electrical equipment, the temperature impact value, and the average temperature in the cabinet.

[0027] Temperature changes in electrical equipment not only affect its own operating status but can also affect other nearby devices through heat conduction, radiation, and convection. Within an electrical cabinet, the layout and interconnections of equipment can lead to temperature accumulation or propagation. Therefore, understanding how the temperature of one device affects the operation of other equipment is crucial.

[0028] The placement of devices in an electrical cabinet determines the distance between them and the effectiveness of heat dissipation. If devices are closely spaced, the high temperature of one device can be transferred to other devices through thermal radiation and convection, increasing the risk of overheating. By calculating the influence coefficient between devices, the effect of temperature transfer can be quantified, helping to determine which devices are most likely to be affected. Different electrical devices form a network through electrical connections. Temperature changes between certain devices can affect other devices through electrical loops, load sharing, or current transfer. For example, if one device overheats, it may also cause the electrical connections to other devices to heat up, leading to circuit failure or short circuit. Therefore, understanding electrical connection relationships is crucial for assessing temperature impacts.

[0029] When a device in an electrical cabinet overheats, not only can that device malfunction, but other devices in the entire cabinet may also experience a chain reaction. For example, rising temperatures can cause insulation degradation, poor electrical contact, or even arcing or fire. By combining temperature information with the spatial layout and electrical connections of electrical equipment, a more comprehensive temperature impact profile can be constructed. Temperature changes in different devices may have time delays and nonlinear effects, and relying solely on device temperature data may not accurately predict the temperature trend of the entire system. However, by comprehensively considering the mutual impact of devices, future temperature trends of each device in the electrical cabinet can be better predicted.

[0030] In this embodiment, the electrical connection relationship includes an upper end connection device and a lower end connection device, and the position relationship includes upper, lower, left, right, front, back and the distance between the upper, lower, left, right, front and back; The specific method for calculating the temperature impact value of electrical equipment is: in, is the temperature impact value of the electrical equipment, For the The temperature of the electrical equipment with the electrical connection relationship, is the temperature of the electrical equipment, For the The influence coefficient of the electrical connection relationship, is the number of electrical connection relationships, For the The temperature of electrical equipment in different positions, For the The distance of the position relationship, For the The influence coefficient of the position relationship, is the number of positional relationships, , , , are all positive integers, .

[0031] By calculating the equipment spontaneous combustion assessment value, it is possible to accurately identify high-risk equipment within a wide range of electrical equipment. Temperature, temperature impact value, and average cabinet temperature contribute to the calculation of spontaneous combustion risk. Temperature is measured in degrees Celsius. The combination of these factors provides a more comprehensive reflection of the equipment's actual condition. For example, some equipment may have a relatively low spontaneous combustion risk even at higher temperatures due to a smaller temperature impact value or poor heat transfer from other equipment. Other equipment, however, may still present a higher spontaneous combustion risk even at lower temperatures due to their location or increased contact with heat sources. Relying solely on a single factor, such as temperature, to determine a device's spontaneous combustion risk can lead to an overly biased assessment. Temperature is only one important factor affecting equipment spontaneous combustion, but equipment layout, heat transfer paths, and cabinet temperature distribution can also influence the final spontaneous combustion risk. By combining multiple factors, such as temperature, temperature impact value, and average cabinet temperature, a more comprehensive and accurate risk assessment system can be formed. This assessment takes into account the effects of heat transfer between equipment and other devices, preventing the omission of equipment with lower surface temperatures that poses a higher risk due to temperature transfer or concentrated heat.

[0032] In this embodiment, the spontaneous combustion evaluation value is calculated based on the temperature of the electrical equipment, the temperature impact value, and the average temperature in the cabinet. The method for assessing the risk of spontaneous combustion of electrical equipment is as follows: in, is the spontaneous combustion assessment value, is the temperature of the electrical equipment, is the temperature impact value of electrical equipment, is the average temperature inside the electrical cabinet, For the suitable working temperature of electrical equipment, is the upper limit of operating temperature; current When the device is in use, there is a risk of spontaneous combustion of electrical equipment; in, Evaluate thresholds for spontaneous combustion of electrical equipment.

