Powered cable fire sign identification and early warning analysis method and training system

Abstract

The present invention discloses a method and a training system for identifying and analyzing the signs of fire in energized cables; the method determines the occurrence and danger level of cable fires by establishing a cable fire warning model, thereby activating different fire extinguishing devices in a timely manner, and can also determine the time for firefighters to enter the scene according to the fire extinguishing process; the information collected by the device is used to train the cable fire warning model, which includes a mounting bracket, a power supply module, a data acquisition module, a cable laying module and an alarm and fire extinguishing module; the cable laying module is used to lay cables; the power supply module is configured to supply power to and overload cables; the data acquisition module is used to collect environmental data when the cables are burning; the alarm and fire extinguishing module is used to report the cable fire status and extinguish the fire; the present invention simultaneously monitors multiple indicators, and obtains learning samples by establishing a test device, thereby determining the type of energized cable fire in real time, effectively improving the accuracy of cable fire alarms and reducing the false alarm rate.

Description

Technical Field

[0001] The present invention relates to the technical field of cable fire safety, and in particular to a method and a training system for identifying and warning fire signs of energized cables. Background Art

[0002] With the continuous development and expansion of power systems, cables, as an important component of power transmission, play an irreplaceable role in modern society. However, the safety of energized cable systems has always been one of the focuses of attention in the power industry and related fields. Especially in terms of fire risk, since cables may be affected by various factors such as the external environment and electrical faults during use, cable fire accidents pose a serious threat to people's lives and property. Traditional fire prevention and control methods are often limited to passive response measures and are unable to predict and actively intervene in the occurrence of fires in advance. Therefore, the development of a system that can identify the signs of energized cable fires and conduct early warning analysis has important theoretical and practical significance. By accurately identifying and predicting the signs of energized cable fires, it will help relevant departments and personnel take timely measures to reduce the losses caused by fire accidents.

[0003] Through research, it was found that the Chinese patent with authorization announcement number CN111854855B discloses a method and device for online monitoring of cable fire characteristics in multiple scenarios. The invention relates to the field of automated intelligent monitoring technology, and in particular to a method for online monitoring of cable fire characteristics in multiple scenarios, including: monitoring a specified type of gas at a specified location in the measured scene at a set frequency; monitoring the temperature at a specified location in the measured scene at a set frequency; monitoring the smoke particles at a specified location in the measured scene at a set frequency, and comparing them with the set gas concentration, temperature and smoke particle concentration alarm thresholds. At the same time, when the gas monitoring results of any type at any location in the gas monitoring results, and / or the temperature monitoring results at any location in the temperature monitoring results; and / or the smoke particle monitoring results at any location in the smoke particle monitoring results reach the alarm threshold, an alarm is issued. At the same time, the present invention also requests protection for a device for online monitoring of cable fire characteristics in multiple scenarios. Through the above method and device, 24-hour uninterrupted remote real-time monitoring of cables can be achieved, thereby improving the cable fire early warning capability.

[0004] While this cable fire monitoring device has, to some extent, addressed the technical issue of cable fire warnings, it can experience a delay in changes in fire-related parameters when cables are installed in cable troughs, making early warnings ineffective. Furthermore, cable fires typically manifest as no fire or smoldering in their early stages, with small changes in parameters like temperature or smoke. This can lead to subtle changes in cable fire data, resulting in false or delayed alarms, posing a fire hazard and presenting certain limitations. Summary of the Invention

[0005] In a first aspect, the present application provides a method for identifying and analyzing fire signs of live cables, which adopts the following technical solutions:

[0006] A method for identifying and analyzing fire signs of energized cables, comprising the following steps:

[0007] S1, building a cable fire warning model;

[0008] S2, based on the training samples, trains the cable fire warning model constructed in S1 to obtain a mature cable fire warning model;

[0009] S3, using the cable fire warning model obtained in S2 to monitor and determine the fire situation of the energized cable in real time. If the fire situation in the carriage is no fire, continue monitoring; if the cable fire situation is open flame or smoldering, execute step S4;

[0010] S4, determine the burning speed of the cable, adopt different fire extinguishing procedures according to the difference in burning speed, and report the cable fire to the fire department;

[0011] S5, using a mature cable fire early warning model, determines whether the fire is extinguished. If so, step S6 is executed; if not, the process returns to S4.

