Engineering building detection method and system

Abstract

The invention discloses an engineering building detection method and system, and relates to the technical field of building fire detection and early warning. Through systematic area division and multi-dimensional sensor network monitoring, environment and equipment operation data of a key area can be collected in real time; and obtaining and evaluating a layout closeness index Ji, a cable overload coefficient Oi, an electrical load change index Li and a toxic gas leakage coefficient Ci of the ith monitoring area, thereby improving the early warning capability of the fire risk. Through combination of smoke diffusion analysis and fire risk index evaluation, the fire risk can be accurately judged, and a fire control instruction can be triggered in a targeted manner. Besides, based on a smoke diffusion index correction mechanism, the system can dynamically adjust the operation states of the spraying equipment and the ventilation facility, so that fire control precision and resource saving are realized, and the fire safety and emergency response efficiency of the building are effectively improved.

Description

Technical Field

[0001] The present invention relates to the technical field of building fire detection and early warning, and specifically to an engineering building detection method and system. Background Art

[0002] The safety detection and assessment of engineering buildings are key links to ensure the stability of building structures, the safety of electrical systems, and the environmental suitability. Especially in high-risk places such as computer rooms, substations, and underground spaces, precise monitoring and assessment technologies are particularly important. With the continuous expansion of the scale and the increase in complexity of construction projects, traditional manual inspection methods are difficult to meet the requirements of efficient and accurate monitoring. Therefore, intelligent and automated building monitoring and assessment methods have gradually become the focus of research.

[0003] Traditional systems have insufficient ability to assess the risk of fire spread. After a fire occurs, the spread speed of smoke and fire often exceeds expectations, and traditional monitoring means cannot collect smoke spread data in real time and analyze it. This limitation makes it difficult to effectively predict the fire spread range, delays the best time for spread control, and thus exacerbates the threat of fire to buildings and personnel. In addition, although some engineering buildings are equipped with automatic sprinkler systems, their layout spacing, spraying accuracy, and linkage control capabilities still need to be optimized and cannot achieve dynamic adjustment based on the development trend of the fire. Most existing fire-fighting equipment is mainly triggered uniformly, such as simply turning on the sprinkler devices in the entire area. This "global trigger" mode not only causes waste of water resources but also may cause unnecessary damage to equipment and the environment in unaffected areas. Summary of the Invention

[0004] In view of the deficiencies of the prior art, the present invention provides an engineering building detection method and system to solve the problems mentioned in the background art.

[0005] To achieve the above objectives, the present invention is realized through the following technical solutions: An engineering building detection method includes the following steps:

[0006] Step 1: Divide the computer room into multiple monitoring areas in advance and arrange a multi-dimensional sensor network in each monitoring area; collect the real-time temperature of cables, electrical load current, humidity, smoke concentration, and toxic gas concentration in each monitoring area in real time, and establish a monitoring data set;

[0007] Step 2: Install a sprinkler array in each monitoring area. The sprinkler array includes several nozzles arranged to form a first nozzle array, and the initial spacing d of the first nozzle array is set to 5 - 6 meters;

[0008] Step 3: Based on the monitoring data set, calculate the layout tightness index J of the i-th monitoring area in real time i, cable overload factor O i , electrical load change index L i and toxic gas leakage factor C i , and correlate the layout tightness index J i of the i-th monitoring area, the cable overload factor O i , the electrical load change index L i and the toxic gas leakage factor C i to obtain the fire risk index Hz i of the i-th monitoring area, and evaluate to determine whether there is a fire risk in the i-th monitoring area. If there is, classify the fire level and trigger the corresponding fire control instruction;

[0009] Step Four: After triggering the corresponding fire control instruction, collect in real time the diffusion data of the smoke from each monitoring area to other monitoring areas, establish and train a smoke diffusion model to construct the smoke diffusion index I diffuse,i of the i-th monitoring area, and preset a diffusion risk threshold X2. If the smoke diffusion index I diffuse,i of the i-th monitoring area is higher than the diffusion risk threshold X2, trigger a diffusion control instruction and make corresponding corrections to the first fire control instruction and the second fire control instruction.

[0010] Preferably, Step One includes:

[0011] S11. Obtain the building structure drawings of the computer room. The building structure drawings include the floor plan of the computer room, including the positions of walls, electrical equipment, doors and windows, and sprinkler arrays;

[0012] S12. Use CAD software to create an electronic map based on the building structure drawings of the computer room;

[0013] S13. Divide the computer room into multiple monitoring areas in the electronic map, and make digital markings in the electronic map. The multiple monitoring areas include wiring monitoring areas, ceiling monitoring areas, air conditioning and ventilation monitoring areas, passage monitoring areas, and equipment monitoring areas.

[0014] Preferably, Step One further includes:

[0015] S14. Arrange a multi-dimensional sensor network in each monitoring area. The sensors include visual sensors, temperature sensors, current sensors, humidity sensors, gas sensors, and smoke sensors;

[0016] S15. Real-time collect the real-time temperature of the cables, electrical load current, humidity, smoke concentration, and toxic gas concentration in each monitoring area through the multi-dimensional sensor network, and establish a monitoring data set.

[0017] Preferably, Step Two includes:

[0018] S21. Select the type of sprinkler array according to the size of the monitored area and the fire control requirements. The types of sprinkler arrays include linear sprinkler arrays, grid sprinkler arrays, and circular sprinkler arrays;

[0019] S22. Mark the installation positions of the sprinkler arrays in the monitored area to ensure that each sprinkler can cover the target area without affecting the activities of other equipment or personnel;

[0020] S23. Select the first initial spacing of the sprinkler arrays to be 5 - 6 meters, and clean and maintain the sprinkler arrays once a month.

[0021] Preferably, step three includes:

[0022] S31. Extract the occupied area A of the electrical equipment in the i-th monitored area in the monitoring dataset k , the total area A of the area i and the gap distance d between each electrical equipment k . After dimensionless processing, calculate and obtain the layout tightness index J of the i-th monitored area through the following formula i :

[0023]

[0024] Among them, n represents the total number of electrical equipment in the area. By combining the ratio of the occupied area of the electrical equipment to the area of the area and the reciprocal of the average distance between the electrical equipment, the layout tightness index J of the i-th monitored area is obtained i ;

[0025] S32. Extract the real-time temperature T of the cables in the i-th monitored area in the monitoring dataset env and the real-time humidity H env . After dimensionless processing, calculate the temperature influence coefficient αT e and the humidity influence coefficient βR h through the following formula:

[0026]

[0027] Among them, T ref is the reference temperature of the monitored area, and T threshold is the temperature threshold;

[0028] S33. Real-time collect the current value I of the cables in the monitored area t , take the maximum value I of the current in the i-th monitored area max , and calculate and obtain the cable overload coefficient O of the i-th monitored area through the following formula i :

[0029]

[0030] Among them, I rated is the rated current, with the unit of A, which is extracted from the cable specification;

[0031] S34. During the first time interval t1 to the second time interval t2, collect the load power P t of the electrical equipment, and obtain the load fluctuation frequency f t through the Fourier transform equation, and take the maximum value P max and the minimum value P min of the load power, and obtain the real-time load change value ΔP t through the following formula:

[0032] ΔP t = P max - P min ;

[0033] And transform the load power P t of the electrical equipment through the Fourier transform equation, analyze the frequency domain characteristics, and obtain the main frequency f main and the frequency amplitude A f . Combine the real-time load change value ΔP t , after dimensionless processing, calculate and obtain the electrical load change index L i of the i-th monitoring area through the following formula:

[0034]

[0035] Among them, P avg represents the average value of the load power, which is used for normalization to eliminate the influence of different load scales on the results;

[0036] S35. Collect the concentration C t of the toxic gas monitored by the gas sensor, and calculate and obtain the toxic gas leakage coefficient C i of the i-th monitoring area through the following formula:

[0037]

[0038] Among them, the toxic gas leakage coefficient C i > 1, and it is necessary to exhaust immediately and evacuate the staff.

