Multi-information fusion locomotive fire early warning and prevention and control system

Abstract

The invention discloses a multi-information fusion locomotive fire early warning and prevention and control system, and the system comprises the steps: determining whether a detection region is abnormal or not through a method for collecting the detection information of a thermal imaging detector at low frequency; when abnormity exists, visible light images / visible light videos are obtained every high-frequency time interval, whether abnormity exists in the detection area or not is determined for the second time, and false alarm of fire behavior in the detection area is prevented; and according to the fire related fault information and the fire spreading trend, the danger degree of the fire is determined so as to execute fire extinguishing operation on the detection area. According to the method, the problem of fire failure due to unreasonable threshold setting can be effectively prevented, the risk of fire occurrence is effectively reduced, and the loss is reduced. Meanwhile, after the existence of the abnormity is determined, the probability of false alarm can be reduced by combining the visible light image / visible light video acquired at each high-frequency time interval with the fault information related to the fire.

Description

Technical Field

[0001] The present invention relates to the technical field of fire warning, and in particular to a multi-information fusion locomotive fire warning and prevention system. Background Art

[0002] When the locomotive is in operation, the pantograph needs to be used to draw power to the locomotive battery and provide kinetic energy for the on-board equipment. For electronic equipment in long-term operation, due to the high power and current value, failure or abnormality of electronic components, overheating due to damage to cables and components, and poor heat dissipation environment of many electronic components are all causes of locomotive fires. Existing fire prevention and control systems mostly monitor temperature, smoke concentration and other information in real time through smoke detectors, temperature detectors, etc. When the fire alarm conditions are met, the feedback is fed back to the display unit for the driver and crew to judge whether to perform related operations manually or automatically. This traditional fire extinguishing method has a certain lag and false alarm probability due to the limited amount of information collected. Summary of the invention

[0003] The present invention discloses a multi-information fusion locomotive fire early warning and prevention and control system to overcome the above technical problems.

[0004] In order to achieve the above object, the technical solution of the present invention is:

[0005] A multi-information fusion locomotive fire warning and prevention system, including: a control unit, a detector group, a TCMS system interface unit, and a linkage prevention and control actuator;

[0006] The detector group is used to obtain detection information corresponding to the detectors in the detector group in the detection area; the detectors in the detector group include a plurality of thermal imaging detectors and a plurality of physical detectors;

[0007] The physical detectors include: smoke detectors, CO concentration detectors, temperature detectors, and temperature sensing cables;

[0008] The detection information of the thermal imaging detector includes: thermal imaging images, thermal imaging videos, visible light images, and visible light videos;

[0009] The TCMS system interface unit is used to obtain the fault information of the locomotive, so as to obtain the fire-related fault information according to the fault information of the locomotive;

[0010] The control unit is used to determine whether there is an abnormality in the detection area according to the detection information of the physical detector by low-frequency acquisition of the detection information of the thermal imaging detector, so as to prevent underreporting when a fire has occurred in the detection area; when there is an abnormality in the detection area or the detection information of the physical detector is greater than or equal to the set warning threshold data, a visible light image / visible light video is acquired at each high-frequency time interval to determine for the second time whether there is an abnormality in the detection area, so as to prevent false alarms of the fire in the detection area; and when it is determined for the second time that there is an abnormality in the detection area, the location of the fire is determined according to the fire-related fault information, and then the degree of danger of the fire is determined according to the spread trend of the fire;

[0011] The linkage prevention and control execution mechanism is used to execute fire extinguishing operations on the detection area according to the danger level of the fire, fire-related fault information and the location where the fire occurred.

[0012] Furthermore, the fire-related fault information includes: fault name, name of the component to which the fault belongs, fault code, fault description information, and fault handling suggestions;

[0013] The method for obtaining fire-related fault information is as follows: based on the fire characteristics of the materials used in the locomotive, the power consumption of the equipment, and the difficulty of loss, combined with the fault information of the locomotive, the fault information associated with the fire is screened; based on the electrical schematic diagram, the position and state of the collected variables in the fault entry are extracted, and the abnormal point of the equipment is located in combination with the fault description and related logic;

[0014] After determining whether there is an abnormality in the detection area, it also includes determining the location where the abnormality occurs based on the fire-related fault information of the locomotive.

[0015] Furthermore, the method for determining whether there is an abnormality in the detection area is as follows:

[0016] When the detection data of the physical detector is less than the set warning threshold data:

[0017] If the detection data of the physical detector is less than the first warning data, a thermal imaging image / thermal imaging video is acquired at every first time interval to determine whether there is an abnormality in the detection area;

[0018] If the detection data of the physical detector is greater than or equal to the nth warning data and less than the n+1th warning data, a thermal imaging image / thermal imaging video is acquired at every n+1th time interval to determine whether there is an abnormality in the detection area; n is a natural number, n=1,…,n,n+1,…,N; N represents the total number of low-frequency time intervals; wherein the Nth warning data is the set warning threshold data;

[0019] When the detection data of the physical detector is greater than or equal to the set warning threshold data, there is an abnormality in the detection area;

[0020] The first time interval>...>nth time interval>n+1th time interval>...>Nth time interval>high-frequency time interval;

[0021] 1st warning data<...<nth warning data<n+1th warning data<...<Nth warning data.

[0022] Furthermore, according to the spread trend of the fire, the method for determining the degree of fire danger is:

[0023] According to the thermal imaging detector / temperature sensing detector / temperature sensing cable, the temperature increase rate of the detector area is obtained:

[0024] If the temperature increase rate per unit time is greater than or equal to the first temperature rise threshold, the fire danger level is a high danger level;

[0025] If the temperature increase rate per unit time is less than the first temperature rise threshold and greater than or equal to the second temperature rise threshold, the fire danger level is a medium danger level;

[0026] If the temperature increase rate per unit time is less than the second temperature rise threshold, the fire danger level is a low danger level;

[0027] Among them, the first temperature rise threshold>the second temperature rise threshold>the third temperature rise threshold.