[0012] S6, start the energized cable smoke exhaust system, and further execute step S7;

[0013] S7, real-time monitoring of environmental toxicity and determination of air quality until the air quality is deemed safe, at which point firefighters will conduct emergency repairs.

[0014] Furthermore, the component cable fire warning basic model in step S1 includes the following steps:

[0015] S11, define the cable fire warning model as a three-layer network topology, which includes an input layer, a hidden layer, and an output layer; and record the number of nodes in the input layer, hidden layer, and output layer as: n, q, m respectively; where a is a constant, which takes the value of 0 or 1;

[0016] S12, gives the output equations of the hidden layer and output layer neuron nodes, and initializes the weights. The initial connection weight coefficients are a set of random small non-zero values;

[0017] Hidden layer neuron node c k The output equation is:

[0018] Sigmoid function is used here

[0019] The neuron node y in the output layer j The output equation is:

[0020] Sigmoid function is used here

[0021] Where: x i represents i type of monitoring indicator; v ki Represents the connection weight coefficient between the input layer and the hidden layer; w jk Represents the connection weight coefficient between the hidden layer and the output layer.

[0022] Furthermore, the training of the cable fire warning model in step S2 includes the following steps:

[0023] S21, forward propagation process: input the training data into the current model, calculate the error between the output of the current model and the theoretical output, the error calculation formula is:

[0024]

[0025] Where, represents the expected output value of the p-th sample neuron j; represents the actual output value of the p-th sample neuron j; E p represents the error of the pth learning sample; E represents the global error;

[0026] S22, back propagation process: according to the calculated error E p , the connection weight coefficient in the correction model, the connection weight coefficient w between the hidden layer and the output layer jk The correction formula is:

[0027]

[0028] The connection weight coefficient v between the input layer and the hidden layer ki The correction formula is:

[0029]

[0030] Where η is the learning rate, and 0<η<1; S j =W j X is the net input value of the jth neuron; S k =W k X is the net input value of the kth neuron; after the correction is completed, return to step S21 until all training data are input and the training is completed. The obtained connection weight coefficient is the connection weight coefficient in the mature cable fire warning model.

[0031] Furthermore, in step S4, the cable burning speed is determined by the following model:

[0032]

[0033] Where: A = r c 2 ρ c c c (T c -T0)+(r p 2 -r c 2 )ρ p [c p (T p -T0)+ΔH]

[0034] B=2πr c h fc (T f -T c )

[0035] C=2πr p h fp (T f -T p )

[0036] T f is the flame center temperature; this value represents the coupling effect of the flame on the thermal convection and thermal radiation of the cable core and outer sheath respectively. fc and χ fp are the heat transfer coefficients of the core and sheath, which represent the heat loss during the flame heat transfer process to the cable core and sheath respectively. ΔH is the latent heat of thermal decomposition of the cable sheath; T p 、T ∞ are the pyrolysis temperature of polyethylene insulation layer and the ambient temperature respectively; r c ,ρ c ,c c are the core radius, core density and core constant pressure heat capacity respectively. p , p c c p h is the sheath radius, core density and core constant pressure heat capacity. fc and h fp is the comprehensive heat transfer coefficient of the core and sheath, which represents the coupling effect of flame on the thermal convection and thermal radiation of the cable core and sheath respectively;

[0037] L f The length of the cable along the line wrapped by flame is defined as the flame wrapped length, and the length of the outer sheath pyrolysis area along the line is defined as the pyrolysis length L p , L pr L is the length of the outer sheath preheating zone along the line direction, pis the projected length of the flame on the outer sheath;

[0038] When V fh If it does not grow rapidly in a short period of time and basically shows a stable trend, it means that the fire is slow and the fire can be extinguished by using a fire extinguisher after power is cut off. fh When the fire grows rapidly in a short period of time, it means that the fire is fierce. At this time, cut off the power supply first and use water injection and suffocation to extinguish the fire.