[0039] Preferably, step three further includes:

[0040] S36. The layout tightness index J i of the i-th monitoring area obtained in S31 - S35, the cable overload coefficient O i , the electrical load change index L i and the toxic gas leakage coefficient Ci Correlate them, and calculate the fire risk index Hz of the i-th monitoring area through the following formula i :

[0041] Hz i = J i * w1 + O i * w2 + L i * w3 + C i * w4;

[0042] In the formula, w1, w2, w3, and w4 respectively represent the layout tightness index J of the i-th monitoring area i , the cable overload factor O i , the electrical load change index L i and the toxic gas leakage factor C i weights, 0 < w1 < 1, 0 < w2 < 1, 0 < w3 < 1, 0 < w4 < 1, and w1 + w2 + w3 + w4 = 1.

[0043] Preferably, step three further includes:

[0044] S37. Preset the fire risk threshold X1, and compare and evaluate the fire risk index Hz of the i-th monitoring area i with the risk threshold X to determine whether there is a fire risk in the i-th monitoring area, including:

[0045] If the fire risk index Hz of the i-th monitoring area i ≥ the fire risk threshold X1 × 150%, it means that there is a first-level fire risk in this monitoring area, and issue a first fire control instruction, including: triggering the audible and visual alarm device to remind the on-site personnel, marking the risk area on the monitoring area map, and after powering off the current monitoring area, activating the automatic fire extinguishing equipment in the monitoring area, that is, turning on the sprinkler array, and turning on 81%-90% of the sprinkler heads of the sprinkler array and setting the sprinkler flow rate of the sprinkler array to 1.67 cubic meters per minute;

[0046] If the fire risk threshold X1 ≤ if the fire risk index Hz of the i-th monitoring area i < the fire risk threshold X1 × 150%, it means that there is a second-level fire risk in this monitoring area, and issue a second fire control instruction, including: triggering the audible and visual alarm device to remind the on-site personnel, marking the risk area on the monitoring area map, and after powering off the current monitoring area, activating the automatic fire extinguishing equipment in the monitoring area, that is, turning on the sprinkler array, and turning on 60 - 80% of the sprinkler heads of the sprinkler array and setting the sprinkler flow rate of the sprinkler array to 1.34 cubic meters per minute;

[0047] If the fire risk index Hz of the i-th monitoring area i<The fire risk threshold X1 indicates that there is no fire risk in the monitored area, and continuous monitoring is carried out.

[0048] Preferably, step four includes:

[0049] S41. After triggering the first fire control instruction or the second fire control instruction, collect the diffusion data of the smoke from each monitored area to other monitored areas in real time, and use a smoke sensor to collect the smoke concentration C of the i-th monitored area smoke,i , and obtain the timestamp information;

[0050] S42. For each monitored area i, identify the monitored area number and spatial location in the electronic map, and connect the relative positions and connection relationships between each monitored area. The connection relationships include pipeline connection, pedestrian passage connection or ceiling connection;

[0051] S43. Establish a smoke diffusion model through the finite element model technology. After training, calculate and obtain the smoke concentration change rate ΔC from the i-th monitored area to the j-th monitored area through the following formula smoke :

[0052] ΔC smoke =C smoke,i -C smoke,j ;

[0053] Where, C smoke,i represents the smoke concentration of the i-th monitored area at the same timestamp, and C smoke,j represents the smoke concentration of the j-th monitored area at the same timestamp;

[0054] S44. Use an air velocity sensor to collect the air velocity v of the i-th monitored area flow,i , combined with the smoke concentration change rate ΔC from the i-th monitored area to the j-th monitored area smoke , calculate and obtain the diffusion rate V of the i-th monitored area through the following formula diffuse,i :

[0055]

[0056] Where, D ij represents the Euclidean distance between the i-th monitored area and the j-th monitored area, and δ is the distance attenuation constant;

[0057] S45. Collect the number of sprinkler heads, the opening ratio of sprinkler heads and the sprinkler flow rate of the i-th monitored area, and construct the sprinkler efficiency F of the i-th monitored area through the following formula spray,i :

[0058]

[0059] Where, Nspray,i represents the number of activated sprinklers in the i-th monitoring area, Q spray,i represents the spray flow rate of the current spray array in the i-th monitoring area, N max is the maximum available number of sprinklers, Q max is the maximum spray flow rate; d represents the first initial spacing; ε represents the attenuation constant, which is obtained by fitting through the exponential decay relationship between the experimental spray efficiency and the sprinkler spacing;

[0060] S46. Combine the smoke concentration C of the i-th monitoring area obtained in S41 - S45 smoke,i the change rate of smoke concentration ΔC from the i-th monitoring area to the j-th monitoring area smoke the diffusion rate V of the i-th monitoring area diffuse,i and the spray efficiency F of the i-th monitoring area spray,i , after dimensionless processing, the smoke diffusion index I of the i-th monitoring area is calculated through the following formula diffuse,i :

[0061]

[0062] Among them, j≠1 means not to repeatedly calculate the influence of the i-th monitoring area on its own area. The goal is to calculate the smoke diffusion effect of the i-th monitoring area on other areas.

[0063] Preferably, step four further includes:

[0064] S47. Preset a diffusion risk threshold X2, and compare and evaluate the smoke diffusion index I of the i-th monitoring area diffuse,i with the diffusion risk threshold X2 to judge the smoke diffusion risk:

[0065] If the smoke diffusion index I of the i-th monitoring area diffuse,i > the diffusion risk threshold X2, it means that there is a diffusion risk in the i-th monitoring area, and a diffusion control instruction is triggered, including: closing the ventilation ducts between the i-th monitoring area and the j-th monitoring area until the smoke diffusion index I of the i-th monitoring area diffuse,i returns within the diffusion risk threshold X2 or the fire is extinguished, adjusting the first initial array spacing d of the spray array to 3 - 4 meters, and increasing the ventilation speed of the smoke discharge port to 150% of the initial ventilation speed;

[0066] And correct the first fire control, including: turning on 91% - 100% of the sprinkler heads of the spray array and setting the spray flow rate of the spray array to 1.75 - 1.85 cubic meters per minute, and turning on 61% - 80% of the sprinkler heads of the spray array of the adjacent j-th monitoring area and setting the spray flow rate of the spray array to 1.20 - 1.30 cubic meters per minute to form a first barrier effect;

[0067] And modify the second fire control instruction, including: turning on 81%-90% of the sprinkler heads in the sprinkler array and setting the sprinkler flow rate of the sprinkler array to 1.40-1.50 cubic meters per minute; and turning on 40%-60% of the sprinkler heads in the sprinkler array of the adjacent jth monitoring area and setting the sprinkler flow rate of the sprinkler array to 1.10-1.19 cubic meters per minute to form a second barrier effect;

[0068] If the smoke diffusion index I of the ith monitoring area diffuse,i ≤ the diffusion risk threshold X2, it means that there is no diffusion risk in the ith monitoring area, and continuous monitoring is carried out.

[0069] Engineering building detection system, including:

[0070] Construct an electronic map unit for obtaining the building structure drawings of the computer room and creating an electronic map based on the building structure drawings of the computer room using CAD software;

[0071] A regional division unit for dividing the computer room in the electronic map, and the regional division includes a wiring monitoring area, a ceiling monitoring area, an air conditioning and ventilation monitoring area, a passage monitoring area, and an equipment monitoring area;

[0072] A data acquisition unit for arranging a multi-dimensional sensor network in each monitoring area to collect the real-time temperature of the cable, the electrical load current, humidity, smoke concentration, and toxic gas concentration in each monitoring area in real time, and establishing a monitoring data set;

[0073] A sprinkler array installation unit for installing a sprinkler array in each monitoring area. The sprinkler array includes a number of sprinkler heads arranged to form a first sprinkler head array, and the initial spacing d of the first sprinkler head array is set to 5-6 meters;

[0074] A fire risk analysis unit for calculating the layout tightness index J of the ith monitoring area in real time based on the monitoring data set i , the cable overload factor O i , the electrical load change index L i and the toxic gas leakage coefficient C i , and correlating the layout tightness index J of the ith monitoring area i , the cable overload factor O i , the electrical load change index L i and the toxic gas leakage coefficient C i to obtain the fire risk index Hz of the ith monitoring area i , and evaluating to determine whether there is a fire risk in the ith monitoring area. If there is, divide the fire grade and trigger the corresponding fire control instruction;

[0075] The diffusion analysis unit is used to collect the diffusion data of smoke from each monitoring area to other monitoring areas in real time after the corresponding fire control command is triggered, establish and train the smoke diffusion model to construct the smoke diffusion index I of the i-th monitoring area. diffuse,i ;

[0076] The correction unit is used to preset the diffusion risk threshold X2. If the smoke diffusion index I of the i-th monitoring area diffuse,i If it is higher than the diffusion risk threshold X2, the diffusion control instruction is triggered, and the first fire control instruction and the second fire control instruction are modified accordingly.