[0028] Furthermore, according to the spread trend of the fire, the method for determining the degree of fire danger is:

[0029] According to the thermal imaging detector, the rate of increase of the fire area in the detector area is obtained:

[0030] If the rate of increase of the fire area per unit time is greater than or equal to the first area change threshold, the fire danger level is a high danger level;

[0031] If the rate of increase of the fire area per unit time is less than the first area change threshold and greater than or equal to the second area change threshold, the fire danger level is a medium danger level;

[0032] If the rate of increase of the fire area per unit time is less than the second area change threshold, the fire danger level is a low danger level;

[0033] Among them, the first area change threshold>the second area change threshold>the third area change threshold.

[0034] Furthermore, the method for performing the fire extinguishing operation on the detection area is as follows:

[0035] When the fire danger level is low, start the nozzle corresponding to the location where the fire occurs and spray a single volume of fire extinguishing agent to extinguish the fire;

[0036] When the fire is at a medium level, the nozzles corresponding to the location where the fire is generated are activated to spray double the volume of fire extinguishing agent to extinguish the fire;

[0037] When the fire danger level is high, start all the sprinklers and spray three times the capacity of fire extinguishing agent to extinguish the fire.

[0038] Furthermore, before performing the fire extinguishing operation on the detection area, it also includes:

[0039] According to the detection information of the thermal imaging detector, the abnormal status of the faulty components in the detection area is obtained: if it is possible to determine that there is an arc or spark at the location where the fire occurred based on the optical image / visible light video, or if it is possible to determine that there is a short circuit or phase loss at the location where the fire occurred through the fire-related fault information, the source of the fire can be located and the driver can be requested to cut off the power.

[0040] Further, it also includes a LKJ interface unit;

[0041] The LKJ interface unit is used to obtain the current running position information of the locomotive; and according to the current running position information of the locomotive, determine the distance between the locomotive and the previous station and the next station, and then be able to know the distance between the current locomotive position and the in and out platform; to determine whether it is necessary to start the emergency processing procedure;

[0042] Furthermore, by performing image segmentation on thermal imaging images and visible light images, the location of the fire can be determined to guide the fire-fighting operations of the joint prevention and control executive agencies.

[0043] Furthermore, the detector group also includes one or two of an atmospheric pressure sensor and an air flow velocity sensor, and the atmospheric pressure sensor and the air flow velocity sensor are used to obtain the gas flow velocity in a confined space, so as to evaluate the impact of the spread of flames when a fire has occurred, so as to guide the fire extinguishing operation of the linkage prevention and control actuator.

[0044] Beneficial effects: A multi-information fusion locomotive fire warning and prevention and control system of the present invention can determine whether there is an abnormality in the detection area according to the detection information of the physical detector by low-frequency acquisition of the detection information of the thermal imaging detector, so as to prevent missed reports when a fire has occurred in the detection area, and avoid the traditional fire detection method. In the case of an abnormality in the detection area or when the detection information of the physical detector is greater than or equal to the set warning threshold data, a visible light image / visible light video is obtained at each high-frequency time interval to determine whether there is an abnormality in the detection area for the second time to prevent false reports of the fire in the detection area; and when it is determined for the second time whether there is an abnormality in the detection area, the location of the fire is determined according to the fire-related fault information, and then the degree of danger of the fire is determined according to the spread trend of the fire; and the fire extinguishing operation of the detection area is performed through the linkage prevention and control execution mechanism. The present invention can effectively reduce the risk of fire and reduce losses. At the same time, after determining that there is an abnormality, the fire can be located by combining the visible light image / visible light video obtained at each high-frequency time interval with the fire-related fault information, which can reduce the probability of false alarms. BRIEF DESCRIPTION OF THE DRAWINGS

[0045] In order to more clearly illustrate the embodiments of the present invention or the technical solutions in the prior art, the following briefly introduces the drawings required for use in the embodiments or the description of the prior art. Obviously, the drawings described below are some embodiments of the present invention. For ordinary technicians in this field, other drawings can be obtained based on these drawings without paying creative labor.

[0046] Figure 1 It is a composition diagram of the locomotive fire warning and prevention linkage system of the present invention;

[0047] Figure 2 A flowchart of determining whether a detection area has an abnormality in an embodiment of the present invention;

[0048] Figure 3 It is a software flow chart of the main program of the locomotive fire warning and prevention and control controller in the embodiment of the present invention;

[0049] Figure 4 The figure is a flow chart of thermal imaging software of a locomotive fire warning and prevention controller in an embodiment of the present invention. DETAILED DESCRIPTION

[0050] In order to make the purpose, technical solution and advantages of the embodiments of the present invention clearer, the technical solution in the embodiments of the present invention will be clearly and completely described below in conjunction with the drawings in the embodiments of the present invention. Obviously, the described embodiments are part of the embodiments of the present invention, not all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by ordinary technicians in this field without creative work are within the scope of protection of the present invention.

[0051] This embodiment introduces a multi-information fusion locomotive fire warning and prevention system. Figure 1-4 As shown, it includes a control unit, a detector group composed of various environmental sensing devices, a TCMS system interface unit, and a linkage prevention and control actuator;

[0052] The detector group is used to obtain detection information corresponding to the detectors in the detector group in the detection area. The detectors in the detector group include thermal imaging detectors and several physical detectors; the detection information of the thermal imaging detector can be selected from ambient temperature, thermal imaging images, thermal imaging videos, visible light images, and visible light videos; the physical detector can be selected from smoke detectors, CO concentration detectors, temperature detectors, temperature sensing cables, three-in-one detectors, etc.