[0039] Furthermore, the air quality is determined in step S7 in the following manner:

[0040] definition:

[0041]

[0042] Where MAX[CO] is the maximum CO value obtained by measurement; LC 50,CO The value is 5700ppm. When [CO2]<5%, the m value is -18 and the b value is 122000; [CO] is the measured carbon monoxide concentration; [CO2] is the measured carbon dioxide concentration.

[0043] When FED1≥1, or FED2≥1 or FED3≥1, it means that the air quality does not meet the standard; otherwise it means that the air quality meets the standard.

[0044] On the other hand, the present application also provides a live cable fire sign identification and early warning training system, which adopts the following technical solutions:

[0045] A fire sign identification and early warning training system for energized cables, comprising a mounting bracket, a power supply module, a data acquisition module, a cable laying module, and an alarm and fire extinguishing module;

[0046] The cable laying module is fixedly connected to the mounting bracket and is used for laying cables;

[0047] The power supply modules are provided on both sides of the mounting bracket for supplying power to and overloading the cables;

[0048] The data acquisition module is used to collect environmental data when the cable is burning; the alarm and fire extinguishing module is set next to the mounting bracket, and is used to report the fire status of the cable and extinguish the fire;

[0049] The cable laying module includes multiple supporting vertical bars and multiple supporting cross bars. The multiple supporting vertical bars are fixedly connected to both sides of the mounting bracket at intervals. The supporting cross bars are fixedly connected between two opposite supporting vertical bars, and the cables are fixedly connected to the supporting cross bars by tying wires.

[0050] Furthermore, a cable trough bottom plate is provided on the supporting cross bar, and a cable trough top plate and two cable trough side plates are spliced ​​on the cable trough bottom plate to form a frame; the cable passes through the frame, and a middle cover plate is provided between the cable trough top plate and the cable trough bottom plate, and a plurality of heat radiation plates are provided on the cable trough top plate, and a plurality of empty slots matching the heat radiation plates are opened on the middle cover plate.

[0051] Furthermore, the data acquisition module includes a pulley bracket, which is slidably connected to a mounting rod in the vertical direction, and the mounting rod is also provided with a fixing bolt for limiting its own sliding; a plurality of thermocouple rods for measuring temperature are arranged at intervals in the horizontal direction on the mounting rod, and one end of the thermocouple rod is inserted into the frame and fixedly connected to a flue gas composition collection probe.

[0052] Furthermore, the data acquisition module also includes a droplet receiving plate, a mass collector and a camera. The droplet receiving plate is placed under the cable, and the mass collector is used to collect the overall mass of the droplet receiving plate; the camera is placed next to the cable laying device to collect brightness information and image information.

[0053] Furthermore, the alarm and fire extinguishing module includes a fire extinguisher, a sprinkler, a fire extinguishing rack and a smoke and exhaust hood; the fire extinguishing rack is arranged next to the cable laying device, the fire extinguisher and the sprinkler are arranged on the fire extinguishing rack, and the smoke and exhaust hood is arranged at the upper end of the cable laying device.

[0054] In summary, this application has the following beneficial technical effects:

[0055] Compared with the prior art, the present invention has the following beneficial effects:

[0056] Fire alarm detectors generally have the risk of high false alarm rates. The proposed live cable fire detection algorithm incorporates a deep learning algorithm, simultaneously monitoring indicators such as gas composition, image information, smoke concentration, temperature distribution, and brightness changes. By building a test model to obtain learning samples, it can then determine the type of live cable fire in real time, including no fire, smoldering, and open flames. This algorithm can effectively improve the accuracy of cable fire alarms and reduce false alarm rates.

[0057] 2. This application focuses on the changes in CO, CO2 and O2 concentrations during the combustion process of bent cables, proposes a fire smoke toxicity assessment method suitable for cables, makes real-time judgments on the toxicity of cable fire smoke, and divides safety standards, thereby effectively reducing the risks faced by firefighters during emergency repairs and rescue operations in fire accidents.