[0033] Step 3: Obtain environmental data inside the electrical cabinet, and determine the fire risk of the electrical cabinet by calculating the fire risk coefficient of the electrical cabinet based on the spontaneous combustion assessment value and location distribution of the electrical equipment. Determine the deflagration probability of the electrical cabinet based on the combustible gas concentration, oxygen concentration, and arc probability, and calculate the fire assessment value based on the deflagration probability and fire risk coefficient.

[0034] Obtaining the spontaneous combustion assessment values ​​and location distribution of electrical equipment enables a more comprehensive assessment of fire risk within electrical cabinets. Relying solely on the spontaneous combustion assessment values ​​of individual devices to determine fire risk may be inaccurate, as the location distribution of devices also significantly influences fire risk. If two devices with high spontaneous combustion assessment values ​​are located far apart, the fire risk between them will not simply increase. However, if they are located in close proximity, the fire risk may be even higher due to heat conduction and mutual influence. The spatial distribution of devices within an electrical cabinet influences the accumulation and transfer of heat between them. If devices are unevenly distributed, some devices may be affected by heat radiation or conduction from other devices, leading to excessively high temperatures and an increased fire risk. Accurately obtaining device location data on a coordinate plane allows for better modeling and calculation of these interactions. For example, if two devices with high spontaneous combustion assessment values ​​are located close together, the incremental risk at that location can be estimated based on spatial density.

[0035] Spatial risk density refers to the concentration of spontaneous combustion assessment values ​​for equipment at different locations within an electrical cabinet. High-risk areas are characterized by a higher concentration of equipment with high spontaneous combustion assessment values, posing a higher fire risk. If electrical equipment in certain areas is densely distributed and has high spontaneous combustion assessment values, the probability of fire in that area is higher. The overall fire risk of an electrical cabinet depends not only on the spontaneous combustion assessment values ​​of individual equipment but also on their spatial distribution. Densely distributed high-risk equipment has a greater impact on the overall fire risk of the cabinet.

[0036] Gaussian kernel density estimation is a nonparametric statistical method used to estimate the probability density of data points distributed in space. For equipment in electrical cabinets, Gaussian kernel density estimation can be used to estimate the risk density at each location. The basic idea behind Gaussian kernel density calculation is to use the spontaneous combustion assessment value of each device as the basis and calculate the density of that point and the surrounding area using a smoothing function (Gaussian kernel). Gaussian kernel density eliminates the excessive impact of individual device locations on risk by performing weighted smoothing on the spontaneous combustion assessment value of each device. This helps to more accurately assess the fire risk of the entire area. Gaussian kernel density fully considers the spatial distribution of equipment. Closely located devices with higher spontaneous combustion assessment values ​​are assigned a higher risk density, reflecting the greater fire risk in these areas.

[0037] In this embodiment, the environmental data includes smoke concentration, combustible gas concentration, ambient temperature, and oxygen concentration; The specific method for determining the fire risk of an electrical cabinet by calculating the fire risk coefficient of the electrical cabinet is as follows: The process of obtaining the location distribution of electrical equipment is as follows: Obtain the dimensional data of the electrical cabinet, which includes the length, width, and height of the electrical cabinet in centimeters. Use the panel of the electrical cabinet as the coordinate plane, the bottom edge as the horizontal coordinate, the vertical edge as the vertical coordinate, the lower left corner as the coordinate origin, and the center point of the electrical equipment as the coordinate point. Calculate the spatial risk density of each electrical equipment based on the coordinate point and the spontaneous combustion assessment value using the Gaussian kernel density: in, For the The risk density of each electrical equipment location, For the The spontaneous combustion assessment value of each electrical equipment, For the The coordinate position of each electrical device, is the risk diffusion range, is the number of electrical equipment in the electrical cabinet, is the width of the electrical cabinet, is the length of the electrical cabinet; Calculate the fire risk coefficient based on the spatial risk density: in, is the fire risk factor.