[0053] The TCMS system interface unit is used to obtain locomotive fault information, locomotive operation data, and detection area component information; and can interact with the controller, the TCMS system sends control instructions, and the controller feeds back fire alarms, collected image / video data, etc.;

[0054] The control unit is used to determine whether there is an abnormality in the detection area based on the detection information of the physical detector by low-frequency acquisition of the detection information of the thermal imaging detector, so as to prevent missed reports when a fire has occurred in the detection area; when an abnormality exists in the detection area or the detection information of the physical detector is greater than or equal to the set warning threshold data, a method of high-frequency acquisition of visible light images / visible light videos is used, that is, visible light images / visible light videos are acquired at each high-frequency time interval to determine for the second time whether there is an abnormality in the detection area, so as to prevent false reports of fires in the detection area; and after determining for the second time whether there is an abnormality in the detection area, the location of the fire is determined based on the fire-related fault information, and then the fire danger level is determined based on the fire spread trend. The fire danger levels include high danger level, medium danger level and low danger level;

[0055] Specifically, during the fire identification process, the TCMS system interface unit obtains locomotive fault information, and based on the fire characteristics of the locomotive materials, equipment power consumption, and wear and tear, combined with the locomotive fault information, screens the fault information associated with the fire; based on the electrical schematic diagram, extracts the location and state of the collected variables in the fault entry, and locates the equipment abnormal point in combination with the fault description and related logic;

[0056] Specifically, the control unit also transmits the above information to the TCMS system and issues an audible and visual alarm. The thermal imaging detector used in this embodiment is a binocular detector, which can acquire visible light images and videos as well as thermal imaging images and videos. By using a method of low-frequency acquisition of the detection information of the thermal imaging detector when the physical detector does not reach the set warning threshold data, an image recognition algorithm or a manual recognition method is used to determine whether there is an abnormality in the detection area, which can effectively avoid the problem of underreporting caused by unreasonable threshold setting. At the same time, when an abnormality is detected in the detection area, the visible light image / visible light video is identified and analyzed manually by using an image recognition algorithm or manually, and the problem causing the fire is determined based on the fire-related fault information of the locomotive, to prevent false alarms of fires in the detection area.

[0057] The linkage prevention and control execution mechanism is used to execute fire extinguishing operations on the detection area according to the danger level of the fire, fire-related fault information and the location where the fire occurred.

[0058] Preferably, the method for determining whether there is an abnormality in the detection area is as follows:

[0059] When the detection data of the physical detector is less than the set warning threshold data:

[0060] If the detection data of the physical detector is less than the first warning data, a thermal imaging image / thermal imaging video is acquired at every first time interval to determine whether there is an abnormality in the detection area;

[0061] If the detection data of the physical detector is greater than or equal to the nth warning data and less than the n+1th warning data, a thermal imaging image / thermal imaging video is acquired at every n+1th time interval to determine whether there is an abnormality in the detection area; n is a natural number, n=1,…,n,n+1,…,N; N represents the total number of low-frequency time intervals; wherein the Nth warning data is the set warning threshold data; in this embodiment, when the detection data of the physical detector is greater than or equal to the N-1th warning data and less than the Nth warning data, that is, the set warning threshold data, a thermal imaging image / thermal imaging video is acquired at every Nth time interval,

[0062] When the detection data of the physical detector is greater than or equal to the set warning threshold data, there is an abnormality in the detection area;

[0063] The first time interval>...>nth time interval>n+1th time interval>...>Nth time interval>high-frequency time interval;

[0064] 1st warning data<...<nth warning data<n+1th warning data<...<Nth warning data.

[0065] In this embodiment, a specific method for determining whether there is an abnormality in the detection area is as follows:

[0066] When the detection data of the physical detector is less than the set warning threshold data:

[0067] If the detection data of the physical detector is less than the first warning data, a thermal imaging image / thermal imaging video is acquired at every first time interval to determine whether there is an abnormality in the detection area;

[0068] If the detection data of the physical detector is greater than or equal to the first warning data and less than the second warning data, a thermal imaging image / thermal imaging video is acquired at every second time interval to determine whether there is an abnormality in the detection area;

[0069] If the detection data of the physical detector is greater than or equal to the second warning data and less than the set warning threshold data, a thermal imaging image / thermal imaging video is obtained at every third time interval to determine whether there is an abnormality in the detection area;

[0070] When the detection data of the physical detector is greater than or equal to the set warning threshold data, it is directly determined that there is an abnormality in the detection area;

[0071] The first time interval>the second time interval>the third time interval>the high frequency time interval;

[0072] The first warning data < the second warning data < the warning threshold data.

[0073] Specifically, the method used to determine whether there is an abnormality in the detection area through thermal imaging images / thermal imaging videos is performed through existing image recognition methods or manual recognition methods, so it is not described in detail here.

[0074] Specifically, in this embodiment, existing fires are often caused by inconspicuous hidden fires, and it is difficult to detect such hidden fire signs using existing ordinary monitoring systems. However, thermal imaging detectors can quickly and effectively detect these hidden fires, and can accurately determine the location and scope of the fire, that is, they can timely detect the ignition point through temperature / smoke / carbon monoxide concentration, etc., and extinguish the fire in its infancy.

[0075] Specifically, the detection data of the temperature detector / temperature sensing cable in this embodiment is temperature data; the detection data of the smoke detector is smoke concentration data; the detection data of the CO concentration detector is CO concentration data;

[0076] Specifically, when the detection data of the temperature detector / temperature sensing cable / smoke detector / CO concentration detector / thermal imaging detection temperature does not reach the set alarm data, it is in the early stage of the fire. In order to identify the fire as early as possible and avoid delays or missed reports due to unreasonable setting of warning parameters, for the situation when the warning threshold data is not reached, low-frequency image acquisition is performed (that is, thermal imaging images or thermal imaging videos are collected at longer intervals), that is, thermal imaging images and thermal imaging videos are collected and identified at regular intervals, and the image is pre-processed to determine whether there is an abnormality in the detection area, so that missed fires can be discovered in time.

[0077] Specifically, this embodiment takes the detection information of the temperature detector / temperature sensing cable as an example for explanation as follows: when the detection temperature of the temperature detector / temperature sensing cable reaches the normal operating temperature of the equipment and is less than the first warning temperature (optionally 80 degrees), a thermal imaging image / thermal imaging video is collected once every first time interval (such as 30 minutes); when the temperature is greater than or equal to the first warning temperature (optionally 80 degrees) and less than the second warning temperature (optionally 120 degrees), a thermal imaging image / thermal imaging video is collected once every second time interval (such as 10 minutes); when the temperature is greater than or equal to the second warning temperature and less than the set warning temperature data (optionally 150 degrees), a thermal imaging image / thermal imaging video is collected once every third time interval (such as 5 minutes); by collecting the thermal imaging image / thermal imaging video, a thermal imaging image is obtained to determine whether there is an abnormality in the detection area, that is, through the thermal imaging image, determine whether there is a temperature increase value per unit time greater than the set temperature rise threshold;

[0078] Specifically, when it is determined that there is an abnormality in the detection area, there are two situations. One is that the detection temperature of the temperature detector / temperature sensing cable is lower than the set warning temperature data, then the visible light image and visible light video are acquired at each high-frequency time interval, that is, the visible light image / visible light video is acquired at a shorter interval (such as setting 10 seconds or 1 second to acquire and identify the thermal imaging image / thermal imaging video) to judge the fire situation, which can prevent false alarms of the fire situation. The second situation is that when the detection temperature of the temperature detector / temperature sensing cable is greater than or equal to the set warning temperature data (optional 150 degrees), the visible light image and visible light video are acquired at each high-frequency time interval, that is, the visible light image / visible light video is acquired at a shorter interval (such as setting 10 seconds or 1 second to acquire and identify the thermal imaging image / thermal imaging video) to judge the fire situation, which can prevent false alarms of the fire situation.

