Intelligent safety management system based on municipal engineering

Abstract

The invention relates to the technical field of safety management, in particular to an intelligent safety management system based on municipal engineering, which comprises a construction area risk dynamic assessment module, a hierarchical dynamic inspection frequency optimization module, an inspection path optimization module, an abnormal behavior monitoring module and a high-risk area early warning module. According to the method, quantitative safety assessment basis is provided for different types of operation by utilizing the environmental data of the construction area, combining with historical accident records and calculating the construction risk level, the inspection frequency is dynamically adjusted according to the risk level of the construction area, the self construction progress and abnormal records, so that the inspection resources are inclined to a high-risk area, and the inspection efficiency is improved. The polling efficiency is improved, meanwhile, the pertinence of safety management is enhanced, in combination with positioning information of constructors and boundary data of a high-risk area, the moving track of the constructors is analyzed, early warning is triggered when the constructors enter the high-risk area, safety risks are intervened in time, the safety supervision capacity of municipal engineering construction is improved, and the probability of construction safety accidents is reduced.

Description

Technical Field

[0001] The present invention relates to the field of safety management technology, and in particular to a smart safety management system based on municipal engineering. Background Art

[0002] The field of safety management technology involves the identification and control of risks across the entire process, encompassing factors such as the operating environment, operational processes, personnel, and equipment and facilities. It aims to reduce the probability of safety incidents through systematic and standardized approaches. This technical area encompasses risk assessment models, safety incident monitoring mechanisms, safety early warning and response processes, safety behavior supervision methods, and the design of information-based safety management systems. Applications include industries such as construction, transportation, energy production, and municipal engineering. The focus is on visualizing operational safety, digitizing management decisions, and intelligently controlling risk, thereby improving safety management efficiency and performance.

[0003] The Municipal Engineering Smart Safety Management System aims to build a safety management platform for municipal construction scenarios. This platform combines multiple functions, including task zoning management, safety hazard identification, safety inspection process control, operational behavior monitoring, and violation tracing, to achieve comprehensive oversight of all safety factors during municipal engineering construction. Its purpose is to enhance construction site safety management capabilities, shorten safety response times, and strengthen personnel safety behavior constraints, thereby reducing the probability of accidents and improving the overall safety assurance capabilities of municipal engineering project execution.

[0004] Traditional management system safety management relies on static standards for safety assessment, which makes it difficult to accurately reflect the risk status of different construction stages, resulting in some high-risk areas not being supervised in a timely manner. Inspection tasks are performed according to fixed cycles, and inspection resources are not flexibly allocated according to actual risk conditions, which easily leads to waste of resources in low-risk areas. Inspection efforts in high-risk areas are insufficient, and the inspection route arrangement fails to fully consider the complex layout of the work site. Inspection personnel may repeat inspections or miss key areas during the execution process, which reduces inspection efficiency. Construction personnel's violations mainly rely on manual observation and post-analysis by on-site management personnel, which results in detection delays and easily leads to safety hazards not being discovered in time. For personnel supervision in high-risk work areas, existing technologies rely on setting physical isolation measures or on-site signs, and fail to effectively track the actual behavior trajectories of personnel, resulting in the risk of illegal entry into high-risk areas still being high, affecting the overall effectiveness of construction safety management. Summary of the Invention

[0005] The purpose of the present invention is to solve the shortcomings of the existing technology and propose a smart safety management system based on municipal engineering.

[0006] In order to achieve the above objectives, the present invention adopts the following technical solutions: a smart safety management system based on municipal engineering, the system comprising:

[0007] The construction area risk dynamic assessment module obtains construction data from municipal engineering construction sites, assigns a corresponding risk level to each construction area based on the construction type of each area and the environmental impact of the current construction area, and obtains construction area risk level classification information;

[0008] The hierarchical dynamic inspection frequency optimization module is based on the risk level classification information of the construction area, calls the construction progress status and inspection abnormality records of the area, and calculates the inspection frequency adjustment coefficient in combination with the current time period, adjusts the inspection frequency of the construction area, and obtains the inspection frequency adjustment information of the construction area;

[0009] The inspection path optimization module calls the inspection frequency adjustment information of the construction area, calls the coordinates of the inspection points in the construction area and the current position of the inspection personnel, adjusts the inspection path, and obtains the inspection task path optimization information;

[0010] The abnormal behavior monitoring module obtains real-time monitoring data from cameras in the municipal construction area based on the inspection task path optimization information, calculates the matching degree between the construction personnel's skeleton coordinate data and known abnormal behaviors, identifies abnormal behaviors based on the matching degree, and immediately notifies safety management personnel and construction personnel to obtain abnormal behavior notification information;

[0011] The improvements of the present invention are that the risk level classification information of the construction area specifically includes the construction process stability influencing factor, the environmental adaptability assessment coefficient and the basic risk numerical weight ratio; the inspection frequency adjustment information of the construction area includes the inspection time interval and the inspection frequency adjustment coefficient; the inspection task path optimization information specifically includes the inspection point sequence number, the inspection personnel path deviation rate and the inspection path dynamic adjustment parameter; the abnormal behavior notification information includes the abnormal behavior matching degree, the abnormal behavior type identification and the abnormal behavior spatiotemporal distribution characteristics.

[0012] The present invention is improved in that the construction area risk dynamic assessment module includes:

[0013] The construction process risk analysis submodule obtains construction data from municipal engineering construction sites, extracts the construction types of each area, including underground pipeline burial, road and bridge construction, and deep foundation pit excavation, calls the safety accident database, screens similar construction records, and obtains the probability and severity of construction accidents using the formula:

[0014]

[0015] Calculate the construction process risk index;

[0016] Among them, Rs Represents the construction process risk index, P i represents the probability of accident occurrence of type i construction technology, S i represents the average severity of accidents of the i-th type of construction technology, n represents the total number of construction technologies in the area, Represents the average probability of construction process accidents;

[0017] The environmental adaptability assessment submodule obtains environmental data of the construction area, including geological type, groundwater level changes, and meteorological conditions, based on the construction process risk index, calculates environmental impact factors, extracts the degree of accident impact under similar environmental conditions in the construction data, calculates the environmental adaptability score, and obtains the construction environment adaptability parameters;

[0018] The risk level classification submodule calculates the risk level value of the construction area based on the construction process risk index and the construction environment adaptability parameter, and classifies the construction area risks according to the municipal engineering risk assessment standard to generate construction area risk level classification information.