[0038] In this embodiment, the method for determining the probability of deflagration is: in, is the probability of deflagration, is the calibration factor, is the combustible gas concentration, , are the lower and upper limits of combustible gas explosion concentrations, is the oxygen concentration, is the probability of arcing; The method for calculating the fire assessment value based on the probability of deflagration and the abnormal temperature rise coefficient is: in, is the fire assessment value, , are the weights of deflagration probability and abnormal temperature rise coefficient, + .

[0039] By considering combustible gas concentration, oxygen concentration, and arc probability, a more accurate assessment can be made of the potential risk of a deflagration within an electrical cabinet. Changes in combustible gas and oxygen concentrations directly influence whether combustion conditions meet deflagration requirements. If arcing occurs during operation of electrical equipment, it is highly likely to cause a fire or deflagration. A comprehensive analysis of these three factors can determine whether a deflagration will occur. Equipment and wiring within an electrical cabinet may release combustible gases (such as carbon monoxide) when exposed to high temperatures or electrical faults. Increased concentrations of these gases increase the risk of a deflagration. Monitoring combustible gas concentrations can promptly detect precursors to a deflagration. Arcing is a common high-energy release process in electrical equipment, and the occurrence of arcing can quickly cause a fire or even a deflagration. Therefore, assessing the probability of arcing helps assess the risk of a deflagration within an electrical cabinet.

[0040] Combining the deflagration probability and fire risk factor allows for a more dynamic and comprehensive risk assessment. The fire risk factor typically considers factors such as the equipment's spontaneous combustion assessment value and location distribution, while the deflagration probability is based on more dynamic environmental factors, such as gas concentration and the potential for arcing. This comprehensive assessment captures potential hazardous changes within the electrical cabinet in real time and calculates a more accurate fire assessment.

[0041] Step 4: Assign an emergency coefficient to each electrical cabinet, determine the loss estimation coefficient of the electrical room based on the fire assessment value and fire level, and construct a fire extinguishing relationship model between the fire extinguishing agent, the fire assessment value, and the fire level based on the historical fire extinguishing effect of the fire extinguishing agent.

[0042] In this embodiment, the specific method for determining the loss estimation coefficient of the power room is: in, is the loss estimation coefficient of the power room, For fire level, For the The emergency factor allocated to each electrical cabinet.

[0043] BP neural networks are able to effectively handle complex nonlinear relationships. In fire scenarios, the relationship between the effectiveness of fire extinguishing agents and the fire assessment value and fire severity level is often nonlinear. For example, the severity of the fire, the effect of the amount of fire extinguishing agent used, and the fire assessment value all affect the fire extinguishing effect. BP neural networks can automatically learn and capture these complex nonlinear relationships, thereby providing more accurate predictions of fire extinguishing effectiveness. By establishing a model that can adapt to complex environments and changing conditions, it is possible to more accurately predict the fire extinguishing effect of the fire extinguishing agent dosage at different fire assessment values ​​and fire severity levels, ensuring that the fire extinguishing agent dosage can be selected in actual operations.

[0044] In this embodiment, the fire extinguishing relationship model between the fire extinguishing agent and the fire assessment value and fire level is constructed based on a BP neural network, which specifically includes an input layer, a hidden layer, and an output layer: The amount of fire extinguishing agent used, fire assessment value, and fire severity level are used as model input data. The fire assessment value and fire severity level after the fire extinguishing agent is used are used as model output data to train the fire extinguishing relationship model. The model input data is input into the trained fire extinguishing relationship model to predict the fire assessment value and fire severity level after the fire extinguishing agent is used. Input layer: in, Input data to the model, The amount of fire extinguishing agent used, is the fire assessment value, is the fire grade; Hidden layer: in, is the weighted input of the hidden layer, is the output of the hidden layer, is the weight matrix of the hidden layer, is the bias term of the hidden layer, is the activation function of the hidden layer; Output layer: in, is the weighted input of the output layer, Output data for the output model, is the weight matrix of the output layer, is the bias term of the output layer, is the activation function of the output layer.