[0058] 3. The present invention constructs an energized cable fire test device, which can study the impact of different cable numbers, types (size, type), and whether they are restricted (cable troughs) on the cable fire behavior; in addition, a heat radiation plate is set up to further study the flame behavior of cables after radiation aging and obtain relevant experimental parameters, thereby providing important parameter support for fire early warning.

[0059] 4. By changing these parameters, more accurate early warning results can be obtained, further classifying cable fire, smoldering, and no-fire states. Furthermore, the cable's operating status can be determined in real time, revealing the cable's lifecycle and providing guidance for cable selection. Furthermore, the present invention provides a fire spread velocity characterization model for energized cables, providing guidance on the scope of post-disaster hazards posed by energized cables. BRIEF DESCRIPTION OF THE DRAWINGS

[0060] Figure 1 This is an overall flow chart of a method for identifying and analyzing fire signs of energized cables in this application;

[0061] Figure 2 It is an overall structural diagram of a fire sign identification and early warning training system for energized cables;

[0062] Figure 3 yes Figure 2 The overall structure diagram of the power supply module;

[0063] Figure 4 yes Figure 2 Partial structural diagram of the cable laying module;

[0064] Figure 5 yes Figure 2 Partial structure diagram of the data acquisition module;

[0065] Figure 6 yes Figure 2 Partial structure diagram of the data acquisition module.

[0066] Description of reference numerals:

[0067] 1. Mounting bracket; 2. Power supply module; 21. Power supply box; 22. Circuit breaker; 23. Cable connection sheath; 3. Data acquisition module; 31. Pulley bracket; 32. Mounting rod; 33. Fixing bolt; 34. Thermocouple rod; 35. Smoke composition collection probe; 36. Droplet receiving plate; 37. Mass collector; 38. Camera; 4. Cable laying module; 41. Support vertical rod; 42. Support horizontal rod; 43. Tie wire; 44. Cable trough bottom plate; 45. Cable trough side plate; 46. Cable trough top plate; 47. Middle cover plate; 471. Empty trough; 48. Heat radiation plate; 5. Alarm and fire extinguishing module; 51. Fire extinguisher; 52. Sprinkler; 53. Fire extinguishing rack; 54. Smoke collection and exhaust hood. DETAILED DESCRIPTION

[0068] The following is combined with Figure 1-6 This application is described in further detail.

[0069] The present application discloses a method for identifying and analyzing fire signs of energized cables, including the following steps:

[0070] S1, building a basic model for cable fire warning;

[0071] S11, define the cable fire warning model as a three-layer network topology, which includes an input layer, a hidden layer, and an output layer; and record the number of nodes in the input layer, hidden layer, and output layer as: n, q, m respectively; where a is a constant, which takes the value of 0 or 1;

[0072] S12, gives the output equations of the hidden layer and output layer neuron nodes, and initializes the weights. The initial connection weight coefficients are a set of random small non-zero values;

[0073] Hidden layer neuron node c k The output equation is:

[0074] Sigmoid function is used here

[0075] The neuron node y in the output layer j The output equation is:

[0076] Sigmoid function is used here

[0077] Where: x i represents i type of monitoring indicator; v ki Represents the connection weight coefficient between the input layer and the hidden layer; w jk Represents the connection weight coefficient between the hidden layer and the output layer.

[0078] S2, obtain training samples through experiments, and train the cable fire warning model constructed in S1 based on the training samples to obtain a mature cable fire warning model;

[0079] S21, forward propagation process: input the training data into the current model, calculate the error between the output of the current model and the theoretical output, the error calculation formula is:

[0080]

[0081] Where, represents the expected output value of the p-th sample neuron j; represents the actual output value of the p-th sample neuron j; E prepresents the error of the pth learning sample; E represents the global error;

[0082] S22, back propagation process: according to the calculated error E p , the connection weight coefficient in the correction model, the connection weight coefficient w between the hidden layer and the output layer jk The correction formula is:

[0083]

[0084] The connection weight coefficient v between the input layer and the hidden layer ki The correction formula is:

[0085]

[0086] Where η is the learning rate, and 0<η<1; S j =W j X is the net input value of the jth neuron; S k =W k X is the net input value of the kth neuron; after the correction is completed, return to step S21 until all training data are input and the training is completed. The obtained connection weight coefficient is the connection weight coefficient in the mature cable fire warning model.