[0019] The present invention is improved in that the hierarchical dynamic inspection frequency optimization module includes:

[0020] The construction area status acquisition submodule calls the construction area risk level classification information, extracts the risk level value of the corresponding construction area, calls the construction progress status and inspection abnormality records, extracts the construction status data of the current time period, and establishes the construction area status parameters;

[0021] The inspection frequency adjustment coefficient calculation submodule is based on the construction area status parameters and uses the formula:

[0022]

[0023] Calculate and obtain the inspection frequency adjustment coefficient to obtain the inspection frequency adjustment parameter of the construction area;

[0024] Among them, F adj represents the inspection frequency adjustment coefficient, D represents the risk level value of the construction area, P c Represents the importance coefficient of the current construction progress, A j represents the impact weight of the jth inspection abnormal record, T j Represents the weight of the time period corresponding to the jth inspection abnormal record, N F Represents the total number of inspection exception records in the current time period of the construction area, N s Represents the total number of inspection points within the construction area;

[0025] The inspection frequency adjustment submodule calls the construction area inspection frequency adjustment parameter, combines the basic inspection frequency setting value, calculates the adjusted construction area inspection frequency, adjusts the construction area inspection frequency, and obtains the construction area inspection frequency adjustment information.

[0026] The present invention is improved in that the inspection path optimization module includes:

[0027] The inspection task generation submodule calls the inspection frequency adjustment information of the construction area, combines it with the construction site layout information, screens the locations that need to be inspected in the construction area, and generates an inspection task list based on the regional characteristics. It calls the coordinates of the inspection points in the construction area and the current location of the inspection personnel, matches the inspection points with the dispatchable inspection personnel, and establishes inspection task allocation information;

[0028] The inspection priority calculation submodule extracts the construction area attribute corresponding to the inspection task based on the inspection task allocation information, and calculates the inspection priority by the formula according to the inspection task urgency value and the construction area operation status:

[0029]

[0030] Calculate inspection priorities for construction areas;

[0031] Among them, P A Represents the inspection priority of the construction area, T A is the inspection task urgency value of the construction area, D is the risk level value of the construction area, K A is the environmental impact coefficient of the construction area, V A is the inspection point distribution density in the construction area, M A The number of historical inspections in the construction area;

[0032] The inspection path adjustment submodule calls the inspection task allocation information based on the inspection priority of the construction area, constructs the initial inspection path of the inspection personnel, calculates the path cost between the inspection points, adjusts the order of the inspection points according to the urgency value of the inspection task, optimizes the inspection route, and obtains the inspection task path optimization information.

[0033] The present invention is improved in that the abnormal behavior monitoring module includes:

[0034] The skeleton point coordinate extraction submodule calls the inspection task path optimization information, obtains the real-time monitoring data of the camera in the municipal construction area, extracts the skeleton point coordinate data of the construction personnel based on the construction personnel image data, and obtains the skeleton point coordinate information of the construction personnel;

[0035] The abnormal behavior matching calculation submodule calculates the matching degree between the skeleton point coordinate data and the known abnormal behavior skeleton feature data based on the skeleton point coordinate information of the construction personnel, using the formula:

[0036]

[0037] Calculate the matching degree of abnormal behavior of construction workers;

[0038] Among them, N ab represents the abnormal behavior matching degree, X ak represents the three-dimensional coordinates of the kth skeleton point of the construction worker, X bk The three-dimensional coordinates of the kth skeleton point representing known abnormal behavior, N B Represents the total number of key bone points, W k represents the matching weight of the k-th bone point, D max Represents the maximum bone point displacement reference value;

[0039] The abnormal behavior notification submodule calls the abnormal behavior matching information and determines the abnormal behavior determination status based on the abnormal behavior determination threshold. If the matching degree meets the abnormal behavior determination status, the abnormal behavior category data is obtained, and the safety management personnel and construction personnel are notified immediately to obtain abnormal behavior notification information.

[0040] The present invention is improved in that the system further comprises:

[0041] The high-risk area warning module obtains the abnormal behavior notification information, the construction personnel location information, extracts the high-risk area boundary data, calculates the intersection of the construction personnel's movement trajectory and the high-risk area, and determines whether the construction personnel have entered the high-risk area. If the construction personnel have entered the high-risk area, the construction personnel and safety management personnel are notified immediately to obtain high-risk area warning management information;

[0042] The high-risk area early warning management information specifically refers to the high-risk area intrusion judgment value, personnel movement risk index and early warning level classification.

[0043] The present invention is improved in that the high-risk area early warning module includes:

[0044] The high-risk area data extraction submodule calls the abnormal behavior notification information, obtains the construction personnel's location information, extracts the high-risk area boundary data, including the regional boundary data of the aerial work platform, the tunnel excavation face, and the underground pipe network inspection shaft, and establishes a high-risk area boundary data set;

[0045] The construction worker trajectory calculation submodule extracts the construction worker movement trajectory based on the high-risk area boundary dataset and calculates the intersection of the construction worker movement trajectory and the high-risk area using the formula:

[0046]

[0047] Calculate the status value of construction workers entering high-risk areas;

[0048] Among them, I represents the status value of construction workers entering high-risk areas, d m Represents the distance between the construction worker’s mth position and the nearest high-risk area boundary, d th Represents the distance threshold for determining high-risk areas, N p represents the number of points in the construction workers' moving trajectory, ∈ represents the smoothing factor;

[0049] The early warning notification submodule calls the status value of the construction personnel entering the high-risk area, and determines whether the construction personnel have entered the high-risk area based on the high-risk area early warning threshold. If the construction personnel have entered the high-risk area, the safety management personnel and construction personnel are notified immediately to obtain the high-risk area early warning management information.

[0050] Compared with the prior art, the advantages and positive effects of the present invention are:

[0051] In the present invention, by utilizing the environmental data of the construction area and combining it with historical accident records, the construction risk level is calculated to provide a quantitative safety assessment basis for different types of operations. The inspection frequency is dynamically adjusted according to the risk level of the construction area, its own construction progress and abnormal records, so that inspection resources are tilted towards high-risk areas, improving inspection efficiency while enhancing the pertinence of safety management. In the inspection route planning, the inspection priority and site layout are combined to calculate the optimal inspection route, so that inspection personnel can complete the inspection task with the optimal route, reduce inspection blind spots, match known abnormal behaviors by skeleton point coordinates, realize real-time violation identification, improve the degree of automation of operation behavior supervision, combine construction personnel positioning information and high-risk area boundary data, analyze personnel movement trajectory, trigger early warning when entering high-risk areas, intervene in safety risks in time, improve the safety supervision capacity of municipal engineering construction, and reduce the probability of construction safety accidents. BRIEF DESCRIPTION OF THE DRAWINGS

[0052] Figure 1 is a system flow chart of the present invention;

[0053] Figure 2 This is a flow chart of the construction area risk dynamic assessment module of the present invention;

[0054] Figure 3 This is a flow chart of the hierarchical dynamic inspection frequency optimization module of the present invention;

[0055] Figure 4 This is a flow chart of the inspection path optimization module of the present invention;

[0056] Figure 5 This is a flow chart of the abnormal behavior monitoring module of the present invention;

[0057] Figure 6 This is a flow chart of the high-risk area early warning module of the present invention. DETAILED DESCRIPTION

[0058] In order to make the purpose, technical solutions and advantages of the present invention more clearly understood, the present invention will be further described in detail below with reference to the accompanying drawings and embodiments. It should be understood that the specific embodiments described herein are only used to explain the present invention and are not intended to limit the present invention.