[0079] Specifically, in this embodiment, the collected low-frequency image video is used to determine whether there is a missed fire, and the image and video processing can be achieved through computer vision algorithms or manual recognition. During manual recognition, the relevant information can be displayed through the display screen of the controller itself, or through the original display screen of the locomotive. The driver and crew will manually judge whether there is a fire or not, and give information about the presence or absence of a fire, and feed it back to the controller through the TCMS system. If the collection of high-frequency images and videos is used to dynamically track the fire status, and no abnormality is found after manual or computer algorithm recognition, it is considered to be a false alarm information from the detector. At this time, you can request to retransmit the image data for the next time period. After re-confirmation, if no abnormality is found, ignore the alarm information; record the false alarm data for subsequent statistics.

[0080] Preferably, the fire-related fault information includes: fault name, name of the component to which the fault belongs, fault code, fault description information, fault handling suggestions, etc. The method for obtaining the fire-related fault information of the locomotive is: based on the fire characteristics of the materials used in the locomotive, the power consumption of the equipment, and the difficulty of loss, combined with the locomotive fault information, the fault information associated with the fire is screened; according to the electrical schematic diagram, the position and state of the collected variables in the fault entry are extracted, and the abnormal point of the equipment is located in combination with the fault description and related logic;

[0081] After determining whether there is an abnormality in the detection area, it also includes determining the location where the abnormality occurs based on the fire-related fault information of the locomotive.

[0082] Specifically, this embodiment screens the fault information associated with the fire, extracts the location and state of the acquisition variables in the fault entry according to the electrical schematic diagram, and locates the abnormal point of the equipment in combination with the fault description and related logic. And with the help of the visible light image acquisition function of the thermal imaging detector, the abnormal state of the faulty component is analyzed and identified, and the source of the fire is identified and located.

[0083] Specifically, three conditions are required for a general fire to burn: combustibles, combustion-supporting materials, and ignition sources. Locomotive equipment is mostly installed in electrical cabinets and packaged in metal shells. The main sources of locomotive fires include cables, connectors, circuit boards, and relays, contactors, IGBTs, etc. that operate switches, such as failure or abnormality of electronic devices, overheating of cables and devices due to damage, short circuits caused by poor heat dissipation of electronic devices, overheating, etc., short circuits of circuit boards and connectors, sparks caused by relay jamming, arcs, or leakage caused by humid environments. Due to the high voltage, small space, many equipment, heavy load, and high power consumption of locomotives, when a fire occurs, it is not as obvious as in some vast lands with wide views. Electrical equipment is mostly installed in independent electrical cabinets, and there is no personnel guarding the machine room after the locomotive is running. During normal operation, the driver and crew concentrate on driving the locomotive, and can only check the status of the machine room through monitoring equipment when there is an abnormality, which poses a safety hazard. These abnormal situations will also be reflected in locomotive failures. Therefore, by combining locomotive fault information, the faulty equipment can be quickly located, so that effective fire alarm and fire extinguishing measures can be taken. The faulty equipment can also be isolated by equipment isolation or power off to reduce the occurrence of secondary hazards.

[0084] Specifically, this embodiment combines the acquisition of visible light images or visible light videos at each high-frequency time interval with the fire-related fault information of the locomotive to determine the location of the fire and locate the fire. Then, the fire spread is dynamically tracked and identified through the existing image recognition algorithm that can recognize the visible light images or visible light videos acquired at high frequencies, and the fire danger level is determined based on the fire spread trend.

[0085] Preferably, a method of determining the danger level of a fire according to the spreading trend of the fire in this embodiment is:

[0086] According to the thermal imaging device / temperature sensing detector / temperature sensing cable, the temperature increase rate of the detector area is obtained:

[0087] If the temperature increase rate per unit time is greater than or equal to the first temperature rise threshold, the fire danger level is a high danger level;

[0088] If the temperature increase rate per unit time is less than the first temperature rise threshold and greater than or equal to the second temperature rise threshold, the fire danger level is a medium danger level;

[0089] If the temperature increase rate per unit time is less than the second temperature rise threshold, the fire danger level is a low danger level;

[0090] Among them, the first temperature rise threshold>the second temperature rise threshold>the third temperature rise threshold.

[0091] This embodiment uses the thermal imaging image or thermal imaging video acquired by the thermal imaging detector, and with the help of the visible light image acquisition function of the thermal imaging detector, analyzes and identifies the abnormal state of the faulty component, identifies and locates the source of the fire; specifically, obtains the temperature difference per unit time in the detection area to evaluate the spread trend of the fire. By obtaining the temperature difference of each part of the detection area, the fire scene is dynamically tracked, the trend of the maximum temperature change of the flame is identified, and the spread trend of the fire is estimated by identifying the rapid change of the temperature difference in a short period of time. Specifically, the temperature difference per unit time and the speed of temperature difference change, such as the temperature difference rising rate of more than 80 degrees within 1 minute under continuous monitoring, the fire is rising rapidly and a high danger level is given. If the temperature difference rising rate of the fire is more than 50 degrees within 1 minute under continuous monitoring, a medium danger level is given. Otherwise, a low danger level is given.