[0087] S3, using the cable fire warning model obtained in S2 to monitor and determine the fire situation of the energized cable in real time. If the fire situation in the carriage is no fire, continue monitoring; if the cable fire situation is open flame or smoldering, execute step S4;

[0088] S4, uses the model to determine the burning speed of the cable, adopts different fire extinguishing procedures according to the difference in burning speed, and reports the cable fire to the fire department;

[0089] The specific model is:

[0090] Where: A = r c 2 ρ c c c (T c -T0)+(r p 2 -r c 2 )ρ p [c p (T p -T0)+ΔH]

[0091] B=2πr c h fc (T f -T c )

[0092] C=2πr p h fp (T f -T p )

[0093] T f is the flame center temperature; this value represents the coupling effect of the flame on the thermal convection and thermal radiation of the cable core and outer sheath respectively. fc and χ fp are the heat transfer coefficients of the core and sheath, which represent the heat loss during the flame heat transfer process to the cable core and sheath respectively. ΔH is the latent heat of thermal decomposition of the cable sheath; T p 、T ∞ are the pyrolysis temperature of polyethylene insulation layer and the ambient temperature respectively; r c , ρ c , c c are the core radius, core density and core constant pressure heat capacity respectively. p , p c c p h is the sheath radius, core density and core constant pressure heat capacity. fc and h fp is the comprehensive heat transfer coefficient of the core and sheath, which represents the coupling effect of flame on the thermal convection and thermal radiation of the cable core and sheath respectively;

[0094] L f The length of the cable along the line wrapped by flame is defined as the flame wrapped length, and the length of the outer sheath pyrolysis area along the line is defined as the pyrolysis length L p , L pr L is the length of the outer sheath preheating zone along the line direction, p is the projected length of the flame on the outer sheath;

[0095] When V fh If it does not grow rapidly in a short period of time and basically shows a stable trend, it means that the fire is slow and the fire can be extinguished by using a fire extinguisher after power is cut off. fh If the fire grows rapidly in a short period of time, it means that the fire is fierce. At this time, cut off the power supply first and use water injection and suffocation to extinguish the fire.

[0096] S5, using a mature cable fire early warning model, determines whether the fire is extinguished. If so, step S6 is executed; if not, the process returns to S4.

[0097] S6, start the energized cable smoke exhaust system, and further execute step S7;

[0098] S7, real-time monitoring of environmental toxicity and determination of air quality until the air quality is deemed safe, at which point firefighters will conduct emergency repairs.

[0099] definition:

[0100]

[0101] Where MAX[CO] is the maximum CO value obtained by measurement; LC 50,CO The value is 5700ppm. When [CO2]<5%, the m value is -18 and the b value is 122000; [CO] is the measured carbon monoxide concentration; [CO2] is the measured carbon dioxide concentration.

[0102] When FED1≥1, or FED2≥1 or FED3≥1, it means that the air quality does not meet the standard; otherwise it means that the air quality meets the standard.

[0103] In order to obtain the training samples mentioned in the above step S2, the present application also provides a power cable fire sign identification and warning test system; the system includes a mounting bracket 1, a power supply module 2, a data acquisition module 3, a cable laying module 4 and an alarm and fire extinguishing module 5.