[0059] In the description of the present invention, it should be understood that the terms "length", "width", "up", "down", "front", "back", "left", "right", "vertical", "horizontal", "top", "bottom", "inside", "outside", etc., indicating directions or positional relationships, are based on the directions or positional relationships shown in the accompanying drawings and are only for the convenience of describing the present invention and simplifying the description. They do not indicate or imply that the devices or elements referred to must have a specific direction, be constructed and operate in a specific direction, and therefore should not be understood as limiting the present invention. In addition, in the description of the present invention, the meaning of "plurality" is two or more, unless otherwise clearly and specifically defined.

[0060] See also Figure 1 The present invention provides a technical solution: a smart safety management system based on municipal engineering, the system includes:

[0061] The construction area risk dynamic assessment module obtains construction data from municipal engineering construction sites. Based on the construction type of each area, including underground pipeline burial, road and bridge construction, and deep foundation pit excavation, it extracts the probability and severity of safety accidents recorded in similar construction records. Combined with the environmental impact of the current construction area, it calculates the risk level value of the construction area, assigns a corresponding risk level to each construction area, and obtains the construction area risk level classification information;

[0062] The hierarchical dynamic inspection frequency optimization module is based on the risk level classification information of the construction area, calls the construction progress status and inspection abnormality records of the area, and calculates the inspection frequency adjustment coefficient in combination with the current time period, adjusts the inspection frequency of the construction area, and obtains the inspection frequency adjustment information of the construction area;

[0063] The inspection path optimization module calls the inspection frequency adjustment information of the construction area, obtains the inspection task list and construction site layout information, calculates the inspection priority of the construction area, calls the coordinates of the inspection points in the construction area and the current location of the inspection personnel, adjusts the inspection path, and obtains the inspection task path optimization information;

[0064] The abnormal behavior monitoring module obtains real-time monitoring data from cameras in municipal construction areas based on patrol task path optimization information, extracts the coordinate data of construction personnel skeleton points, calculates the matching degree between the coordinate data of construction personnel skeleton points and known abnormal behaviors, and identifies abnormal behaviors based on the matching degree. If the matching degree reaches the abnormal behavior determination state, the abnormal behavior category data is obtained, and safety management personnel and construction personnel are notified immediately to receive abnormal behavior notification information.

[0065] The high-risk area early warning module obtains abnormal behavior notification information, construction personnel location information, and extracts high-risk area boundary data, including regional boundary data of aerial work platforms, tunnel excavation faces, and underground pipeline inspection wells. It calculates the intersection of construction personnel movement trajectories and high-risk areas, and determines whether construction personnel have entered high-risk areas. If construction personnel have entered high-risk areas, the module immediately notifies the construction personnel and safety management personnel to obtain high-risk area early warning management information.

[0066] The risk level classification information for construction areas specifically includes the construction process stability influencing factor, the environmental adaptability assessment coefficient and the basic risk numerical weight ratio; the inspection frequency adjustment information for construction areas includes the inspection time interval and the inspection frequency adjustment coefficient; the inspection task path optimization information specifically includes the inspection point sequence number, the inspection personnel path deviation rate and the inspection path dynamic adjustment parameter; the abnormal behavior notification information includes the abnormal behavior matching degree, the abnormal behavior type identification and the abnormal behavior spatiotemporal distribution characteristics; the high-risk area early warning management information specifically refers to the high-risk area intrusion judgment value, the personnel movement risk index and the early warning level classification.

[0067] See also Figure 2 ,The construction area risk dynamic assessment module includes:

[0068] The construction process risk analysis submodule obtains construction data from municipal engineering construction sites, extracts the construction types of each area, including underground pipeline burial, road and bridge construction, and deep foundation pit excavation, calls the safety accident database, screens similar construction records, and obtains the probability and severity of construction accidents using the formula:

[0069]

[0070] Calculate the construction process risk index;

[0071] Among them, R s Represents the construction process risk index, P i represents the probability of accident occurrence of type i construction technology, S i represents the average severity of accidents of the i-th type of construction technology, n represents the total number of construction technologies in the area, Represents the average probability of construction process accidents;

[0072] The construction process risk analysis submodule obtains the construction data of the municipal engineering construction site, extracts the construction types of each area, including underground pipeline laying, road and bridge construction, and deep foundation pit excavation, extracts historical construction accident records from the database based on the construction type, and calculates the probability and severity of construction accidents. The probability of accident occurrence P iThe severity of the construction accident S is obtained by dividing the number of accidents of the construction type in the historical data by the total number of construction of this type. i The factors such as casualties, economic losses, and construction delays are quantitatively evaluated based on historical accident data. The calculation formula is S i =w1L+w2C+w3T, where L represents the level of casualties, C represents the amount of economic losses, and T represents the construction delay time. The weights w1, w2, and w3 are set based on the analysis of the proportion of factors affecting construction accidents, specifically from the statistics of the historical accident database. In the accident data of the past five years, casualties accounted for 50% of the impact, economic losses accounted for 30%, and construction delays accounted for 20%. Therefore, w1=0.5, w2=0.3, and w3=0.2 are set. If the data statistics change in the future, the weight values can be adjusted to reflect the actual situation. Based on this, S is calculated. i , substitute the accident probability and accident severity of all construction processes into the construction process risk index calculation formula The average accident probability Assume that a certain area contains three construction processes (underground pipeline burial, road and bridge construction, and deep foundation pit excavation), with accident probabilities of 0.02, 0.05, and 0.08, and accident severity levels of 3.2, 4.5, and 5.7, respectively. The calculation results are Finally, the construction process risk index is obtained.

[0073] The environmental adaptability assessment submodule obtains environmental data of the construction area, including geological types, groundwater level changes, and meteorological conditions, based on the construction process risk index. It calculates environmental impact factors, extracts the impact of accidents under similar environmental conditions in the construction data, calculates environmental adaptability scores, and obtains construction environment adaptability parameters.