[0045] Step 5: Obtain the remaining dosage of the fire extinguishing agent. Using the linear programming method, take the dosage of the fire extinguishing agent used in each electrical cabinet as the decision target, minimize the loss estimation coefficient of the power room as the objective function, and use the remaining dosage of the fire extinguishing agent and the maximum safe concentration as constraints to calculate the dosage used in each electrical cabinet.

[0046] Linear programming is a powerful mathematical tool that helps optimize the allocation of resources (i.e., fire extinguishing agents) within given constraints. In this case, linear programming can precisely allocate fire extinguishing agents based on the fire severity, estimated loss factor, and remaining extinguishing agent quantity for each electrical panel. This optimizes fire extinguishing agent usage, ensuring that each panel receives the appropriate amount of extinguishing agent. This ensures that limited fire extinguishing agent resources are used efficiently, avoiding waste or shortages. The appropriate amount of extinguishing agent is allocated to each panel to maximize fire extinguishing effectiveness and minimize losses. In practical applications, fire extinguishing agent usage must meet a series of constraints, such as a maximum safe concentration (to prevent excessive extinguishing agent from negatively impacting equipment or the environment) and the remaining amount of extinguishing agent (to prevent premature depletion of extinguishing agent). Linear programming can find the optimal solution within these constraints, minimizing the estimated loss factor while remaining within any constraints. This ensures that firefighting operations are both effective and safe, preventing the negative impacts on equipment and personnel caused by excessive extinguishing agent use and avoiding the problem of insufficient extinguishing agent at critical moments.

[0047] In this embodiment, the specific calculation method for minimizing the loss estimation coefficient of the power room to be the objective function is: in, is the objective function for minimizing the loss estimation coefficient of the power room, The dosage of the fire extinguishing agent output by the fire extinguishing relationship model After the fire level, For the The emergency factor assigned to each electrical cabinet, The dosage of the fire extinguishing agent output by the fire extinguishing relationship model Fire assessment value after The calculation method with the remaining dose of fire extinguishing agent and the maximum safe concentration as constraints is: in, For the The amount of fire extinguishing agent used in each electrical cabinet, is the remaining amount of fire extinguishing agent, is the extinguishing agent volume conversion coefficient, is the background concentration of fire extinguishing agent in the power room when in use, is the upper limit of safe concentration, is the volume of the power room.

[0048] The above formulas are all dimensionless and numerical calculations. The formulas are obtained by collecting a large amount of data and performing software simulation to obtain the most recent real situation. The preset parameters in the formulas are set by technicians in this field according to actual conditions.

[0049] The above embodiments can be implemented in whole or in part by software, hardware, firmware, or any other combination thereof. When implemented using software, the above embodiments can be implemented in whole or in part in the form of a computer program product. Those skilled in the art will appreciate that the units and algorithm steps of each example described in conjunction with the embodiments disclosed herein can be implemented by electronic hardware, or a combination of computer software and electronic hardware. Whether these functions are performed by hardware or software depends on the specific application and design constraints of the technical solution.

[0050] The units described as separate components may or may not be physically separate, and the components shown as units may or may not be physical units, and may be located in one place or distributed across multiple network units. Some or all of these units may be selected to achieve the purpose of this embodiment as needed.

[0051] The above is only a specific implementation method of the present application, but the scope of protection of the present application is not limited thereto. Any technician familiar with this technical field can easily think of changes or replacements within the technical scope disclosed in this application, which should be covered by the scope of protection of the present application.