[0104] The cable laying module 4 includes multiple support vertical rods 41 and multiple support cross rods 42. The multiple support vertical rods 41 are fixedly connected on both sides of the mounting bracket 1 at intervals. The support cross rod 42 is fixedly connected between the two opposite support vertical rods 41. The cable is fixedly connected to the support cross rod 42 through a wire tie 43. A cable trough bottom plate 44 is provided on the support cross rod 42. The cable trough bottom plate 44 is spliced ​​with a cable trough top plate 46 and two cable trough side plates 45 to form a frame. The cable passes through the frame, and multiple card slots are arranged at intervals on the cable trough bottom plate 44. The cable trough side plates 45 are connected to the cable trough bottom plate 44 through the card slots. Therefore, selecting different card slots can change the width of the frame, so that it is suitable for cables of different sizes.

[0105] A middle cover plate 47 is provided between the cable trough top plate 46 and the cable trough bottom plate 44, a plurality of heat radiation plates 48 are provided on the cable trough top plate 46, and a plurality of empty slots 471 matching the heat radiation plates 48 are opened on the middle cover plate 47; the heat radiation plates 48 are mainly used to simulate the heating effect of heat radiation from the external environment, so as to be suitable for energized cable fire simulation in various scenarios; it should be noted that the cable trough is an experimental option, and the sample data obtained subsequently needs to indicate whether the cable is burning in the cable trough; thereby more comprehensively training the cable fire warning model in the aforementioned method.

[0106] The power supply module 2 includes two power supply boxes 21, with cables installed between them. A circuit breaker 22 and a cable connection sheath 23 are also installed between the cables and the power supply box 21. The circuit breaker 22 protects the cables from fires by automatically tripping when the voltage exceeds a certain threshold, which can be set based on the cable type. The cable connection sheath 23 is used to power the cables. It should be noted that different sizes and quantities of cable connection sheaths 23 can be installed, depending on the cable type and number of cables, to study the cable fire behavior in different installation scenarios.

[0107] The data acquisition module 3 is provided with multiple types, which are used to collect environmental data when the cable is burning, and the environmental data includes: gas composition, image information, smoke concentration, temperature distribution, and brightness change information; the data acquisition module 3 includes a pulley bracket 31, the pulley bracket 31 is slidably connected to a mounting rod 32 in the vertical direction, and the mounting rod 32 is also provided with a fixing bolt 33 for limiting its own sliding; a plurality of thermocouple rods 34 for measuring temperature are arranged at intervals in the horizontal direction on the mounting rod 32, one end of the thermocouple rod 34 is inserted into the frame and fixedly connected to a smoke composition collection probe 35; the electric thermocouple rod is used to collect temperature distribution information, and the smoke composition collection probe 35 is used to collect gas composition information.

[0108] The data acquisition module 3 also includes a droplet receiving plate 36, a quality collector 37 and a camera 38. The droplet receiving plate 36 is placed under the cable laying module 4. The quality collector 37 is used to collect the overall quality of the droplet receiving plate 36. When the cable starts to burn, it can be determined whether the cable is in the cable trough based on whether the data of the quality collector 37 changes. The camera 38 is placed next to the cable laying device to collect brightness information and image information. It is also responsible for collecting L f , L p , L pr data.

[0109] The alarm and fire extinguishing module 5 includes a fire extinguisher 51, a sprinkler 52, a fire extinguishing rack 53 and a smoke exhaust hood; the fire extinguishing rack 53 is arranged next to the cable laying device, the fire extinguisher 51 and the sprinkler 52 are arranged on the fire extinguishing rack 53, and the smoke exhaust hood is arranged at the upper end of the cable laying device; the smoke exhaust hood can collect smoke concentration information and at the same time complete the function of exhaust and fire extinguishing.

[0110] The training system can repeat the test many times. During the test, the objective environment of cable combustion can be changed to obtain different combustion environment data. In the subsequent training of the cable fire warning model, the environmental data obtained from the repeated tests will be input into the basic model to obtain a model that can accurately judge cable fires, which is conducive to the prevention and treatment of cable fires.

[0111] The above are all preferred embodiments of the present application, and are not intended to limit the scope of protection of the present application. Therefore, any equivalent changes made based on the structure, shape, and principle of the present application should be included in the scope of protection of the present application.