[0092] Preferably, another method of determining the danger level of a fire according to the spreading trend of the fire in this embodiment is:

[0093] According to the thermal imaging device, the rate of increase of the fire area in the detector area is obtained:

[0094] If the rate of increase of the fire area per unit time is greater than or equal to the first area change threshold, the fire danger level is a high danger level;

[0095] If the rate of increase of the fire area per unit time is less than the first area change threshold and greater than or equal to the second area change threshold, the fire danger level is a medium danger level;

[0096] If the rate of increase of the fire area per unit time is less than the second area change threshold, the fire danger level is a low danger level;

[0097] Among them, the first area change threshold>the second area change threshold>the third area change threshold.

[0098] This embodiment also obtains the area of ​​the flame and the direction of the flame spread through the obtained visible light image or visible light video to evaluate the spread trend of the flame and the degree of fire hazard; to guide the fire-fighting operation of the linkage prevention and control executive agency. Specifically, through visible light images or visible light videos, it is possible to dynamically track the fire scene in the detection area through computer algorithms or manually, identify the direction and trend of the fire spread, and estimate the flame area to guide the fire-fighting operation of the linkage prevention and control executive agency. Specifically, the visible light image identifies the fire and area, such as the fire area increases by 10cm within 1 minute of continuous monitoring. 2 If the fire area increases by 5cm within 1 minute of continuous monitoring, the fire spreads rapidly and a high risk level is given. 2 , a medium risk level is given. Otherwise, a low risk level is given.

[0099] Preferably, the method for performing the fire extinguishing operation on the detection area is as follows:

[0100] When the fire danger level is low, start the nozzle corresponding to the location where the fire occurs and spray a single volume of fire extinguishing agent to extinguish the fire;

[0101] When the fire is at a medium level, the nozzles corresponding to the location where the fire is generated are activated to spray double the volume of fire extinguishing agent to extinguish the fire;

[0102] When the fire danger level is high, start all the sprinklers and spray three times the capacity of fire extinguishing agent to extinguish the fire.

[0103] Preferably, before performing the fire extinguishing operation on the detection area, the method further includes:

[0104] If the optical image / visible light video can be used to determine that there is an arc or spark at the location where the fire occurred, or there are traces of burns caused by sparks, or if it is determined that there is a short circuit or a cable fall-off or other phase loss at the location where the fire occurred, the source of the fire can be identified and located. At the same time, the driver is requested to cut off the power.

[0105] Specifically, with the help of the visible light image acquisition function of the thermal imaging detector, the abnormal state of the faulty component can be analyzed and identified, and the source of the fire can be identified and located;

[0106] Specifically, if the locomotive fault information is determined to be a short-circuit fault through the visible light video or visible light image, or there is a relay jam (need to be identified in combination with the image) with arc or spark, it is necessary to request the driver to cut off the power supply for processing, and at the same time, start the fire extinguishing process in combination with the fire spread. When it is determined through the fire-related fault information in the detection area that there are multiple fire-related faults in the detection area, it is also necessary to request the driver to cut off the power supply for processing, and at the same time, start the fire extinguishing process in combination with the fire spread. When it is detected that the fire has begun to spread, if it is a medium-risk level, only the sprinklers at the fire site are activated to spray the fire extinguishing agent, but double the capacity of the fire extinguishing agent is used; when it reaches a high-risk level, triple the capacity of the fire extinguishing agent is used, and all sprinklers are activated to spray the fire extinguishing agent; otherwise, the conventional fire extinguishing process is used, and only one-time fire extinguishing treatment is performed on the fire site.

[0107] Preferably, it also includes a LKJ interface unit;

[0108] The LKJ interface unit is used to obtain the current running position information of the locomotive; and based on the current running position information of the locomotive, determine the distance between the locomotive and the previous station and the next station, and then be able to know the distance between the current locomotive position and the entry and exit platform; to determine whether it is necessary to start the emergency handling procedure.

[0109] Specifically, when the emergency handling procedure needs to be initiated: first, the driver and passengers activate the siren to alert the crowd and call the police; then the fire extinguishing operation in the detection area is performed through the linkage prevention and control execution mechanism.

[0110] Specifically, in this embodiment, when the distance between the locomotive and the platform is less than a set distance threshold, or the time it takes for the locomotive to arrive at the platform is less than a set time threshold, an emergency handling procedure is adopted;

[0111] Preferably, the detector group also includes one or two of an atmospheric pressure sensor and an air flow velocity sensor, and the atmospheric pressure sensor and the air flow velocity sensor are used to obtain the gas flow velocity in the enclosed space, so as to evaluate the impact of the spread of flames when a fire has occurred, so as to guide the fire extinguishing operation of the linkage prevention and control actuator.

[0112] Specifically, the detector group of this embodiment also includes one or two of an atmospheric pressure sensor and an air flow velocity sensor. The atmospheric pressure sensor and the air flow velocity sensor are used to obtain the gas flow velocity of the enclosed space when the detection area is a confined space, so as to evaluate the degree of influence on the spread of the flame when a fire has occurred, so as to guide the fire extinguishing operation of the linkage prevention and control actuator. Specifically, after a fire occurs in a confined space, the pressure in the space will be different from the external pressure. At this time, rashly opening the confined space is likely to cause secondary hazards. Therefore, it is necessary to pay attention to the changes in the atmospheric pressure in the space. If the atmospheric pressure rises by more than 5kpa within 1 minute of continuous monitoring, a high risk level is given. If the atmospheric pressure rises by more than 3kpa within 1 minute of continuous monitoring, a medium risk level is given. Otherwise, a low risk is given.

[0113] Specifically, the detector group in this embodiment can be composed of thermal imaging detectors, smoke detectors, and preferably also CO concentration detectors, temperature detectors, temperature sensing cables, and atmospheric pressure sensors, air flow velocity sensors, etc. can also be added for confined spaces; each detector group is set according to the scene, and the number is N (N is a natural number).

[0114] Specifically, the TCMS system interface unit of this embodiment is connected to the locomotive TCMS system to obtain locomotive operation data, locomotive fault information, and detection area component information; wherein, the detection area component information is such as the voltage value of the component in the detection area, the current value of the component, the load state of the component or the current state of the relay-type equipment (the switching quantity of the relay), etc., wherein the detection area component information can be combined with the locomotive fault information to determine whether the component has defects such as cracks; the locomotive fault information is generally stored and displayed through the locomotive real-time fault table, including but not limited to the component to which the fault belongs, fault code, fault description, and fault handling suggestions. Such as overvoltage, overcurrent, overtemperature or failure of electronic devices, relay jamming, etc. The locomotive operation data includes but is not limited to locomotive running speed, locomotive network voltage, battery voltage, traction / braking force, locomotive running time, driver and crew related operation commands, etc.