[0074] The environmental adaptability assessment submodule is based on the construction process risk index R s , obtain environmental data of the construction area, including geological type, groundwater level changes and meteorological conditions. The environmental impact factor is obtained by weighted summation after standardization of various environmental parameters, such as geological type risk factor E g According to the stratum stability assessment, the value of soft soil layer is set to 0.8, the value of silt layer is set to 0.6, and the value of pebble layer is set to 0.4. The value is set based on the shear strength data of the geological structure. The shear strength of soft soil layer is generally less than 30kPa, the shear strength of silt layer is between 30-60kPa, and the shear strength of pebble layer is greater than 60kPa. Therefore, the risk factor is set according to the inverse relationship of shear strength. The groundwater level change factor E wCalculated based on the range of water level fluctuations, for example, if the water level change is less than 1m, it is set to 0.3, 1-3m is set to 0.5, and greater than 3m is set to 0.8. This factor is set based on the impact of groundwater on the bearing capacity of the foundation. Water level fluctuations less than 1m have little impact on the bearing capacity, so the lowest value is taken. Water level fluctuations of 1-3m have a moderate impact. Fluctuations greater than 3m may lead to foundation softening and increased settlement risks, so the highest value is taken. The meteorological condition factor E m The assessment is based on wind speed, precipitation, and temperature changes. If the wind speed is greater than 15m / s, the precipitation is greater than 50mm, and the temperature change is greater than 20℃, then 0.9 is taken. Otherwise, the environmental adaptability score is reduced in sequence. A s The calculation formula is A s =w4E g +w5E w +w6E m , where weights w4 = 0.4, w5 = 0.3, and w6 = 0.3 are set based on the analysis of the impact of various factors on construction stability. The geological type has the greatest impact on construction stability, so it has the highest weight. The impact of groundwater level change is second, and the impact of meteorological conditions is relatively low but still needs to be considered. Therefore, the weights are set evenly. The geological type of the construction area is set to silt sand layer, the groundwater level change is 2.5m, the meteorological conditions are precipitation of 40mm, wind speed of 12m / s, and temperature change of 15℃. Then A s =0.4××0.6+0.3×0.5+0.3×0.7=0.58, and the construction environment adaptability parameters are calculated.

[0075] The risk level classification submodule calculates the risk level value of the construction area based on the construction process risk index and the construction environment adaptability parameters, and classifies the risks of the construction area according to the municipal engineering risk assessment standards to generate the risk level classification information of the construction area;

[0076] The risk level classification submodule is based on the construction process risk index R s and construction environment adaptability parameter A s , calculate the risk level value of the construction area, the risk level value calculation formula is D = αR s +β(1-A s ), where the weight coefficients α = 0.7 and β = 0.3 are set based on the proportion of construction process risk in the overall construction safety assessment. Construction process risk has a greater impact on the final risk level, so α is taken as a higher value, while environmental adaptability has a relatively low impact on construction safety, so β is taken as a lower value. The specific values are determined by analyzing data from 100 construction areas in the past five years. It was found that the impact of construction process risk on the final accident probability accounted for 70%, and the impact of environmental adaptability accounted for 30%. Based on this, α and β were set, and the R of the construction area was set. s =3.98, A s=0.58, then D = 0.7 × 3.98 + 0.3 × (1-0.58) = 2.79 + 0.126 = 2.916. According to the municipal engineering risk assessment standard, the risk level threshold is set as low risk (D < 2.5), medium risk (2.5 ≤ D < 3.5), and high risk (D ≥ 3.5). The threshold is set based on historical accident statistics. In the past five years, when the accident rate in the construction area was less than 10%, the corresponding D was less than 2.5. When the accident rate was between 10% and 25%, the corresponding D was mainly between 2.5 and 3.5. When the accident rate exceeded 25%, the corresponding D was mainly between 2.5 and 3.5.

[0077] When D is greater than 3.5, the risk level threshold is determined based on this, and the risk level of the area is classified as medium risk. The risk level classification information of the construction area is generated, as shown in Table 1.

[0078] Table 1 Construction area risk level classification table

[0079]

[0080] As shown in Table 1, the risk level value of construction area A is 2.916, which is classified as medium risk. The risk level value of area B is 2.201, which is classified as low risk. The risk level value of area C is 3.498, which is close to the high risk threshold and belongs to the medium risk level.

[0081] See also Figure 3 , the hierarchical dynamic inspection frequency optimization module includes:

[0082] The construction area status acquisition submodule calls the construction area risk level classification information, extracts the risk level value of the corresponding construction area, calls the construction progress status and inspection exception records, extracts the construction status data of the current time period, and establishes the construction area status parameters;

[0083] The construction area status acquisition submodule calls the construction area risk level classification information to obtain the risk level value D of the construction area, calls the construction progress status data to obtain the information of the current construction stage. The construction progress status data includes the completion ratio of the construction plan, the status of the critical path construction nodes, the list of completed construction tasks, etc., and the construction progress importance coefficient P c Calculate the weight of the current construction phase in the overall construction plan. For example, compare the scheduled completion time and impact range of each construction task to determine the impact of the current task on the overall progress. Set the completion rate of the construction plan for a certain construction area to 60%, the current task is on the critical path and involves a construction area of 2,000 square meters, then P cThe weight can be set to 0.75 by calculating the construction area ratio, critical path ratio and other factors. At the same time, the inspection abnormality record is called to extract abnormal data that occurs during the construction process, including safety hazard reports, equipment failure records, quality inspection abnormal items, etc. The inspection abnormality record is given an impact weight A according to the abnormality type. j For example, if the safety hazard is set to 0.8, the equipment failure is set to 0.6, and the quality inspection abnormality is set to 0.4, the impact weight A j The setting of A is based on the impact of inspection anomalies on construction safety. The criteria for determining the impact are: if the anomaly involves a direct threat to construction safety, resulting in casualties or major safety hazards, then A is set. j =0.8-1.0, if the construction machinery fails or the construction progress is hindered, then set A j =0.5-0.7, if the abnormality only affects the construction quality, such as the concrete mix ratio deviation, then set A j =0.3-0.4, the specific weight value is statistically evaluated through historical accident data, and is set based on the normalization of impact level and frequency of occurrence, corresponding to the time period weight T j The time is set based on the proximity between the abnormality occurrence time and the current time. If the abnormality occurs within 24 hours, T j =1, within 48 hours it is set to 0.8, within 72 hours it is set to 0.6, time period weight T j The setting is based on the persistence of the abnormal impact. If the abnormality may have a long-term impact on the construction progress or safety, the time period weight will decrease at a slower rate. On the contrary, if the abnormality has a greater impact on short-term construction, such as equipment failure, then T j The descent rate is accelerated and set to an exponential decrease mode. After calculating the impact weight of each inspection abnormality record, the construction area status parameters are established.