[0077] The present invention provides a method and system for detecting engineering buildings, which have the following beneficial effects:

[0078] (1) The present invention arranges a multi-dimensional sensor network to collect multi-dimensional data including real-time cable temperature, electrical load current, humidity, smoke concentration, and toxic gas concentration, and calculates the layout compactness index J i , Cable overload factor O i , Electrical load variation index L i and toxic gas leakage coefficient C i To obtain the fire risk index Hz of the i-th monitoring area i . It is used to evaluate the fire risk level of the monitored area in real time, ensure the accuracy and timeliness of risk identification, and provide a scientific basis for subsequent control instructions.

[0079] (2) The present invention collects smoke diffusion data during the fire process, establishes a smoke diffusion model and trains it in real time, and dynamically calculates the smoke diffusion index. When the smoke diffusion index exceeds the preset diffusion risk threshold X2, the diffusion control instruction is triggered. This function can effectively predict the smoke diffusion path and range, thereby curbing smoke diffusion in the early stages of a fire, reducing the threat of fire to neighboring areas, and significantly improving the ability to control fire diffusion.

[0080] (3) The building detection system of this project, through systematic area division and multi-dimensional sensor network monitoring, can collect environmental and equipment operation data of key areas in real time, improving the early warning capability of fire risks. Combining smoke diffusion analysis and fire risk index assessment, it can accurately judge fire risks and trigger fire control instructions in a targeted manner. In addition, based on the correction mechanism of the smoke diffusion index, the system can dynamically adjust the operating status of sprinkler equipment and ventilation facilities, realizing precise fire control and resource conservation, and effectively improving the fire safety and emergency response efficiency of buildings.

[0081] (4) In fire control, the present invention dynamically adjusts the working state of the sprinkler array through the dual evaluation of the fire level and the diffusion risk, including the number of nozzles, the nozzle spacing, and the sprinkler flow rate. For example, for different fire levels, the opening ratio of the sprinkler array nozzles (such as 81%-100% or 60%-80%) and the corresponding sprinkler flow rate (such as 1.67 cubic meters per minute or 1.34 cubic meters per minute) are respectively set to achieve precise fire extinguishing. This intelligent regulation strategy not only improves the fire extinguishing efficiency but also avoids the waste of water resources and equipment damage in unaffected areas. BRIEF DESCRIPTION OF THE DRAWINGS

[0082] Figure 1 Schematic diagram of the steps of the engineering building detection method of the present invention;

[0083] Figure 2 Schematic diagram of the process of the engineering building detection system of the present invention. DETAILED DESCRIPTION OF THE EMBODIMENTS

[0084] The following will clearly and completely describe the technical solutions in the embodiments of the present invention with reference to the accompanying drawings in the embodiments of the present invention. Obviously, the described embodiments are only a part of the embodiments of the present invention, rather than all the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by those of ordinary skill in the art without creative efforts shall fall within the protection scope of the present invention.

[0085] Embodiment 1

[0086] Please refer to Figure 1 , the present invention provides an engineering building detection method, including the following steps:

[0087] Step 1: The computer room is pre-divided into multiple monitoring areas, and a multi-dimensional sensor network is arranged in each monitoring area; the real-time temperature of the cables, the electrical load current, the humidity, the smoke concentration, and the concentration of toxic gases in each monitoring area are collected in real time, and a monitoring data set is established;

[0088] Step 2: A sprinkler array is installed in each monitoring area. The sprinkler array includes several nozzles arranged to form a first nozzle array, and the initial spacing d of the first nozzle array is set to 5-6 meters;

[0089] Step 3: Based on the monitoring data set, the layout tightness index J i , the cable overload coefficient O i , the electrical load change index L i , and the toxic gas leakage coefficient C i of the i-th monitoring area are calculated in real time, and the layout tightness index J i , the cable overload coefficient O i , the electrical load change index Li and the toxic gas leakage coefficient C i are correlated to obtain the fire risk index Hz of the i-th monitoring area i , and evaluate to determine whether there is a fire risk in the i-th monitoring area. If there is, classify the fire level and trigger the corresponding fire control instruction;

[0090] Step 4. After triggering the corresponding fire control instruction, collect the diffusion data of the smoke from each monitoring area to other monitoring areas in real time, establish and train a smoke diffusion model to construct the smoke diffusion index I of the i-th monitoring area diffuse,i , and preset a diffusion risk threshold X2. If the smoke diffusion index I of the i-th monitoring area diffuse,i is higher than the diffusion risk threshold X2, trigger a diffusion control instruction and make corresponding corrections to the first fire control instruction and the second fire control instruction.

[0091] In this embodiment, the present invention arranges a multi-dimensional sensor network to collect multi-dimensional data including the real-time temperature of the cable, the electrical load current, humidity, smoke concentration, and toxic gas concentration, and calculates the layout tightness index J i , the cable overload coefficient O i , the electrical load change index L i and the toxic gas leakage coefficient C i to obtain the fire risk index Hz of the i-th monitoring area i . It is used to evaluate the fire risk level of the monitoring area in real time, ensure the accuracy and timeliness of risk identification, and provide a scientific basis for subsequent control instructions.

[0092] The present invention collects the smoke diffusion data after a fire occurs, establishes and trains a smoke diffusion model in real time, and dynamically calculates the smoke diffusion index. When the smoke diffusion index exceeds the preset diffusion risk threshold X2, a diffusion control instruction is triggered. This function can effectively predict the smoke diffusion path and range, thereby containing the smoke diffusion in the initial stage of the fire, reducing the threat of the fire to adjacent areas, and significantly improving the fire diffusion control ability.

[0093] During the fire control process, the present invention dynamically corrects the initial fire control instruction by collecting and analyzing the monitoring data in real time to adapt to the development and change of the fire. For example, according to the smoke diffusion situation, adjust the opening number and flow rate of the nozzles of the sprinkler array to ensure the accuracy and effectiveness of the control measures. This real-time dynamic correction mechanism greatly improves the response ability of the fire monitoring system and shortens the response time.

[0094] Embodiment 2

[0095] This embodiment is an explanatory description based on Embodiment 1. Specifically, Step 1 includes:

[0096] S11. Obtain the architectural structure drawings of the computer room. The architectural structure drawings include the floor plan layout of the computer room, including the positions of walls, electrical equipment, doors and windows, and sprinkler arrays, providing an accurate basis for the subsequent creation of the electronic map and area division, avoiding the errors caused by traditional manual observation or surveying, and improving the reliability and adaptability of the monitoring system.

[0097] S12. Use CAD software to create an electronic map based on the architectural structure drawings of the computer room, which can accurately present the internal layout of the computer room and realize the digital reproduction of the floor plan layout of the computer room. The electronic map is convenient for the subsequent division and marking of the monitoring areas, supports the one-to-one correspondence between the monitoring system and the actual environment, and improves the visualization degree and operation convenience of the system.