[0115] Specifically, the linkage prevention and control actuator of this embodiment, after receiving the fire alarm information sent by the controller, triggers the corresponding area control valve to operate according to the fire danger level, fire-related fault information and the location of the fire, completes the fire extinguishing agent spraying, and realizes the cooling and fire extinguishing function. The actuator includes a fire extinguishing agent bottle group, a connecting pipe, a cylinder pressure gauge, a nozzle, a spraying feedback device, a safety release device, an emergency start switch, a fire fighting pipe network and other components.

[0116] Specifically, the fire warning and prevention system of this embodiment further includes a display unit for displaying fire-related information, including image information, video information, detector data, TCMS unit data, etc.

[0117] Specifically, the display unit can be the original display unit in the locomotive driver's cab, or an independent display screen, which is used to display the visible light video and image, thermal imaging video and image obtained by the thermal imaging detector in the detector group, and the fire alarm information obtained by the control unit and the received TCMS system data. If necessary, send relevant instructions to the control unit to control the linkage prevention and control execution mechanism to execute fire extinguishing spraying and other disposal measures.

[0118] Specifically, this embodiment also includes a 6A interface unit, which is used to obtain environmental information in the machinery room and information such as bogie and axle temperature, which can assist the staff in determining the occurrence and judgment of the fire.

[0119] Specifically, the control unit in this embodiment is a data exchange center, which is responsible for comprehensive analysis and processing of information. It receives detection information from detection devices such as temperature, smoke concentration, equipment fault information, visible light video and images collected by thermal imaging detectors, thermal imaging videos and images, etc. collected by each detection device in the detector group, combined with locomotive operation data, locomotive fault information and detection area component information obtained from the TCMS system interface unit, and gives fire alarm information after fire analysis to control the fire extinguishing device to perform fire extinguishing operations, and transmits the above information to the TCMS system to issue an audible and visual alarm.

[0120] Specifically, this embodiment determines the distance between the locomotive and the platform according to the running position of the locomotive; and determines the time when the locomotive leaves the platform or the time when the locomotive is about to arrive at the platform in combination with the running speed of the locomotive to determine whether it is necessary to start the emergency handling procedure: if the locomotive has just left or is about to enter the platform (for example, according to the current running speed of the locomotive, the distance from the locomotive to the platform is less than ten minutes, or the distance between the locomotive and the platform is not more than 10 kilometers), then the emergency handling procedure needs to be started. At this time, it is considered that there are many people around it and the danger is greater. The driver and crew need to sound the siren to warn the surrounding people to evacuate the personnel, and quickly complete the alarm confirmation and processing. At the same time, because the location is not very remote, when the situation is critical, the ground fire can be used to assist in fire fighting. At this time, the geographical location level is Class A. Then start the linkage prevention and control executive agency to perform the fire extinguishing operation on the detection area. Specifically, after the driver and crew sound the siren to remind the surrounding people to evacuate, the driver and crew can stop the car while ensuring their own safety, and automatically or manually start the fire sprinkler to spray the fire extinguishing agent for fire extinguishing. If necessary, the ground rescue team can be requested for assistance.

[0121] When the locomotive is far away from the platform (for example, according to the current locomotive running speed, the distance for the locomotive to reach the platform is more than ten minutes, or the distance between the locomotive and the platform is more than 10 kilometers), at this time, it is generally considered that the line is relatively remote and sparsely populated, and the linkage prevention and control executive mechanism is directly activated to perform fire extinguishing operations in the detection area. The geographical location level is C. At this time, the locomotive is far away from the platform and the personnel are sparse. For lines running in remote areas with sparse population, there is no need to sound the whistle.

[0122] Specifically, the locomotive fault information in this embodiment is obtained through the fault information table updated in real time by the locomotive. It includes locomotive detection area faults, detection area input end faults and detection area load faults, and the content covers the fault component, fault code, fault description, and fault handling suggestions. Such as overvoltage, overcurrent, overtemperature or failure of electronic devices, relay jamming, etc. Among them, it is distinguished according to the degree of influence on the fire, such as relay jamming faults, such as sparks or arcs, which may cause the relay to burn and cause a large-scale fire, leading to a high-risk state. Therefore, it is judged by combining visible light image recognition of abnormal conditions of faulty components; if it is only mechanical jamming, or there are no other faults such as jamming, but such faults can be detected from the locomotive fault information table, and it is not a continuous fault report. Intermittent good and bad types are mostly poor contact. By combining visible light images to identify abnormal conditions such as no burning marks on the faulty components, the components can be replaced after the locomotive enters the section. Short circuits and overtemperatures caused by device aging can be identified in combination with locomotive operation information, such as abnormal current values ​​of related equipment and sudden changes in equipment temperature.

[0123] Specifically, this embodiment introduces locomotive faults to assist in fire identification because locomotive faults are updated in real time, and the location of the faulty equipment can be identified through fault coding, and then the corresponding position detector is called to obtain detection data to locate the location where the fire may occur, avoiding blind searches and facilitating fault location.

[0124] When a fault is detected, relevant data can be obtained through TCMS. If it is an independent electronic component, the input voltage, current and load condition of the component will be checked. If it is a contactor or relay device, the received command information and status feedback information will be checked, and the degree of fire danger can be comprehensively determined based on the image recognition results.

[0125] In addition, the visible light image video of this embodiment can also monitor mechanical equipment defects, such as mechanical part breakage. Since the failure of such mechanical parts cannot be obtained in real time through the TCMS system, and there is no relevant fault entry in the fault record, it can only be discovered through manual maintenance after the locomotive returns to the section. This invisible fault is also one of the key factors that cause locomotive fire accidents. By adding image detection methods, fire hazards can be discovered early and the hazards caused by such faults can be reduced.