Claims

1. A method for identifying and analyzing fire signs of energized cables, characterized in that The steps include: S1, building a basic model for cable fire warning; S2, obtain training samples through experiments, train the cable fire warning model constructed in S1 based on the training samples, and further construct the cable fire warning model; S3, using the cable fire warning model obtained in S2 to monitor and determine the fire situation of the energized cable in real time. If the fire situation in the carriage is no fire, continue monitoring; if the cable fire situation is open flame or smoldering, execute step S4; S4, determine the burning speed of the cable, adopt different fire extinguishing procedures according to the difference in burning speed, and report the cable fire to the fire department; S5, using the established cable fire warning model, determine whether the fire is extinguished. If so, proceed to step S6; if not, return to S4; S6, start the energized cable smoke exhaust system, and further execute step S7; S7, real-time monitoring of environmental toxicity and assessment of air quality. Firefighters will then conduct emergency repairs until the air quality is deemed safe. In step S4, the burning speed of the cable is determined by the following model: Where: A = r c 2 ρ c c c (T c - T0) + (r p 2 - r c 2 )ρ p [c p (T p - T0) + ΔH] B=2πr c h fc (T f -T c ) C=2πr p h fp (T f -T p ) In addition, T f is the flame center temperature; it represents the coupling effect of flame on the thermal convection and thermal radiation of cable core and outer sheath respectively; X fc With X fp are the heat transfer coefficients of the core and sheath, representing the heat loss during the flame heat transfer process to the cable core and sheath respectively; ΔH is the latent heat of thermal decomposition of the cable sheath; T p and T0 are the pyrolysis temperature of polyethylene insulation layer and the ambient temperature respectively; r c ,ρ c ,c c are the core radius, core density and core constant pressure heat capacity respectively; r p , p c , c p h is the sheath radius, core density and core constant pressure heat capacity; fc and h fp is the comprehensive heat transfer coefficient of the core and sheath; L f The length of the cable wrapped by flame along the line is defined as the flame wrapped length, and the length of the outer sheath pyrolysis area along the line is defined as the pyrolysis length L p , L pr L is the length of the outer sheath preheating zone along the line direction, p is the projected length of the flame on the outer sheath; When V fh If it does not grow rapidly in a short period of time and basically shows a stable trend, it means that the fire is slow and the fire can be extinguished by using a fire extinguisher after power is cut off. fh When the fire grows rapidly in a short period of time, it means that the fire is fierce. At this time, cut off the power supply first and use water injection and suffocation to extinguish the fire.

2. A method for identifying and analyzing fire signs of live cables according to claim 1, characterized in that The construction of the cable fire early warning basic model in step S1 includes the following steps: S11, define the cable fire warning model as a three-layer network topology structure, which includes an input layer, a hidden layer, and an output layer; and record the number of nodes in the input layer, hidden layer, and output layer as: n, q, m respectively; in a is a constant, which can be 0 or 1; S12, gives the output equations of the hidden layer and output layer neuron nodes, and initializes the weights. The initial connection weight coefficients are a set of random small non-zero values; Hidden layer neuron node c k The output equation is: Sigmoid function is used here The neuron node y in the output layer j The output equation is: Sigmoid function is used here Where: x i represents i type of monitoring indicator; v ki Represents the connection weight coefficient between the input layer and the hidden layer; w jk Represents the connection weight coefficient between the hidden layer and the output layer.

3. A method for identifying and analyzing fire signs of live cables according to claim 2, characterized in that The training of the cable fire warning model in step S2 includes the following steps: S21, forward propagation process: input the training data into the current model, calculate the error between the output of the current model and the theoretical output, the error calculation formula is: Where, represents the expected output value of the p-th sample neuron j; Represents the actual output value of the p-th sample neuron j; E p represents the error of the pth learning sample; E represents the global error; S22, back propagation process: according to the calculated error E p , the connection weight coefficient in the correction model, the connection weight coefficient w between the hidden layer and the output layer jk The correction formula is: The connection weight coefficient v between the input layer and the hidden layer ki The correction formula is: Where η is the learning rate, and 0<η<1; S j =W j X is the net input value of the jth neuron; S k =W k X is the net input value of the kth neuron; after the correction is completed, return to step S21 until all training data are input and the training is completed. The obtained connection weight coefficient is the connection weight coefficient in the mature cable fire warning model.