[0098] S13. Divide the computer room into multiple monitoring areas in the electronic map and perform digital marking in the electronic map. The multiple monitoring areas include wiring monitoring areas, ceiling monitoring areas, air conditioning and ventilation monitoring areas, passage monitoring areas, and equipment monitoring areas. The digital marking assigns independent numbers to each area to ensure the efficiency and accuracy of data collection and monitoring management. The area division also facilitates the targeted design of monitoring strategies and improves the refinement degree of fire monitoring.

[0099] S14. Arrange a multi-dimensional sensor network in each monitoring area. The sensors include visual sensors, temperature sensors, current sensors, humidity sensors, gas sensors, and smoke sensors. The multi-dimensional sensor combination can capture comprehensive data that is difficult to identify by traditional single sensors, providing more dimensional information support for fire risk analysis and improving the accuracy and comprehensiveness of monitoring.

[0100] S15. Real-time collect the cable real-time temperature, electrical load current, humidity, smoke concentration, and toxic gas concentration in each monitoring area through the multi-dimensional sensor network to establish a monitoring data set.

[0101] In this embodiment, the implementation of the above steps forms a set of digital and intelligent fire monitoring front-end systems, which can realize the full-process digital management from data collection to area division, providing efficient technical support for subsequent risk analysis, fire control, and the establishment of diffusion models.

[0102] Embodiment 3

[0103] This embodiment is an explanatory description based on Embodiment 1. Specifically, Step 2 includes:

[0104] S21. Select the sprinkler array type according to the size of the monitoring area and the fire control requirements. The sprinkler array types include linear sprinkler arrays, grid sprinkler arrays, and circular sprinkler arrays.

[0105] S22. Mark the installation positions of the sprinkler arrays within the monitoring area to ensure that each sprinkler can cover the target area without affecting the activities of other equipment or personnel.

[0106] S23. Select the first initial spacing of the sprinkler arrays to be 5 - 6 meters and clean and maintain the sprinkler arrays once a month.

[0107] In this embodiment, by flexibly selecting linear, grid or circular sprinkler arrays, customized layouts can be made according to the shapes and fire risk characteristics of different monitoring areas. For example, linear sprinklers are suitable for narrow areas, grid sprinklers are suitable for large - area regions, and circular sprinklers are suitable for irregular areas. Such a design can significantly improve the uniformity and effectiveness of sprinkler coverage and meet the fire - prevention requirements of complex computer room environments.

[0108] Define the installation positions of the sprinklers to ensure that each target area is covered and avoid monitoring blind spots. At the same time, reasonably plan the installation positions to avoid interference of the sprinklers with the activities of other equipment or personnel, thereby improving the safety and applicability of the installation. This step helps to optimize the accuracy of fire control and reduce unnecessary impacts on non - fire areas.

[0109] Set the first initial spacing of the sprinkler arrays within the range of 5 - 6 meters to ensure the best balance of the sprinkler coverage range, avoiding insufficient coverage due to too large a spacing and preventing equipment waste due to too small a spacing. Regularly cleaning and maintaining the sprinkler arrays can ensure the continuous reliability and spraying efficiency of the sprinkler system and avoid affecting the fire - extinguishing effect due to blocked or damaged sprinklers.

[0110] Embodiment 4

[0111] This embodiment is an explanatory description based on Embodiment 1. Specifically, Step 3 includes:

[0112] S31. Extract the occupied area A of electrical equipment, the total area A of the i - th monitoring area in the monitoring dataset k , and the gap distance d between each electrical equipment i . After dimensionless processing, calculate and obtain the layout tightness index J of the i - th monitoring area through the following formula k : i :

[0113]

[0114] Among them, n represents the total number of electrical equipment in the area. By combining the ratio of the occupied area of electrical equipment to the area of the region and the reciprocal of the average distance between electrical equipment, the layout tightness index J of the i - th monitoring area is obtained. iBy combining the ratio of the occupied area of electrical equipment to the total area of the region with the clearance distance between equipment rooms, the layout density of electrical equipment in the monitored area is comprehensively measured. This indicator helps identify areas with overly tight layouts, thereby reducing the risk of fires caused by poor equipment heat dissipation or overload. The introduction of the layout tightness index provides a scientific basis for optimizing the layout of the computer room.

[0115] S32. Extract the real-time temperature T of the cable in the i-th monitored area from the monitoring dataset env and the real-time humidity H env . After dimensionless processing, calculate the temperature influence coefficient αT through the following formula e and the humidity influence coefficient βR h :

[0116]

[0117] where T ref is the reference temperature of the monitored area, and T threshold is the temperature threshold; by analyzing the changes in temperature and humidity in the monitored area in real time, the temperature influence coefficient can reflect the potential hazards of high-temperature environments on electrical equipment, while the humidity influence coefficient effectively evaluates the impact of humidity on the insulation performance of cables. These parameters provide an accurate reference for judging the fire risk in the area and help take preventive measures such as cooling or dehumidification in advance.

[0118] S33. Real-time collect the current value I of the cable in the monitored area t , take the maximum value I max of the current in the i-th monitored area, and calculate the cable overload coefficient O of the i-th monitored area through the following formula i :

[0119]

[0120] where I rated is the rated current, with the unit of A, extracted from the cable specifications; by real-time monitoring the cable current value and calculating the cable overload coefficient, cables operating overloaded can be effectively identified. Overloaded operation is an important cause of cable heating and fire, and the monitoring of the cable overload coefficient can help give early warnings and take load adjustment measures to reduce the probability of fire.

[0121] S34. During the time from the first interval t1 to the second interval t2, collect the load power P of the electrical equipment t , and obtain the load fluctuation frequency f through the Fourier transform equation t , and take the maximum value P max and the minimum value P min of the load power, and obtain the real-time load change value ΔP through the following formula t :

[0122] ΔP t = P max - P min ;

[0123] And perform a transformation on the load power P of the electrical equipment through the Fourier transform equation t to analyze the frequency domain characteristics and obtain the main frequency f main and the frequency amplitude A f . Combine with the real-time change value ΔP of the load t . After dimensionless processing, calculate and obtain the electrical load change index L of the i-th monitoring area through the following formula i :

[0124]

[0125] where P avg represents the average value of the load power, which is used for normalization to eliminate the influence of different load scales on the results; based on the real-time fluctuation analysis of the load power, combined with the Fourier transform to extract the frequency domain characteristics, and deeply evaluate the instability of the electrical equipment load. The calculation of the load change index can identify potential overload or overheating problems caused by sudden load changes, provide dynamic support for fire risk assessment, and guide the optimization of the equipment operation mode.

[0126] S35. Collect the concentration C of the toxic gas monitored by the gas sensor t , and calculate and obtain the toxic gas leakage coefficient C of the i-th monitoring area through the following formula i :

[0127]

[0128] where the toxic gas leakage coefficient C i > 1, and it is necessary to exhaust immediately and evacuate the staff. By real-time monitoring the concentration of the toxic gas and quickly calculating the leakage coefficient, it is ensured that the exhaust and evacuation instructions are triggered immediately when abnormal concentration is detected. The introduction of this coefficient not only improves the safety of the system but also effectively reduces the risk of personnel poisoning in case of fire.

[0129] S36. Correlate the layout tightness index J i , cable overload coefficient O i , electrical load change index L i and toxic gas leakage coefficient C i obtained in S31 - S35, and calculate and obtain the fire risk index Hz of the i-th monitoring area through the following formula i :

[0130] Hz i = J i * w1 + Oi *w2 + L i *w3 + C i *w4;

[0131] Wherein, w1, w2, w3, and w4 respectively represent the layout tightness index J i of the i-th monitoring area, the cable overload factor O i the electrical load change index L i and the toxic gas leakage factor C i weights, 0 < w1 < 1, 0 < w2 < 1, 0 < w3 < 1, 0 < w4 < 1, and w1 + w2 + w3 + w4 = 1.