[0126] Specifically, the present embodiment processes thermal imaging images, thermal imaging videos, visible light images and visible light videos, and the methods that can be used include frame difference method, BP neural network, convolutional neural network, YOLO, time series, optical flow method, background subtraction method, frame difference method, etc., which are not introduced in detail here. The thermal imaging images in the present embodiment are displayed in different colors according to the temperature values ​​of the objects, so the boundary judgment of the objects is weak. The visible light images identify the fire and the area, and the temperature change and the severity of the fire are analyzed by thermal imaging.

[0127] Specifically, when a fire has occurred, this embodiment evaluates the spread of the fire: including: adding one or two of an atmospheric pressure sensor and an air flow velocity sensor to the confined space to detect the gas flow velocity and evaluate the impact of the fire spread.

[0128] Preferably, image segmentation is performed on thermal imaging images and visible light images to assist in determining the location of the fire and guide the fire-fighting operation of the linkage prevention and control executive agency.

[0129] Specifically, this embodiment can also perform image segmentation on thermal images and visible light images through the image recognition and warning display unit, which helps to determine the relationship between targets and target positioning, so as to determine the location of the fire and guide the fire extinguishing operation of the linkage prevention and control executive agency. Among them, image segmentation is a conventional technology in the field and will not be described in detail here.

[0130] The specific real-time scenario of this embodiment is as follows:

[0131] Scenario 1: (Low-frequency image detection stage): A single physical detector reaches the threshold, and detects whether there is an abnormality in the detection area through low-frequency image recognition or manual analysis methods: By obtaining the data of each detector, it is found that a single detector detects the warning threshold, such as the temperature detected by the thermal imager is greater than 180°C, the smoke reaches the alarm threshold and lasts for 60 seconds, and the temperature rise exceeds 3°C within 60 seconds; through the image recognition algorithm, start the recognition of the low-frequency collected thermal imaging image to determine whether there is a fire. If no suspected smoke or flame temperature is found during manual detection or preprocessing, the suspected fire signal 0 is fed back to the fire controller through the display screen, which is confirmed as a false alarm; if smoke or flames are detected, the image recognition confirms that there is an abnormality, and there is a possibility of a fire at this time. Next, combined with the locomotive fault information related to the fire, continue with the subsequent operations.

[0132] Scenario 2 (high-frequency image detection stage): At this point, it has been determined that there is an abnormality in the detection area. The locomotive operation information and locomotive fault information are obtained from the TCMS data to check whether there is any fault information about the equipment in the detection area. If there is no fault information about the equipment in this area, the status of the faulty component is automatically or manually analyzed through the high-frequency collected visible light video / image and thermal imaging image / video to determine whether there is a fire. Otherwise, it is necessary to further obtain TCMS related data, such as component input voltage, current, load condition, relay status, etc., to obtain image information of faulty components that are prone to mechanical failures that cannot be reported, and to identify and verify whether there are mechanical cracks, loose connectors, detachment, relay jams, etc. After the fire is identified manually or by an algorithm, the fire signal is fed back to the fire controller through the display screen, high-frequency image acquisition is started, the fire trend is analyzed, and the fire extinguishing device is started for fire extinguishing treatment. The locomotive is stopped and the power is cut off. According to the distance from the platform, it is analyzed whether it is necessary to request reinforcements from ground personnel, whether the towed carriage can be unhooked, and the connection between the dangerous source and other carriages can be disconnected.

[0133] Scenario 3 (high-frequency image detection stage): At this point, it has been determined that there is an abnormality in the detection area. The locomotive operation information and locomotive fault information are obtained from the TCMS data to check whether there is any fault information about the equipment in the detection area. If there is no fault information about the equipment in this area, the visible light, thermal imaging image or video is retrieved, and the faulty component status is automatically or manually analyzed to check whether there is a fire. Otherwise, it is necessary to further obtain TCMS related data, such as component input voltage, current, load condition, relay status, etc., to obtain image information of faulty components that are prone to mechanical failures that cannot be reported, and to identify and verify whether there are mechanical cracks, loose connectors, detachment, relay jams, etc. After the fire is identified manually or by an algorithm, the fire signal can be fed back to the fire controller through the display screen, high-frequency image acquisition can be started, the fire trend can be analyzed, the fire extinguishing device can be started for fire extinguishing treatment, and the locomotive can be stopped and powered off. If the fire trend analysis is determined to be high-risk, the fire extinguishing device can be started for fire extinguishing treatment, the locomotive can be stopped and powered off, and whether it is necessary to request reinforcements from ground personnel based on the distance from the platform; whether the towed carriage can be unhooked to disconnect the connection between the dangerous source and other carriages.

[0134] Specifically, if the fire trend analysis shows a medium danger level and will not endanger other carriages, the fire extinguishing device will be activated to extinguish the fire, and the locomotive will be stopped and the power will be cut off. Based on the distance from the platform, it will be analyzed whether it is necessary to request reinforcement from ground personnel. If it has spread to a larger area, the fire extinguishing dosage will be increased. If the fire trend analysis shows a low danger level, the fire extinguishing device will be activated to extinguish the fire, and the locomotive will be stopped and the power will be cut off.

[0135] This embodiment collects environmental information through a variety of environmental sensing devices, and combines the above information with the locomotive TCMS data and LKJ data. The control unit determines the alarm status and provides the fire alarm level. The display unit displays fire-related information, receives instructions from drivers and passengers, and can provide linkage control and execute prevention and control when necessary.

[0136] Finally, it should be noted that the above embodiments are only used to illustrate the technical solutions of the present invention, rather than to limit it. Although the present invention has been described in detail with reference to the aforementioned embodiments, those skilled in the art should understand that they can still modify the technical solutions described in the aforementioned embodiments, or replace some or all of the technical features therein with equivalents. However, these modifications or replacements do not cause the essence of the corresponding technical solutions to deviate from the scope of the technical solutions of the embodiments of the present invention.