[0084] The inspection frequency adjustment coefficient calculation submodule is based on the construction area status parameters and uses the formula:

[0085]

[0086] Calculate and obtain the inspection frequency adjustment coefficient to obtain the inspection frequency adjustment parameter of the construction area;

[0087] Among them, F adj represents the inspection frequency adjustment coefficient, D represents the risk level value of the construction area, P c Represents the importance coefficient of the current construction progress, T j represents the impact weight of the jth inspection abnormal record, T j Represents the weight of the time period corresponding to the jth inspection abnormal record, N F Represents the total number of inspection exception records in the current time period of the construction area, N s Represents the total number of inspection points within the construction area;

[0088] The inspection frequency adjustment coefficient calculation submodule calculates the inspection frequency adjustment coefficient F based on the construction area status parameters. adj , where the construction area risk level value D represents the basic risk level of the construction area, and the construction progress importance coefficient P c Represents the importance of the current construction stage, and the inspection abnormality record affects the weight A j Combined with the time period weight T j Calculate the total impact of inspection anomalies and the total number of inspection points N s Determined by the inspection layout plan of the construction area, for example, if the risk level value of a construction area is set to D = 3.5, the current construction progress importance coefficient P c =0.75, there are 3 inspection abnormality records in the current time period of the construction area, namely equipment failure, safety hazard and quality inspection abnormality, their impact weights are set to A1=0.6, A2=0.8, A3=0.4, and the corresponding time period weights are set to T1=0.8, T2=1.0, T3=0.6, respectively. The total number of inspection points in the construction area is N s =10, calculate the inspection frequency adjustment coefficient:

[0089]

[0090] Construction progress importance coefficient P c The setting of P is based on the impact scope and time urgency of the construction task. If the current construction task is a key node in the critical path and its delay may lead to an extension of the overall construction period, then P is set. c =0.7-0.9, if the task is a common construction task, then set P c =0.4-0.6, if the task can adjust the construction sequence or there is an alternative plan, then set P c =0.2-0.3. This value can be normalized and set by historical engineering data and construction period calculation to calculate the inspection frequency adjustment coefficient F of the construction area. adj =0.4145, and obtain the inspection frequency adjustment parameter of the construction area.

[0091] The inspection frequency adjustment submodule calls the construction area inspection frequency adjustment parameters, combines the basic inspection frequency setting value, calculates the adjusted construction area inspection frequency, adjusts the construction area inspection frequency, and obtains the construction area inspection frequency adjustment information.

[0092] The inspection frequency adjustment submodule calls the construction area inspection frequency adjustment parameter F adj , combined with the basic inspection frequency setting value F base , calculate the adjusted inspection frequency F of the construction area new , basic inspection frequency F baseBased on the original risk level of the construction area, if the initial inspection frequency of the construction area is set to 4 times a day, the formula for calculating the inspection frequency after adjustment is:

[0093] F new =F base ×(1+F adj );

[0094] Set the initial inspection frequency F of the construction area base =4, calculate:

[0095] F new =4×(1+0.4145)=4×1.4145=5.658;

[0096] After adjustment, the inspection frequency is rounded to 6 times / day, and the inspection frequency adjustment information of the construction area is obtained, as shown in Table 2.

[0097] Table 2 Inspection frequency adjustment table for construction areas

[0098]

[0099] Basic inspection frequency F base The setting is based on the initial risk assessment results of the construction area. If the risk level D of the construction area is lower than 2.5, then F is set. base = 3 times / day, if D is between 2.5-3.5, set F base = 4 times / day, if D is higher than 3.5, set F base = 5-6 times / day. This setting is based on an analysis of historical engineering accidents. Areas with higher risk levels require more frequent inspections to reduce risks. As shown in Table 2, the inspection frequency of construction area A is adjusted from 4 times per day to 6 times, area B is adjusted from 3 times to 4 times, and area C is adjusted from 5 times to 7 times. This ensures that the inspection frequency is dynamically adjusted as the construction status changes.

[0100] See also Figure 4 , the inspection path optimization module includes:

[0101] The inspection task generation submodule calls the inspection frequency adjustment information of the construction area, combines it with the construction site layout information, screens the locations that need to be inspected in the construction area, and generates an inspection task list based on the regional characteristics. It calls the coordinates of the inspection points in the construction area and the current location of the inspection personnel, matches the inspection points with the dispatchable inspection personnel, and establishes the inspection task allocation information.

[0102] The inspection task generation submodule calls the inspection frequency adjustment information of the construction area to obtain the inspection frequency after the adjustment. Combined with the construction site layout information, it determines the distribution of inspection points in the construction area and selects the locations that need to be inspected in the construction area. The screening process is based on the inspection frequency, construction progress status and historical inspection exception records. After the inspection points are selected, an inspection task list is generated based on the functional attributes of the construction area. For example, the inspection tasks in the underground pipeline construction area include checking the sealing of pipeline interfaces, monitoring groundwater infiltration, and detecting pipeline installation errors. The inspection tasks in the road and bridge construction area include checking the stability of the supporting structure, evaluating the quality of the bridge pavement, and monitoring bridge deformation data. The inspection tasks in the deep foundation pit construction area include checking the foundation pit support structure, measuring the displacement of the pit wall, and evaluating the changes in the groundwater level. The coordinates of the inspection points in the construction area and the current location of the inspection personnel are called, the Euclidean distance between the inspection point and the inspection personnel is calculated, and the inspection personnel with a close distance and moderate workload are selected based on the urgency of the inspection task and the task load of the inspection personnel to establish the inspection task allocation information.

[0103] The inspection priority calculation submodule extracts the construction area attributes corresponding to the inspection task based on the inspection task allocation information. According to the inspection task urgency value and the construction area operation status, the formula is used:

[0104]

[0105] Calculate inspection priorities for construction areas;

[0106] Among them, P A Represents the inspection priority of the construction area, T A is the inspection task urgency value of the construction area, D is the risk level value of the construction area, K A is the environmental impact coefficient of the construction area, V A is the inspection point distribution density in the construction area, M A The number of historical inspections in the construction area;

[0107] The inspection priority calculation submodule extracts the construction area attributes corresponding to the inspection task based on the inspection task allocation information, obtains the risk level value D of the construction area, and calculates the urgency value T of the inspection task based on the inspection task allocation information. A The inspection tasks are sorted and the urgency value is set according to the impact range of the inspection tasks, construction progress requirements, and historical abnormal data. For example, the urgency of inspection tasks involving key construction nodes is set higher. The inspection priority calculation formula is:

[0108]

[0109] Among them, the environmental impact coefficient K A It is set according to the geological conditions, weather conditions and surrounding environmental factors of the construction area, such as K in soft soil foundation area.A =0.8, hard rock area K A =0.4, inspection point distribution density V A According to the number of inspection points per unit area, the risk level of a construction area is set to D = 3.5, and the inspection task urgency value is T A =7, environmental impact coefficient K A =0.8, inspection point distribution density V A =15, historical inspection times M A =20, calculate inspection priority:

[0110]

[0111] Calculate the inspection priority P of the construction area A =6.05.