[0132] In this embodiment, the layout tightness index J i of the i-th monitoring area, the cable overload factor O i the electrical load change index L i and the toxic gas leakage factor C i are weighted and correlated through a formula to calculate the fire risk index Hz i of the i-th monitoring area. The fire risk index can quantify the fire risk level of the area, transforming the risk assessment from single-dimensional analysis to multi-dimensional comprehensive calculation; the weights are adjusted according to specific scenarios to adapt to the fire risk characteristics of different types of buildings, such as data machine rooms or industrial factories, etc.; the dynamic calculation of the fire risk index combines real-time monitoring data and can accurately reflect the fire hazard situation of each area at the current time point; compared with traditional static assessment methods, this method greatly improves the timeliness of fire risk analysis.

[0133] Embodiment 5

[0134] This embodiment is an explanatory description based on Embodiment 4. Specifically, Step 3 further includes:

[0135] S37. Preset a fire risk threshold X1, and compare and evaluate the fire risk index Hz i of the i-th monitoring area with the risk threshold X to determine whether there is a fire risk in the i-th monitoring area, including:

[0136] If the fire risk index Hz i of the i-th monitoring area ≥ the fire risk threshold X1 × 150%, it indicates that there is a first-level fire risk in this monitoring area, and a first fire control instruction is issued, including: triggering the audible and visual alarm device to remind the on-site personnel, marking the risk area on the monitoring area map, and after powering off the current monitoring area, activating the automatic fire extinguishing equipment in the monitoring area, that is, turning on the spray array and opening 81% - 90% of the nozzle quantity of the spray array and setting the spray flow rate of the spray array to 1.67 cubic meters per minute;

[0137] If the fire risk threshold X1 ≤ the fire risk index Hz of the i-th monitored area i < the fire risk threshold X1 × 150%, indicating that there is a second-level fire risk in this monitored area, and a second-level fire control instruction is issued, including: triggering the audible and visual alarm device to remind the on-site personnel, marking the risk area on the monitored area map, and after power-off control of the current monitored area, activating the automatic fire extinguishing equipment in the monitored area, that is, turning on the sprinkler array, and turning on 60%-80% of the sprinkler heads of the sprinkler array and setting the sprinkler flow rate of the sprinkler array to 1.34 cubic meters per minute;

[0138] If the fire risk index Hz of the i-th monitored area i < the fire risk threshold X1, indicating that there is no fire risk in this monitored area, continue to monitor, monitor continuously in a risk-free state, avoid unnecessary intervention, and ensure the efficient use of system resources.

[0139] In this embodiment, the first-level fire risk and the second-level fire risk are set to achieve hierarchical response: for the first-level fire risk of severe risk, generate the first-level fire control instruction and take stronger fire extinguishing measures (such as higher sprinkler flow rate, more sprinkler heads opened) to turn on 81%-90% of the sprinkler heads, and the sprinkler flow rate is 1.67 cubic meters per minute to maximize the fire extinguishing efficiency and effectively contain the spread of the fire.

[0140] For the second-level fire risk, take medium-level response measures to avoid waste of resources and reduce possible equipment damage at the same time. Turn on 60%-80% of the sprinkler heads, and the sprinkler flow rate is 1.34 cubic meters per minute to effectively control the spread of the fire.

[0141] Under both the first-level fire risk and the second-level fire risk, trigger the audible and visual alarm device and mark the risk area on the electronic map to quickly remind the on-site personnel, avoid personnel staying in the dangerous area, automatically perform power-off control, eliminate the secondary disasters that may be caused by the operation of electrical equipment, and improve the overall fire safety.

[0142] The hierarchical response and precise fire extinguishing measures avoid the waste of resources caused by taking excessive fire extinguishing measures in the case of minor fires, and at the same time reduce the damage to equipment and the environment caused by fire extinguishing measures.

[0143] The fire risk threshold X1 = 0.6;

[0144] The following is an example shown in a figure:

[0145]

[0146]

[0147] Embodiment 6

[0148] This embodiment is an explanatory description carried out in Embodiment 5. Specifically, Step 4 includes:

[0149] S41. After triggering the first fire control instruction or the second fire control instruction, the diffusion data of the smoke from each monitoring area to other monitoring areas is collected in real time. The smoke concentration C of the i-th monitoring area is collected by a smoke sensor smoke,i , and the timestamp information is obtained. By collecting the smoke concentration data of the i-th monitoring area in real time and obtaining the timestamp information, the diffusion process of the smoke in each monitoring area can be dynamically tracked. This provides real-time and accurate data support for a rapid response after a fire occurs, enables timely assessment of the smoke diffusion trend, and provides a basis for subsequent fire control decisions.

[0150] S42. For each monitoring area i, the monitoring area number and spatial position are marked on the electronic map, and the relative positions and connection relationships between the monitoring areas are connected. The connection relationships include pipe connections, pedestrian passage connections, or ceiling connections. Marking the area numbers, spatial positions, and their mutual connection relationships can clearly present the relative positions between the areas during a fire. This step helps to understand the smoke diffusion path, facilitates subsequent diffusion simulation and scheduling of control devices. It ensures that the control instructions can be accurately transmitted to the relevant areas, improving the collaborative efficiency of fire emergency response.

[0151] S43. A smoke diffusion model is established through finite element model technology. After training, the diffusion of fire smoke is usually affected by factors such as air flow, temperature change, structural obstacles (such as walls and doors), and the building's ventilation system. Through the finite element model, these factors are used as input variables to establish the relationship between smoke concentration, time, and space. The continuous physical space is discretized, and the space of the monitoring area is divided into small grid units (usually tetrahedral or hexahedral grids), and each grid unit represents a small part of the physical space. Through this discretization method, the change in smoke concentration in each area can be accurately calculated. According to the simulation time step, the time axis is also discretized into small time periods. This enables the model to gradually simulate the change in smoke concentration over time. The smoke diffusion model is trained using existing smoke diffusion experimental data or historical data to optimize the model's parameters so that it can accurately predict the smoke diffusion behavior. During the training process, the error between the model prediction results and the actual data is analyzed, and the assumptions and parameters in the model are gradually adjusted to make the model better fit the actual situation. The trained model is applied to new test data to verify the accuracy and generalization ability of the model. The cross-validation method is used to evaluate the performance of the model to ensure that it can adapt to different fire scenarios.

[0152] The smoke concentration change rate ΔC from the i-th monitoring area to the j-th monitoring area is calculated by the following formula smoke :

[0153] ΔC smoke = C smoke,i - C smoke,j ;

[0154] where C smoke,i represents the smoke concentration in the i-th monitoring area at the same timestamp, and C smoke,j represents the smoke concentration in the j-th monitoring area at the same timestamp; by using the finite element model technology to calculate the change rate of smoke concentration from the i-th monitoring area to the j-th monitoring area, the diffusion dynamics of smoke can be quantified. This provides mathematical model support for predicting the smoke diffusion speed between different areas, making the fire risk assessment more accurate, helping decision-makers take effective control measures at the initial stage of the fire, and preventing the spread of the fire.

[0155] S44. Collect the air velocity v of the i-th monitoring area using an anemometer flow,i , combine it with the change rate of smoke concentration ΔC from the i-th monitoring area to the j-th monitoring area smoke , and calculate and obtain the diffusion rate V of the i-th monitoring area through the following formula diffuse,i :

[0156]

[0157] where D ij represents the Euclidean distance between the i-th monitoring area and the j-th monitoring area, and δ is the distance attenuation constant; the introduction of the anemometer can directly affect the calculation of the smoke diffusion rate. Combining the change rate of smoke concentration and wind speed data can more accurately estimate the smoke diffusion speed between different areas. This data is the key to formulating a reasonable fire control strategy, which can help the fire response system adjust the resource allocation in real time and improve the flexibility and accuracy of the emergency response.