Claims

1. A multi-information fusion locomotive fire warning and prevention system, characterized in that: include: Control unit, detector group, TCMS system interface unit, linkage prevention and control actuator; The detector group is used to obtain detection information corresponding to the detectors in the detector group in the detection area; the detectors in the detector group include a plurality of thermal imaging detectors and a plurality of physical detectors; The physical detectors include: smoke detectors, CO concentration detectors, temperature detectors, and temperature sensing cables; The detection information of the thermal imaging detector includes: thermal imaging images, thermal imaging videos, visible light images, and visible light videos; The TCMS system interface unit is used to obtain the fault information of the locomotive, so as to obtain the fire-related fault information according to the fault information of the locomotive; The control unit is used to determine whether there is an abnormality in the detection area according to the detection information of the physical detector by low-frequency acquisition of the detection information of the thermal imaging detector, so as to prevent underreporting when a fire has occurred in the detection area; when there is an abnormality in the detection area or the detection information of the physical detector is greater than or equal to the set warning threshold data, a visible light image / visible light video is acquired at each high-frequency time interval to determine for the second time whether there is an abnormality in the detection area, so as to prevent false alarms of the fire in the detection area; and when it is determined for the second time that there is an abnormality in the detection area, the location of the fire is determined according to the fire-related fault information, and then the degree of danger of the fire is determined according to the spread trend of the fire; The linkage prevention and control execution mechanism is used to execute fire extinguishing operations on the detection area according to the danger level of the fire, fire-related fault information and the location where the fire occurred.

2. A multi-information fusion locomotive fire warning and prevention system according to claim 1, characterized in that: The fire-related fault information includes: fault name, name of the component to which the fault belongs, fault code, fault description information, and fault handling suggestions; The method for obtaining fire-related fault information is as follows: based on the fire characteristics of the materials used in the locomotive, the power consumption of the equipment, and the difficulty of loss, combined with the fault information of the locomotive, the fault information associated with the fire is screened; based on the electrical schematic diagram, the position and state of the collected variables in the fault entry are extracted, and the abnormal point of the equipment is located in combination with the fault description and related logic; After determining whether there is an abnormality in the detection area, it also includes determining the location where the abnormality occurs based on the fire-related fault information of the locomotive.

3. The multi-information fusion locomotive fire warning and prevention system according to claim 1 is characterized in that: The method for determining whether there is an abnormality in the detection area is as follows: When the detection data of the physical detector is less than the set warning threshold data: If the detection data of the physical detector is less than the first warning data, a thermal imaging image / thermal imaging video is acquired at every first time interval to determine whether there is an abnormality in the detection area; If the detection data of the physical detector is greater than or equal to the nth warning data and less than the n+1th warning data, a thermal imaging image / thermal imaging video is acquired at every n+1th time interval to determine whether there is an abnormality in the detection area; n is a natural number, n=1,…,n,n+1,…,N; N represents the total number of low-frequency time intervals; wherein the Nth warning data is the set warning threshold data; When the detection data of the physical detector is greater than or equal to the set warning threshold data, there is an abnormality in the detection area; The first time interval>...>nth time interval>n+1th time interval>...>Nth time interval>high-frequency time interval; 1st warning data<...<nth warning data<n+1th warning data<...<Nth warning data.

4. The multi-information fusion locomotive fire warning and prevention system according to claim 1 is characterized in that: According to the spread trend of fire, the method to determine the degree of fire danger is: According to the thermal imaging detector / temperature sensing detector / temperature sensing cable, the temperature increase rate of the detector area is obtained: If the temperature increase rate per unit time is greater than or equal to the first temperature rise threshold, the fire danger level is a high danger level; If the temperature increase rate per unit time is less than the first temperature rise threshold and greater than or equal to the second temperature rise threshold, the fire danger level is a medium danger level; If the temperature increase rate per unit time is less than the second temperature rise threshold, the fire danger level is a low danger level; Among them, the first temperature rise threshold>the second temperature rise threshold>the third temperature rise threshold.

5. The multi-information fusion locomotive fire warning and prevention system according to claim 1 is characterized in that: According to the spread trend of fire, the method to determine the degree of fire danger is: According to the thermal imaging detector, the rate of increase of the fire area in the detector area is obtained: If the rate of increase of the fire area per unit time is greater than or equal to the first area change threshold, the fire danger level is a high danger level; If the rate of increase of the fire area per unit time is less than the first area change threshold and greater than or equal to the second area change threshold, the fire danger level is a medium danger level; If the rate of increase of the fire area per unit time is less than the second area change threshold, the fire danger level is a low danger level; Among them, the first area change threshold>the second area change threshold>the third area change threshold.

6. The multi-information fusion locomotive fire warning and prevention system according to claim 1 is characterized in that: The method of performing fire extinguishing operation on the detection area is as follows: When the fire danger level is low, start the nozzle corresponding to the location where the fire occurs and spray a single volume of fire extinguishing agent to extinguish the fire; When the fire is at a medium level, the nozzles corresponding to the location where the fire is generated are activated to spray double the volume of fire extinguishing agent to extinguish the fire; When the fire danger level is high, start all the sprinklers and spray three times the capacity of fire extinguishing agent to extinguish the fire.

7. The multi-information fusion locomotive fire warning and prevention system according to claim 1 is characterized in that: Before executing the fire extinguishing operation in the detection area, it also includes: According to the detection information of the thermal imaging detector, the abnormal status of the faulty components in the detection area is obtained: if it is possible to determine that there is an arc or spark at the location where the fire occurred based on the optical image / visible light video, or if it is possible to determine that there is a short circuit or phase loss at the location where the fire occurred through the fire-related fault information, the source of the fire can be located and the driver can be requested to cut off the power.

8. The multi-information fusion locomotive fire warning and prevention system according to claim 1 is characterized in that: Also includes LKJ interface unit; The LKJ interface unit is used to obtain the current running position information of the locomotive; and according to the current running position information of the locomotive, determine the distance between the locomotive and the previous station and the next station, and then be able to know the distance between the current locomotive position and the in and out platform; To determine whether emergency procedures need to be initiated.

9. The multi-information fusion locomotive fire warning and prevention system according to claim 1 is characterized in that: By performing image segmentation on thermal imaging images and visible light images, we can assist in determining the location of the fire and guide the fire-fighting operations of the joint prevention and control executive agencies.

10. The multi-information fusion locomotive fire warning and prevention system according to claim 1, characterized in that: The detector group also includes one or two of an atmospheric pressure sensor and an air flow velocity sensor, and the atmospheric pressure sensor and the air flow velocity sensor are used to obtain the gas flow velocity in a confined space, so as to evaluate the impact of the spread of flames when a fire has occurred, so as to guide the fire extinguishing operation of the linkage prevention and control actuator.

Images