4. A method for identifying and analyzing fire signs of live cables according to claim 3, characterized in that In step S7, the air quality is determined in the following manner: definition: Where MAX[CO] is the maximum CO value obtained by measurement; LC 50,CO The value is 5700ppm. When [CO2]<5%, the m value is -18 and the b value is 122000. [CO] is the measured carbon monoxide concentration; [CO2] is the measured carbon dioxide concentration. When FED1≥1, FED2≥1 or FED3≥1, it means that the air quality does not meet the standard; otherwise, it means that the air quality meets the standard.

5. A training system for identifying and analyzing the signs of fire in live cables, used for training the method for identifying and analyzing the signs of fire in live cables according to claim 4, characterized in that: It comprises a mounting bracket (1), a power supply module (2), a data acquisition module (3), a cable laying module (4) and an alarm and fire extinguishing module (5); The cable laying module (4) is fixedly connected to the mounting bracket (1) and is used for laying cables; The power supply module (2) is arranged on both sides of the mounting bracket (1) and is used to supply power to and overload the cable; The data acquisition module (3) is used to collect environmental data when the cable is burning; the alarm and fire extinguishing module is arranged beside the mounting bracket (1) and is used to report the fire status of the cable and extinguish the fire; The cable laying module (4) comprises a plurality of support vertical bars (41) and a plurality of support cross bars (42), wherein the plurality of support vertical bars (41) are fixedly connected to both sides of the mounting bracket (1) at intervals, and the support cross bars (42) are fixedly connected between two opposing support vertical bars (41), and the cables are fixedly connected to the support cross bars (42) by means of tying wires (43).

6. The energized cable fire sign identification and early warning analysis training system according to claim 5, characterized in that: The supporting cross bar (42) is provided with a cable trough bottom plate (44), and the cable trough bottom plate (44) is spliced ​​with a cable trough top plate (46) and two cable trough side plates (45) to form a frame; the cable passes through the frame, and a middle cover plate (47) is provided between the cable trough top plate (46) and the cable trough bottom plate (44), and a plurality of heat radiation plates (48) are provided on the cable trough top plate (46), and a plurality of empty slots (471) matching the heat radiation plates (48) are opened on the middle cover plate (47).

7. The energized cable fire sign identification and early warning analysis training system according to claim 6, characterized in that: The data acquisition module (3) comprises a pulley bracket (31), the pulley bracket (31) being slidably connected to a mounting rod (32) in a vertical direction, the mounting rod (32) being further provided with a fixing bolt (33) for limiting its own sliding; a plurality of thermocouple rods (34) for measuring temperature are arranged at intervals in a horizontal direction on the mounting rod (32), one end of the thermocouple rod (34) being inserted into the frame and fixedly connected to a smoke component collection probe (35).

8. The energized cable fire sign identification and early warning analysis training system according to claim 7, characterized in that: The data acquisition module (3) further comprises a droplet receiving plate (36), a mass acquisition instrument (37) and a camera (38), wherein the droplet receiving plate (36) is placed below the cable, and the mass acquisition instrument (37) is used to collect the overall mass of the droplet receiving plate (36); and the camera (38) is placed beside the cable laying module (4) and is used to collect brightness information and image information.

9. The energized cable fire sign identification and early warning analysis training system according to claim 8, characterized in that: The alarm and fire extinguishing module (5) comprises a fire extinguisher (51), a sprinkler (52), a fire extinguishing rack (53) and a smoke collecting and exhausting hood; the fire extinguishing rack (53) is arranged beside the cable laying module (4), the fire extinguisher (51) and the sprinkler (52) are arranged on the fire extinguishing rack (53), and the smoke collecting and exhausting hood is arranged at the upper end of the cable laying module (4).

Images