Claims

1. A multi-sensor collaborative analysis method for electrical cabinet fire extinguishing, characterized in that: The specific steps include: Step 1: Number the electrical devices inside the electrical cabinet and determine the electrical connection relationships between each device. By acquiring an infrared image of the interior of the cabinet and using image coordinate calibration technology, the physical locations of the electrical devices are spatially aligned with the pixel locations in the infrared image. Based on the aligned position mapping, an infrared temperature feature recognition model is constructed to extract the infrared data of each device. Step 2: Based on the temperature data of the identified electrical equipment and the physical position and electrical connection relationships of the electrical equipment in the cabinet, calculate the temperature impact value between each electrical equipment and the electrical equipment with the same position and electrical connection relationships. Convert the position and electrical connection relationships into a heat conduction network. By integrating the temperature data of the electrical equipment, the heat conduction network, and the average temperature in the cabinet, calculate the spontaneous combustion assessment value of each electrical equipment to assess the spontaneous combustion risk of the electrical equipment; Step 3: Obtain environmental data within the electrical cabinet and determine the fire risk of the electrical cabinet by calculating the fire risk coefficient based on the spontaneous combustion assessment value and location distribution of the electrical equipment. Determine the deflagration probability of the electrical cabinet based on the combustible gas concentration, oxygen concentration, and arc probability. Calculate the fire assessment value based on the deflagration probability and fire risk coefficient. Step 4: Assign an emergency coefficient to each electrical cabinet. Based on the fire assessment value and fire severity, determine the loss estimation coefficient of the electrical room. Based on the historical fire extinguishing effect of fire extinguishing agents, construct a fire extinguishing relationship model between the fire extinguishing agent, the fire assessment value, and the fire severity. Step 5: Obtain the remaining dosage of fire extinguishing agent. Using the linear programming method, with the dosage of fire extinguishing agent used in each electrical cabinet as the decision target, the minimum loss estimation coefficient of the power room as the goal, and the remaining dosage of fire extinguishing agent and the maximum safe concentration as constraints, calculate the dosage used in each electrical cabinet.

2. The electrical cabinet fire extinguishing method using multi-sensor collaborative analysis according to claim 1, characterized in that: The infrared data of the electrical equipment includes the temperature of the electrical equipment, the arc generated by the electrical equipment, the fire level and the average temperature inside the cabinet; The specific method of establishing an infrared temperature recognition model based on the infrared image of the interior of the electrical cabinet and the position of the electrical equipment in the image is as follows: Obtain historical infrared images of electrical cabinets, mark the electrical equipment with the equipment number and infrared data according to their position in the image, and establish an infrared temperature recognition model based on a convolutional neural network. Use the historical infrared images as model input and the marked equipment number and infrared data of the electrical equipment as model output. Train the infrared temperature recognition model, and use real-time infrared images as model input to identify the equipment number and infrared data of the electrical equipment in the infrared image. The infrared temperature recognition model is based on a convolutional neural network, which specifically includes a convolution layer, an activation layer, a pooling layer and an output layer.

3. The electrical cabinet fire extinguishing method using multi-sensor collaborative analysis according to claim 1, characterized in that: The electrical connection relationship includes an upper end connection device and a lower end connection device, and the position relationship includes up, down, left, right, front, back and the distance between up, down, left, right, front and back; The specific method for calculating the temperature impact value of electrical equipment is: in, is the temperature impact value of the electrical equipment, For the The temperature of the electrical equipment with the electrical connection relationship, is the temperature of the electrical equipment, For the The influence coefficient of the electrical connection relationship, is the number of electrical connection relationships, For the The temperature of electrical equipment in different positions, For the The distance of the position relationship, For the The influence coefficient of the position relationship, is the number of positional relationships, , , , are all positive integers, .