[0112] Inspection task urgency value T A According to the impact of construction progress and abnormal historical data, the interval is set to [1, 10]. When the inspection task involves key construction nodes and there are abnormal inspection records in the past 24 hours, T A Take 9-10, if there is no abnormal record but the construction task is in a critical stage, then T A Take 6-8. If the inspection task is in the regular construction area and there is no abnormal impact, then T A Take 1-5, environmental impact coefficient K A According to the geological conditions of the construction area, the soft soil area is set to 0.8, the sandy soil area is set to 0.6, the pebble area is set to 0.4, and the hard rock area is set to 0.3. The inspection point distribution density V A It is calculated by the total number of inspection points within a unit area, and the value range is [5, 20]. A larger value indicates a denser distribution of inspection points.

[0113] The inspection path adjustment submodule calls the inspection task allocation information based on the inspection priority of the construction area, constructs the initial inspection path of the inspectors, calculates the path cost between the inspection points, adjusts the inspection point order according to the inspection task urgency value, optimizes the inspection route, and obtains the inspection task path optimization information;

[0114] The inspection path adjustment submodule calls the inspection task allocation information based on the inspection priority of the construction area, obtains the initial inspection path of the inspection personnel, and calculates the path cost between the inspection points. The path cost is calculated based on the straight-line distance between the inspection points, terrain obstacles, and the distribution of construction equipment. The Euclidean distances of inspection points A, B, C, and D are set to d respectively. AB =500m,d BC =300m,d CD=450m. The impact of construction equipment causes the inspection route to increase the detour distance. The detour increments are set to 50m, 80m, and 60m respectively. The path cost is calculated as follows:

[0115] d AB ′=500+50=550;

[0116] d BC ′=300+80=380;

[0117] d CD '=450+60=510;

[0118] The total cost of the inspection path is calculated as 550+380+510=1440m. According to the inspection task urgency value, the order of the inspection points is adjusted. For example, if the inspection task urgency is T A >T C >T B >T D , then the inspection path is adjusted to A→C→B→D, the inspection route is optimized, and the inspection task path optimization information is obtained, as shown in Table 3.

[0119] Table 3 Construction area inspection route adjustment table

[0120]

[0121] Inspection path adjustment is based on inspection priority P A Sort by setting the adjustment influence coefficient γ, the value range is [0.1, 0.5], when the inspection priority P A When it is higher than 7.0, γ=0.5, the inspection path adjustment weight is larger, and when P A When it is lower than 4.0, γ = 0.1, and the adjustment has little impact. After the adjustment, the total cost of the inspection path is reduced by 60m, which meets the optimization goal and ensures that inspection points with higher inspection task priorities are inspected first.

[0122] See also Figure 5 ,The abnormal behavior monitoring module includes:

[0123] The skeleton point coordinate extraction submodule calls the inspection task path optimization information, obtains the real-time monitoring data of the camera in the municipal construction area, extracts the skeleton point coordinate data of the construction personnel based on the image data of the construction personnel, and obtains the skeleton point coordinate information of the construction personnel;

[0124] The skeleton point coordinate extraction submodule calls the inspection task path optimization information, obtains the real-time monitoring data of the camera in the municipal construction area, and screens the effective monitoring images within the construction site. The screening process is based on the viewing angle coverage of the monitoring camera, the inspection priority of the construction area and the distribution density of personnel, extracts the image data of the construction personnel, and performs key frame screening on the image data. The screening interval is set according to the monitoring video frame rate. For example, if the frame rate is 30fps, the screening interval is set to 10 frames, that is, one frame is obtained every 0.33 seconds. After the key frame is selected, the skeleton point coordinate data of the construction personnel is extracted. The skeleton point coordinate data is extracted based on the key parts of the construction personnel's body (head, shoulder, elbow, wrist, hip, knee, ankle). The three-dimensional coordinate value of each key point is obtained by calculating the projection relationship, such as the coordinate X of the key point of the head h According to the pixel coordinates (x h ,y h ) Combined with the camera focal length f and height H c Calculation, the formula is Correspondingly, the three-dimensional coordinates of other key skeleton points are calculated using the same method to obtain the coordinate information of the construction workers' skeleton points.

[0125] The abnormal behavior matching calculation submodule calculates the matching degree between the skeleton point coordinate data and the known abnormal behavior skeleton feature data based on the construction personnel skeleton point coordinate information, using the formula:

[0126]

[0127] Calculate the matching degree of abnormal behavior of construction workers;

[0128] Among them, M ab represents the abnormal behavior matching degree, X ak represents the three-dimensional coordinates of the kth skeleton point of the construction worker, X bk The three-dimensional coordinates of the kth skeleton point representing known abnormal behavior, N B Represents the total number of key bone points, W k represents the matching weight of the k-th bone point, D max Represents the maximum bone point displacement reference value;

[0129] The abnormal behavior matching calculation submodule calculates the matching degree between the skeletal point coordinate data and the known abnormal behavior skeletal feature data based on the construction worker's skeletal point coordinate information. The matching process calculates the Euclidean distance between the construction worker's skeletal point coordinate data and the abnormal behavior skeletal feature data in three-dimensional space. The abnormal behavior skeletal feature data includes categories such as falls, illegal hand operations, and dangerous area intrusions. Each type of abnormal behavior defines the reference coordinates of key skeletal points. For example, the characteristic data of fall behavior is set as head height less than 0.5 meters, torso tilt angle greater than 45 degrees, and hand key point height less than 0.3 meters. The matching calculation formula is:

[0130]

[0131] The matching weight W k According to the contribution of the skeleton point to the abnormal behavior, the weight of the skeleton point W k Based on the degree of influence on human posture judgment, the head key points are used to judge actions such as falling, bending, and raising the head, so they are given a higher weight W h =0.4, the key points of the hand are used to judge illegal operations, touching dangerous areas, etc., set W h =0.35, the leg key points are used to judge the states of walking, standing, falling, etc., and assign W h =0.25, the weight distribution is determined based on the proportion of different types of abnormal behaviors in the historical data of the construction site, such as fall-related abnormalities account for 45%, illegal operations account for 35%, and standing posture abnormalities account for 20%, so the weight is distributed according to this proportion. The weights are set to 0.4, 0.35, and 0.25, and the maximum bone point displacement reference value D max It is set to 2 meters. This value is based on the statistical setting of the movement amplitude of construction workers during normal operation in the construction site safety management regulations. The value range is based on the construction environment constraints, such as D in the narrow space construction area. max The distance between the skeleton coordinates of a construction worker and the abnormal behavior skeleton feature data is set to 0.3, 0.4, and 0.6 meters respectively. The calculation is as follows:

[0132]

[0133] Calculated abnormal behavior matching degree M ab =0.795.