[0158] S45. Collect the number of sprinkler heads, the opening ratio of sprinkler heads, and the sprinkler flow rate in the i-th monitoring area, and construct the sprinkler efficiency F of the i-th monitoring area through the following formula spray,i :

[0159]

[0160] where N spray,i represents the number of activated sprinkler heads in the i-th monitoring area, Q spray,i represents the sprinkler flow rate of the current sprinkler array in the i-th monitoring area, N max is the maximum available number of sprinkler heads, and Q maxis the maximum spray flow rate; d represents the first initial spacing; ε represents the attenuation constant, which is obtained by fitting the exponential attenuation relationship between the experimental spray efficiency and the nozzle spacing; by collecting the number of nozzles, the opening ratio, and the spray flow rate in the monitoring area in real time, the spray efficiency of each area can be calculated. This calculation helps to dynamically adjust the working state of the spray system according to the actual smoke diffusion situation. By specifically adjusting the number of nozzles and the spray flow rate, unnecessary water resource waste can be avoided, the spread of fire can be maximally controlled, and environmental damage can be reduced.

[0161] S46. Combine the smoke concentration C of the i-th monitoring area obtained from S41 - S45 smoke,i , the change rate of smoke concentration ΔC from the i-th monitoring area to the j-th monitoring area smoke , the diffusion rate V of the i-th monitoring area diffuse,i and the spray efficiency F of the i-th monitoring area spray,i , after dimensionless processing, the smoke diffusion index I of the i-th monitoring area is calculated through the following formula diffuse,i :

[0162]

[0163] where j≠1 means not to repeatedly calculate the influence of the i-th monitoring area on its own area, and the goal is to calculate the smoke diffusion effect of the i-th monitoring area on other areas.

[0164] Combining the smoke concentration, the change rate of smoke concentration, the diffusion rate, and the spray efficiency obtained in the foregoing steps, by calculating the smoke diffusion index, the smoke diffusion effect of each monitoring area can be quantified. This index can accurately reflect the diffusion risk of smoke in different areas, and thus provide real-time and accurate basis for fire control. The dimensionless processed data is more universal and comparable, can realize the comparison between different areas, optimize the resource allocation of spray equipment, and improve the overall effect of fire control.

[0165] In this embodiment, by collecting the smoke concentration, wind speed, diffusion rate, and spray efficiency of each monitoring area in real time, and combining the accurate prediction of smoke diffusion by the finite element model, this method can dynamically monitor the smoke diffusion situation after a fire occurs. Compared with the passive response of traditional systems, the calculation of the smoke diffusion index and the diffusion rate in the steps provides an accurate fire risk assessment, helps to predict the fire diffusion range in real time, and provides a basis for the accurate control of spray equipment. By dynamically adjusting the spray area, flow rate, and nozzle opening ratio, the spray effect can be optimized according to the fire risk level and the smoke diffusion situation, which not only avoids the resource waste and unnecessary environmental damage caused by the "global trigger" mode, but also improves the efficiency and accuracy of fire prevention and control.

[0166] Embodiment 7

[0167] This embodiment is an explanatory description carried out in Embodiment 6. Specifically, Step 4 further includes:

[0168] S47. Preset a diffusion risk threshold X2, and compare and evaluate the smoke diffusion index I of the i-th monitoring area diffuse,i with the diffusion risk threshold X2 to determine the smoke diffusion risk:

[0169] If the smoke diffusion index I of the i-th monitoring area diffuse,i > the diffusion risk threshold X2, it indicates that there is a diffusion risk in the i-th monitoring area, and a diffusion control instruction is triggered, including: closing the ventilation duct between the i-th monitoring area and the j-th monitoring area until the smoke diffusion index I of the i-th monitoring area diffuse,i returns to within the diffusion risk threshold X2 or the fire is extinguished, adjusting the first initial array spacing d of the spray array to 3 - 4 meters, and increasing the ventilation speed of the smoke discharge port to 150% of the initial ventilation speed;

[0170] And correct the first fire control, including: turning on 91% - 100% of the nozzle quantity of the spray array and setting the spray flow rate of the spray array to 1.75 - 1.85 cubic meters per minute, and turning on 61% - 80% of the nozzle quantity of the spray array in the adjacent j-th monitoring area and setting the spray flow rate of the spray array to 1.20 - 1.30 cubic meters per minute to form a first barrier effect;

[0171] And correct the second fire control instruction, including: turning on 81% - 90% of the nozzle quantity of the spray array and setting the spray flow rate of the spray array to 1.40 - 1.50 cubic meters per minute; and turning on 40% - 60% of the nozzle quantity of the spray array in the adjacent j-th monitoring area and setting the spray flow rate of the spray array to 1.10 - 1.19 cubic meters per minute to form a second barrier effect;

[0172] If the smoke diffusion index I of the i-th monitoring area diffuse,i ≤ the diffusion risk threshold X2, it indicates that there is no diffusion risk in the i-th monitoring area, and continuous monitoring is carried out.

[0173] In this embodiment, by calculating and comparing the smoke diffusion index with the diffusion risk threshold X2 in real time, the smoke diffusion risk of each monitoring area can be dynamically and accurately evaluated. This mechanism ensures a real-time response to the spread of fire and avoids delays caused by monitoring lags in traditional systems. The real-time monitoring of the smoke diffusion index can promptly detect the emergence of diffusion risks and quickly take necessary measures when the risks occur, reducing the impact of the spread on other areas.

[0174] For areas with a risk of spread, by triggering diffusion control instructions, such as closing ventilation ducts, adjusting the initial array spacing of the sprinkler array, increasing the ventilation speed of the smoke exhaust outlets, etc., the scope of smoke spread can be maximally controlled. This personalized adjustment helps to control the spread of fire within a local area without causing excessive interference to the entire building. According to different fire spread situations, the system can flexibly adjust the working state of the sprinkler device, rather than a single global switch, avoiding resource waste and ineffective interference to unaffected areas.

[0175] By modifying the first and second fire control instructions, the opening ratio and sprinkler flow rate of the sprinkler array can be adjusted according to different smoke diffusion indices. For example, increasing the opening ratio and flow rate of the nozzles to ensure that the most direct fire area receives sufficient fire extinguishing support, while adjacent areas receive an appropriate barrier effect. This strategy can improve the fire extinguishing efficiency and maximally avoid the spread of fire. Precise correction control is also reflected in the "first barrier effect" and "second barrier effect". Through reasonable adjustment of the sprinkler array, the fire extinguishing synergy between regions is effectively improved, enhancing the coherence and effectiveness of the fire extinguishing process. When the smoke diffusion index of the i-th monitoring area is lower than the threshold X2, the system does not trigger unnecessary control measures but continues to monitor. Through this "demand-based control" strategy, unnecessary resource waste is avoided, such as overstarting the sprinkler system or exhaust equipment. This precise control helps to save water resources, electricity, and the use of other equipment.

[0176] In addition, by optimizing the fire control instructions, the system can achieve multi-region, multi-stage, and multi-level linkage control, thereby improving the fire emergency handling ability of the entire building.

[0177] Example 8

[0178] Please refer to Figure 2 , the engineering building detection system, including:

[0179] Construct an electronic map unit for obtaining the building structure drawings of the computer room and creating an electronic map based on the building structure drawings of the computer room using CAD software;

[0180] A region division unit for dividing the computer room in the electronic map, and the region division includes a wiring monitoring region, a ceiling monitoring region, an air conditioning ventilation monitoring region, a passage monitoring region, and an equipment monitoring region;

[0181] A data acquisition unit for arranging a multi-dimensional sensor network in each monitoring region to collect the real-time temperature of the cable, the electrical load current, humidity, smoke concentration, and toxic gas concentration in each monitoring region in real time, and establishing a monitoring data set;

[0182] A spray array installation unit, used to install a spray array in each monitoring area, the spray array comprising a plurality of spray heads arranged in a manner to form a first spray head array, wherein an initial spacing d of the first spray head array is set to 5-6 meters;

[0183] The fire risk analysis unit is used to calculate the layout tightness index J of the i-th monitoring area in real time based on the monitoring data set. i , Cable overload factor O i , Electrical load variation index L i and toxic gas leakage coefficient C i , and the layout density index J of the i-th monitoring area i , Cable overload factor O i , Electrical load variation index L i and toxic gas leakage coefficient C i Correlate to obtain the fire risk index Hz of the i-th monitoring area i , and evaluate to determine whether there is a fire risk in the i-th monitoring area. If so, classify the fire level and trigger the corresponding fire control command;

[0184] The diffusion analysis unit is used to collect the diffusion data of smoke from each monitoring area to other monitoring areas in real time after the corresponding fire control command is triggered, establish and train the smoke diffusion model to construct the smoke diffusion index I of the i-th monitoring area. diffuse,i ;

[0185] The correction unit is used to preset the diffusion risk threshold X2. If the smoke diffusion index I of the i-th monitoring area diffuse,i If it is higher than the diffusion risk threshold X2, the diffusion control instruction is triggered, and the first fire control instruction and the second fire control instruction are modified accordingly.