4. The electrical cabinet fire extinguishing method using multi-sensor collaborative analysis according to claim 1, characterized in that: According to the temperature of the electrical equipment, the temperature impact value, and the average temperature in the cabinet, the spontaneous combustion assessment value is calculated. The method for assessing the risk of spontaneous combustion of electrical equipment is as follows: in, is the spontaneous combustion assessment value, is the temperature of the electrical equipment, is the temperature impact value of electrical equipment, is the average temperature inside the electrical cabinet, For the suitable working temperature of electrical equipment, is the upper limit of operating temperature; current When the device is in use, there is a risk of spontaneous combustion of electrical equipment; in, Evaluate thresholds for spontaneous combustion of electrical equipment.

5. The electrical cabinet fire extinguishing method using multi-sensor collaborative analysis according to claim 1, characterized in that: The environmental data include smoke concentration, combustible gas concentration, ambient temperature, and oxygen concentration; The specific method for determining the fire risk of an electrical cabinet by calculating the fire risk coefficient of the electrical cabinet is as follows: The process of obtaining the location distribution of electrical equipment is as follows: Obtain the dimensional data of the electrical cabinet, use the panel of the electrical cabinet as the coordinate plane, the bottom edge as the horizontal coordinate, the vertical edge as the vertical coordinate, the lower left corner as the coordinate origin, and the center point of the electrical equipment as the coordinate point. Based on the coordinate point and spontaneous combustion assessment value of each electrical equipment, calculate the spatial risk density using Gaussian kernel density: in, For the The risk density of each electrical equipment location, For the The spontaneous combustion assessment value of each electrical equipment, For the The coordinate position of each electrical device, is the risk diffusion range, is the number of electrical equipment in the electrical cabinet, is the width of the electrical cabinet, is the length of the electrical cabinet; Calculate the fire risk coefficient based on the spatial risk density: in, is the fire risk factor.

6. The electrical cabinet fire extinguishing method using multi-sensor collaborative analysis according to claim 5, characterized in that: The method for determining the deflagration probability is: in, is the probability of deflagration, is the calibration factor, is the combustible gas concentration, , are the lower and upper limits of combustible gas explosion concentrations, is the oxygen concentration, is the probability of arcing; The method for calculating the fire assessment value based on the probability of deflagration and the abnormal temperature rise coefficient is: in, is the fire assessment value, , are the weights of deflagration probability and abnormal temperature rise coefficient, + .

7. The electrical cabinet fire extinguishing method using multi-sensor collaborative analysis according to claim 1, characterized in that: The specific method for determining the loss estimation coefficient of the power room is: in, is the loss estimation coefficient of the power room, For fire level, For the The emergency factor allocated to each electrical cabinet.

8. The electrical cabinet fire extinguishing method using multi-sensor collaborative analysis according to claim 1, characterized in that: The fire extinguishing relationship model between the fire extinguishing agent, the fire assessment value, and the fire level is based on a BP neural network, which specifically includes an input layer, a hidden layer, and an output layer: The amount of fire extinguishing agent used, fire assessment value, and fire level are used as model input data, and the fire assessment value and fire level after using the fire extinguishing agent are used as model output data to train the fire extinguishing relationship model. The model input data is input into the trained fire extinguishing relationship model to predict the fire assessment value and fire level after using the fire extinguishing agent.

9. The electrical cabinet fire extinguishing method using multi-sensor collaborative analysis according to claim 1, characterized in that: The specific calculation method for minimizing the loss estimation coefficient of the power room to obtain the objective function is: in, is the objective function for minimizing the loss estimation coefficient of the power room, The dosage of the fire extinguishing agent output by the fire extinguishing relationship model After the fire level, For the The emergency factor assigned to each electrical cabinet, The dosage of the fire extinguishing agent output by the fire extinguishing relationship model Fire assessment value after The calculation method with the remaining dose of fire extinguishing agent and the maximum safe concentration as constraints is: in, For the The amount of fire extinguishing agent used in each electrical cabinet, is the remaining amount of fire extinguishing agent, is the extinguishing agent volume conversion coefficient, is the background concentration of fire extinguishing agent in the power room when in use, is the upper limit of safe concentration, is the volume of the power room.

Images