[0134] The abnormal behavior notification submodule calls the abnormal behavior matching information and judges the abnormal behavior determination status based on the abnormal behavior determination threshold. If the matching degree meets the abnormal behavior determination status, the abnormal behavior category data is obtained, and the safety management personnel and construction personnel are notified immediately to obtain abnormal behavior notification information.

[0135] The abnormal behavior notification submodule calls the abnormal behavior matching information and judges the abnormal behavior judgment state based on the abnormal behavior judgment threshold. The abnormal behavior judgment threshold M th Set to 0.75. This value is set based on the historical abnormal behavior statistics of the construction site. 1000 cases of abnormal behavior data of the construction site are counted, of which 750 cases have a matching degree greater than 0.75 and are judged as abnormal. The remaining 250 cases have a matching degree less than 0.75 and are judged as normal. This ensures that the misjudgment rate does not exceed 5%. If the abnormal behavior matching degree M ab ≥M th, then it is determined that the construction personnel have abnormal behavior. If the matching degree is lower than the judgment threshold, it is determined to be normal behavior. The abnormal behavior category data, such as falls, illegal operations, dangerous intrusions, etc., are called to obtain the abnormal behavior category information, send abnormal behavior notification information, and notify the construction site safety management personnel, as shown in Table 4.

[0136] Table 4. Detection results of abnormal behavior of construction workers

[0137] Construction worker ID <![CDATA[Bone point matching degree M ab > Abnormal behavior category Determination status P001 0.795 Falling behavior abnormal P002 0.620 - normal P003 0.810 Illegal operations abnormal

[0138] As shown in Table 4, the matching degree of construction workers P001 and P003 exceeds 0.75, which is judged as abnormal behavior. The matching degree of P002 is lower than the threshold and is judged as normal behavior. The system immediately notifies the safety management personnel and records the abnormal behavior category information.

[0139] See also Figure 6 , the high-risk area early warning module includes:

[0140] The high-risk area data extraction submodule calls abnormal behavior notification information, obtains construction personnel location information, extracts high-risk area boundary data, including regional boundary data of aerial work platforms, tunnel excavation faces, and underground pipe network inspection shafts, and establishes a high-risk area boundary data set;

[0141] The high-risk area data extraction submodule calls the abnormal behavior notification information to obtain the construction personnel's location information. The construction personnel's location information is provided by the real-time monitoring system. The positioning device worn by the construction personnel (such as GPS or UWB positioning tags) uploads the current location. The record format is (X p , Y p , Z p ), where X p , Y p , Z p They represent the positions of construction workers in three-dimensional coordinates, call the boundary data of high-risk areas, and filter the boundary data of areas involved in high-altitude operations, tunneling, and underground pipe network maintenance in the construction area. The boundary data of high-risk areas is stored as a polygon coordinate point set {(X b , Y b , Z b )}, where (X b , Y b , Z b) represents the coordinates of the boundary points. When screening boundary points, the regional boundaries are determined based on the characteristics of the construction environment. The boundaries of the aerial work platform are set according to the designed safety height of the work platform. For example, the safety boundary height of the aerial work platform is set to be no less than 2.0 m. The boundaries of the tunnel excavation face are set according to the tunnel cross-sectional dimensions. For example, the diameter of the typical tunnel excavation section is set to range from 5.0 m to 12.0 m, with the actual excavation dimension as the boundary. The boundaries of the underground pipeline inspection well are set according to industry standards. For example, the diameter of the inspection well mouth is set to range from 0.8 m to 1.2 m, and the boundaries are set according to the actual excavation range of the pipeline network in the construction area. After determining the boundary data of each high-risk area, a high-risk area boundary dataset is established, as shown in Table 5.

[0142] Table 5 High-risk area boundary dataset

[0143] Region Type Region ID Boundary point coordinates (some examples) aerial work platforms A1 (10，20，5)，(15，25，5)，(20，20，5) Tunnel boring face B3 (30，40，0)，(35，45，0)，(40，40，0) Underground pipe network inspection well C5 (50，60，-2)，(55，65，-2)，(60，60，-2)

[0144] As shown in Table 5, the high-risk area boundary dataset contains various construction area types and their corresponding boundary coordinates.

[0145] The construction worker trajectory calculation submodule extracts the construction worker movement trajectory based on the high-risk area boundary dataset and calculates the intersection of the construction worker movement trajectory and the high-risk area using the formula:

[0146]

[0147] Calculate the status value of construction workers entering high-risk areas;

[0148] Among them, I represents the status value of construction workers entering high-risk areas, d m Represents the distance between the construction worker’s mth position and the nearest high-risk area boundary, d th Represents the distance threshold for determining high-risk areas, N p represents the number of points in the construction workers' moving trajectory, ∈ represents the smoothing factor;

[0149] The construction worker trajectory calculation submodule extracts the movement trajectory of construction workers based on the high-risk area boundary data set. The movement trajectory data is provided by the construction worker positioning equipment. Each trajectory point records the real-time location of the construction worker in the construction area (X m , Y m , Z m ), after extracting the movement trajectory of the construction workers, the intersection of the movement trajectory of the construction workers and the high-risk area is calculated. The intersection calculation is based on the distance d between the construction workers' trajectory point and the nearest boundary point of the high-risk area m , the calculation formula of the status value I of construction workers entering high-risk areas is:

[0150]

[0151] Among them, dm represents the Euclidean distance from the construction worker's trajectory point to the nearest high-risk area boundary point, d th Represents the distance threshold for determining high-risk areas, N p Represents the number of points in the construction workers' moving trajectory, ∈ is set as a smoothing factor to avoid zero division errors during the calculation process, where the high-risk area determination distance threshold d th It is set according to the construction environment and personnel safety protection requirements. For example, it is required that a 2.0m safety buffer zone be set before construction personnel enter the dangerous area, so d th = 2.0m. This threshold is suitable for early warning judgment when construction personnel approach the boundary. If the construction area involves special dangerous operations (such as deep foundation pit excavation and blasting operations), the threshold can be increased to above 3.0m. The smoothing factor ∈ = 0.01 is set to a small value to prevent d m =d th The denominator is zero, which leads to calculation anomalies. Assume that the movement trajectory of a construction worker contains 4 points, and the distances from the nearest high-risk area boundary point are 1.5m, 1.8m, 2.2m, and 2.5m respectively. Then calculate:

[0152]

[0153] The calculated status value of construction workers entering high-risk areas is I = 0.696.

[0154] The early warning notification submodule calls the status value of construction personnel entering high-risk areas, and judges whether the construction personnel have entered high-risk areas based on the high-risk area early warning threshold. If the construction personnel have entered high-risk areas, the safety management personnel and construction personnel will be notified immediately to obtain high-risk area early warning management information.