[0186] In this embodiment, through systematic area division and multi-dimensional sensor network monitoring, the environment and equipment operation data of key areas can be collected in real time, improving the early warning capability of fire risks. Combining smoke diffusion analysis and fire risk index assessment, it is possible to accurately judge the fire risk and trigger fire control instructions in a targeted manner. In addition, based on the correction mechanism of the smoke diffusion index, the system can dynamically adjust the operating status of sprinkler equipment and ventilation facilities, realize the precision of fire control and resource conservation, and effectively improve the fire safety and emergency response efficiency of buildings.

[0187] The threshold is set to facilitate comparison. The size of the threshold depends on the amount of sample data and the number of bases set by technicians in this field for each group of sample data; as long as it does not affect the proportional relationship between the parameter and the quantized value.

[0188] The above formulas are all obtained by collecting a large amount of data for software simulation and selecting a formula close to the true value. The coefficients in the formula are set by those skilled in the art according to the actual situation. As described above, only the preferred specific embodiments of the present invention are given, but the protection scope of the present invention is not limited thereto. Any person skilled in the art within the technical scope disclosed by the present invention, according to the technical solution and inventive concept of the present invention, makes equivalent substitutions or changes, and should be covered by the protection scope of the present invention.

Claims

1. A method for detecting an engineering building, characterized in that: The following steps are involved: Step 1: Divide the computer room into multiple monitoring areas in advance, and deploy a multi-dimensional sensor network in each monitoring area; Collect the real-time cable temperature, electrical load current, humidity, smoke concentration and toxic gas concentration in each monitoring area in real time to establish a monitoring data set; Step 2: Install a spray array in each monitoring area, the spray array including a plurality of spray heads arranged in a manner to form a first spray head array, and the initial spacing d of the first spray head array is set to 5-6 meters; Step 3: Based on the monitoring data set, calculate the layout density index J of the i-th monitoring area in real time i , Cable overload factor O i , Electrical load variation index L i and toxic gas leakage coefficient C i , and the layout density index J of the i-th monitoring area i , Cable overload factor O i , Electrical load variation index L i and toxic gas leakage coefficient C i Correlate to obtain the fire risk index Hz of the i-th monitoring area i , and evaluate to determine whether there is a fire risk in the i-th monitoring area. If so, classify the fire level and trigger the corresponding fire control command; Step 4: After triggering the corresponding fire control command, real-time data on the diffusion of smoke from each monitoring area to other monitoring areas is collected, and a smoke diffusion model is established and trained to construct the smoke diffusion index I of the i-th monitoring area. diffuse,i , and preset the diffusion risk threshold X2, if the smoke diffusion index I of the i-th monitoring area diffuse,i If it is higher than the diffusion risk threshold X2, the diffusion control instruction is triggered, and the first fire control instruction and the second fire control instruction are modified accordingly.

2. The engineering building detection method according to claim 1, characterized in that: Step one includes: S11. Obtain the architectural structural drawings of the computer room, which include the floor plan of the computer room, including the walls, electrical equipment placement, door and window locations, and sprinkler array locations; S12. Use CAD software to create an electronic map based on the architectural drawings of the computer room; S13. Divide the computer room into areas in the electronic map to obtain multiple monitoring areas, and digitally mark them in the electronic map. The multiple monitoring areas include a wiring monitoring area, a ceiling monitoring area, an air conditioning and ventilation monitoring area, a channel monitoring area, and an equipment monitoring area.

3. The engineering building detection method according to claim 2, characterized in that: Step 1 also includes: S14, deploying a multi-dimensional sensor network in each monitoring area, wherein the sensors include visual sensors, temperature sensors, current sensors, humidity sensors, gas sensors, and smoke sensors; S15. The real-time temperature of the cables, electrical load current, humidity, smoke concentration and toxic gas concentration in each monitoring area are collected in real time through a multi-dimensional sensor network to establish a monitoring data set.

4. The engineering building detection method according to claim 1, characterized in that: Step 2 includes: S21. Select the type of sprinkler array according to the size of the monitoring area and the fire control requirements. The sprinkler array types include linear sprinkler array, grid sprinkler array and circular sprinkler array; S22. Mark the installation location of the sprinkler array in the monitoring area to ensure that each sprinkler can cover the target area without affecting other equipment or personnel activities; S23. Select a first initial spacing of 5-6 meters for the nozzle array, and clean and maintain the nozzle array once a month.

5. The engineering building detection method according to claim 1, characterized in that: Step three includes: S31, extracting the area A occupied by electrical equipment in the ith monitoring area in the monitoring data set k 、Total area of ​​the region A i And the clearance distance d between each electrical device k After dimensionless processing, the layout compactness index J of the i-th monitoring area is calculated by the following formula: i : Where n represents the total number of electrical equipment in the area. The ratio of the area occupied by electrical equipment to the area of ​​the area is combined with the inverse of the average distance between electrical equipment to obtain the layout density index J of the i-th monitoring area. i ; S32, extract the real-time temperature T of the cable in the i-th monitoring area in the monitoring data set env and real-time humidity H env After dimensionless processing, the temperature influence coefficient αT is calculated by the following formula e and humidity influence coefficient βR h : Among them, T ref is the reference temperature of the monitoring area, T threshold is the temperature threshold; S33, real-time acquisition of the current value I of the cable in the monitoring area t , take the maximum value I of the current in the i-th monitoring area max The cable overload coefficient O of the i-th monitoring area is calculated by the following formula: i : Among them, I rated is the rated current in A, extracted from the cable specification; S34, collecting the load power P of the electrical equipment within the first time interval t1 to the second time interval t2 t , and use the Fourier transform equation to obtain the load fluctuation frequency f t , and take the maximum value of load power P max and the minimum load power P min , the real-time load change value ΔP is obtained by the following formula t : ΔP t =P max -P min ; And the load power P of the electrical equipment is calculated by Fourier transform equation t Transform and analyze the frequency domain characteristics to obtain the main frequency f main and frequency amplitude A f , combined with the real-time load change value ΔP t After dimensionless processing, the electrical load change index L of the i-th monitoring area is calculated by the following formula: i : Among them, P avg It represents the average value of load power and is used for normalization to eliminate the impact of different load scales on the results; S35, collect the toxic gas concentration C monitored by the gas sensor t , the toxic gas leakage coefficient C of the i-th monitoring area is calculated by the following formula: i : Among them, the toxic gas leakage coefficient C i >1, exhaust air and evacuate staff immediately.

6. The engineering building detection method according to claim 1, characterized in that: Step three also includes: S36: The layout density index J of the i-th monitoring area obtained in S31-S35 is i , Cable overload factor O i , Electrical load variation index L i and toxic gas leakage coefficient C i The fire risk index Hz of the i-th monitoring area is calculated by the following formula: i : Hz i =J i *w1+O i *w2+L i *w3+C i *w4; Where w1, w2, w3 and w4 represent the layout density index J of the ith monitoring area respectively. i , Cable overload factor O i , Electrical load variation index L i and toxic gas leakage coefficient C i The weights are 0<w1<1, 0<w2<1, 0<w3<1, 0<w4<1, and w1+w2+w3+w4=1.

7. The engineering building detection method according to claim 6, characterized in that: Step three also includes: S37, preset the fire risk threshold X1, and set the fire risk index Hz of the i-th monitoring area i Compare and evaluate with the risk threshold X to determine whether there is a fire risk in the i-th monitoring area, including: If the fire risk index of the i-th monitoring area is Hz i ≥ Fire risk threshold value X1×150%, indicating that the monitoring area has the first fire level risk, and the first fire control instruction is issued, including: triggering the sound and light alarm device to remind the on-site personnel, marking the risk area on the monitoring area map, and after the power is cut off for the current monitoring area, activating the automatic fire extinguishing equipment in the monitoring area, that is, turning on the sprinkler array, turning on 81%-90% of the sprinkler heads in the sprinkler array, and setting the sprinkler flow rate of the sprinkler array to 1.67 cubic meters / minute; If the fire risk threshold X1≤If the fire risk index Hz of the i-th monitoring area i <Fire risk threshold value X1×150%, indicating that the monitoring area has a second fire level risk, and a second fire control instruction is issued, including: triggering the sound and light alarm device to remind the on-site personnel, marking the risk area on the monitoring area map, and after powering off the current monitoring area, activating the automatic fire extinguishing equipment in the monitoring area, that is, turning on the sprinkler array, turning on 60-80% of the number of sprinkler heads in the sprinkler array, and setting the sprinkler array spray flow rate to 1.34 cubic meters / minute; If the fire risk index of the i-th monitoring area is Hz i < Fire risk threshold X1, indicating that there is no fire risk in the monitoring area and continuous monitoring is required.

8. The engineering building detection method according to claim 7, characterized in that: Step 4 includes: S41, after the first fire control command or the second fire control command is triggered, real-time data on the diffusion of smoke from each monitoring area to other monitoring areas is collected, and the smoke concentration C of the i-th monitoring area is collected by using a smoke sensor. smoke,i , and obtain timestamp information; S42, for each monitoring area i, marking the monitoring area number and spatial position in the electronic map, and connecting the relative positions and connection relationships between the monitoring areas, the connection relationship including pipeline connection, pedestrian passage connection or ceiling connection; S43. The smoke diffusion model is established by finite element model technology. After training, the smoke concentration change rate ΔC from the i-th monitoring area to the j-th monitoring area is calculated by the following formula: smoke : ΔC smoke =C smoke,i -C smoke,j ; Among them, C smoke,i represents the smoke concentration of the ith monitoring area at the same time stamp, C smoke,j represents the smoke concentration of the jth monitoring area at the same timestamp; S44, using a wind speed sensor to collect the airflow velocity v in the i-th monitoring area flow,i , combined with the smoke concentration change rate ΔC from the i-th monitoring area to the j-th monitoring area smoke The diffusion rate V of the i-th monitoring area is calculated by the following formula: diffuse,i : Among them, D ij represents the Euclidean distance between the ith monitoring area and the jth monitoring area, and δ is the distance attenuation constant; S45, collect the number of sprinklers, sprinkler opening ratio and sprinkler flow rate in the ith monitoring area, and construct the sprinkler efficiency F of the ith monitoring area through the following formula: spray,i : Among them, N spray,i represents the number of sprinklers activated in the ith monitoring area, Q spray,i represents the spray flow rate of the current spray array in the i-th monitoring area, N max is the maximum number of available nozzles, Q max is the maximum spray flow rate; d represents the first initial spacing; ε represents the decay constant, which is obtained by fitting the exponential decay relationship between the experimental spray efficiency and the nozzle spacing; S46, the smoke concentration C of the i-th monitoring area obtained by combining S41-S45 smoke,i , the smoke concentration change rate from the i-th monitoring area to the j-th monitoring area ΔC smoke , the diffusion rate V of the ith monitoring area diffuse,i And the spraying efficiency F of the i-th monitoring area spray,i After dimensionless processing, the smoke diffusion index I of the i-th monitoring area is calculated by the following formula: diffuse,i : Among them, j≠1 means that the impact of the i-th monitoring area on its own area will not be repeatedly calculated. The goal is to calculate the smoke diffusion effect of the i-th monitoring area on other areas.

9. The engineering building detection method according to claim 1, characterized in that: Step 4 also includes: S47, preset the diffusion risk threshold X2, and set the smoke diffusion index I of the i-th monitoring area diffuse,i Compare and evaluate with the diffusion risk threshold X2 to determine the smoke diffusion risk: If the smoke diffusion index I of the i-th monitoring area diffuse,i >Diffusion risk threshold X2, indicating that there is diffusion risk in the i-th monitoring area, triggering diffusion control instructions, including: closing the ventilation ducts between the i-th monitoring area and the j-th monitoring area until the smoke diffusion index I of the i-th monitoring area is diffuse,i When the fire is restored to within the diffusion risk threshold X2 or the fire is extinguished, the first initial array spacing d of the sprinkler array is adjusted to 3-4 meters, and the ventilation speed of the smoke exhaust port is increased to 150% of the initial ventilation speed; The first fire control is modified, including: opening 91%-100% of the number of sprinklers in the sprinkler array and setting the sprinkler flow rate of the sprinkler array to 1.75-1.85 cubic meters per minute, and opening 61%-80% of the number of sprinklers in the sprinkler array of the adjacent j-th monitoring area and setting the sprinkler flow rate of the sprinkler array to 1.20-1.30 cubic meters per minute, to form a first barrier effect; The second fire control instruction is modified, including: opening 81%-90% of the number of sprinkler heads in the sprinkler array and setting the sprinkler flow rate of the sprinkler array to 1.40-1.50 cubic meters per minute; and opening 40%-60% of the number of sprinkler heads in the sprinkler array of the adjacent j-th monitoring area and setting the sprinkler flow rate of the sprinkler array to 1.10-1.19 cubic meters per minute, forming a second barrier effect; If the smoke diffusion index I of the i-th monitoring area diffuse,i ≤ diffusion risk threshold X2, indicating that there is no diffusion risk in the i-th monitoring area and continuous monitoring is required.

10. An engineering building detection system, comprising the engineering building detection method according to any one of claims 1 to 9, characterized in that: include: Construct an electronic map unit to obtain the architectural drawings of the computer room and use CAD software to create an electronic map based on the architectural drawings of the computer room; A region division unit, used to divide the computer room into regions in the electronic map, wherein the region division includes a wiring monitoring region, a ceiling monitoring region, an air conditioning and ventilation monitoring region, a channel monitoring region, and an equipment monitoring region; A data acquisition unit is used to arrange a multi-dimensional sensor network in each monitoring area, collect the real-time temperature of the cable, electrical load current, humidity, smoke concentration and toxic gas concentration in each monitoring area in real time, and establish a monitoring data set; A spray array installation unit, used to install a spray array in each monitoring area, the spray array comprising a plurality of spray heads arranged in a manner to form a first spray head array, wherein an initial spacing d of the first spray head array is set to 5-6 meters; The fire risk analysis unit is used to calculate the layout tightness index J of the i-th monitoring area in real time based on the monitoring data set. i , Cable overload factor O i , Electrical load variation index L i and toxic gas leakage coefficient C i , and the layout density index J of the i-th monitoring area i , Cable overload factor O i , Electrical load variation index L i and toxic gas leakage coefficient C i Correlate to obtain the fire risk index Hz of the i-th monitoring area i , and evaluate to determine whether there is a fire risk in the i-th monitoring area. If so, classify the fire level and trigger the corresponding fire control command; The diffusion analysis unit is used to collect the diffusion data of smoke from each monitoring area to other monitoring areas in real time after the corresponding fire control command is triggered, establish and train the smoke diffusion model to construct the smoke diffusion index I of the i-th monitoring area. diffuse,i ; The correction unit is used to preset the diffusion risk threshold X2. If the smoke diffusion index I of the i-th monitoring area diffuse,i If it is higher than the diffusion risk threshold X2, the diffusion control instruction is triggered, and the first fire control instruction and the second fire control instruction are modified accordingly.

Images