[0155] The early warning notification submodule calls the status value of construction personnel entering the high-risk area, and judges whether the construction personnel have entered the high-risk area based on the high-risk area early warning threshold. th According to the historical construction safety data analysis, the behavioral characteristics of construction workers entering high-risk areas are considered. If a construction worker approaches a high-risk area many times but does not enter, the warning threshold is lowered to increase the alarm sensitivity. Referring to the data analysis, when I th When ≤0.5, the warning sensitivity is higher, but the false alarm rate increases, so I th =0.5, ensuring that the warning range is moderate. If construction workers enter the high-risk area, that is, I≥I th , then an early warning is triggered, the safety management personnel and construction personnel are notified, and the early warning information is recorded, as shown in Table 6.

[0156] Table 6 High-risk area warning notification information

[0157] Construction worker ID Entering high-risk area status value I Warning status P001 0.696 Early Warning P002 0.300 normal P003 0.810 Early Warning

[0158] As shown in Table 6, the status values of construction workers P001 and P003 exceed 0.5, triggering a high-risk area warning. The status value of construction worker P002 is lower than the threshold and is judged to be in a normal state. The system immediately notifies relevant personnel to handle the situation.

[0159] The above are merely preferred embodiments of the present invention and do not limit the present invention in any other form. Any technician familiar with the profession may use the technical content disclosed above to change or modify it into an equivalent embodiment with equivalent changes and apply it to other fields. However, any simple modification, equivalent change and modification made to the above embodiment based on the technical essence of the present invention without departing from the content of the technical solution of the present invention shall still fall within the scope of protection of the technical solution of the present invention.

Claims

1. A smart safety management system based on municipal engineering, characterized by: The system comprises: The construction area risk dynamic assessment module obtains construction data from municipal engineering construction sites, assigns a corresponding risk level to each construction area based on the construction type of each area and the environmental impact of the current construction area, and obtains construction area risk level classification information; The hierarchical dynamic inspection frequency optimization module is based on the risk level classification information of the construction area, calls the construction progress status and inspection abnormality records of the area, and calculates the inspection frequency adjustment coefficient in combination with the current time period, adjusts the inspection frequency of the construction area, and obtains the inspection frequency adjustment information of the construction area; The inspection path optimization module calls the inspection frequency adjustment information of the construction area, calls the coordinates of the inspection points in the construction area and the current position of the inspection personnel, adjusts the inspection path, and obtains the inspection task path optimization information; The abnormal behavior monitoring module obtains real-time monitoring data from cameras in the municipal construction area based on the inspection task path optimization information, calculates the matching degree between the construction personnel's skeletal point coordinate data and known abnormal behaviors, identifies abnormal behaviors based on the matching degree, and immediately notifies safety management personnel and construction personnel to obtain abnormal behavior notification information.

2. The municipal engineering intelligent safety management system according to claim 1 is characterized in that: The construction area risk level classification information specifically includes the construction process stability influencing factor, the environmental adaptability assessment coefficient and the basic risk numerical weight ratio; the construction area inspection frequency adjustment information includes the inspection time interval and the inspection frequency adjustment coefficient; the inspection task path optimization information specifically includes the inspection point sequence number, the inspection personnel path deviation rate and the inspection path dynamic adjustment parameter; the abnormal behavior notification information includes the abnormal behavior matching degree, the abnormal behavior type identification and the abnormal behavior spatiotemporal distribution characteristics.

3. The municipal engineering intelligent safety management system according to claim 1 is characterized in that: The construction area risk dynamic assessment module includes: The construction process risk analysis submodule obtains construction data from municipal engineering construction sites, extracts the construction types of each area, including underground pipeline burial, road and bridge construction, and deep foundation pit excavation, calls the safety accident database, screens similar construction records, and obtains the probability and severity of construction accidents using the formula: Calculate the construction process risk index; Among them, R s Represents the construction process risk index, P i represents the probability of accident occurrence of type i construction technology, S i represents the average severity of accidents of type i construction process, n represents the total number of construction processes in the area, and P represents the average probability of construction process accidents; The environmental adaptability assessment submodule obtains environmental data of the construction area, including geological type, groundwater level changes, and meteorological conditions, based on the construction process risk index, calculates environmental impact factors, extracts the degree of accident impact under similar environmental conditions in the construction data, calculates the environmental adaptability score, and obtains the construction environment adaptability parameters; The risk level classification submodule calculates the risk level value of the construction area based on the construction process risk index and the construction environment adaptability parameter, and classifies the construction area risks according to the municipal engineering risk assessment standard to generate construction area risk level classification information.

4. The municipal engineering intelligent safety management system according to claim 1 is characterized in that: The hierarchical dynamic inspection frequency optimization module includes: The construction area status acquisition submodule calls the construction area risk level classification information, extracts the risk level value of the corresponding construction area, calls the construction progress status and inspection abnormality records, extracts the construction status data of the current time period, and establishes the construction area status parameters; The inspection frequency adjustment coefficient calculation submodule is based on the construction area status parameters and uses the formula: Calculate and obtain the inspection frequency adjustment coefficient to obtain the inspection frequency adjustment parameter of the construction area; Among them, F adj represents the inspection frequency adjustment coefficient, D represents the risk level value of the construction area, P c Represents the importance coefficient of the current construction progress, A j represents the impact weight of the jth inspection abnormal record, T j Represents the weight of the time period corresponding to the jth inspection abnormal record, N F Represents the total number of inspection exception records in the current time period of the construction area, N s Represents the total number of inspection points within the construction area; The inspection frequency adjustment submodule calls the construction area inspection frequency adjustment parameter, combines the basic inspection frequency setting value, calculates the adjusted construction area inspection frequency, adjusts the construction area inspection frequency, and obtains the construction area inspection frequency adjustment information.

5. The municipal engineering intelligent safety management system according to claim 1 is characterized in that: The inspection path optimization module includes: The inspection task generation submodule calls the inspection frequency adjustment information of the construction area, combines it with the construction site layout information, screens the locations that need to be inspected in the construction area, and generates an inspection task list based on the regional characteristics. It calls the coordinates of the inspection points in the construction area and the current location of the inspection personnel, matches the inspection points with the dispatchable inspection personnel, and establishes inspection task allocation information; The inspection priority calculation submodule extracts the construction area attribute corresponding to the inspection task based on the inspection task allocation information, and calculates the inspection priority by the formula according to the inspection task urgency value and the construction area operation status: Calculate inspection priorities for construction areas; Among them, P A Represents the inspection priority of the construction area, T A is the inspection task urgency value of the construction area, D is the risk level value of the construction area, K A is the environmental impact coefficient of the construction area, V A is the inspection point distribution density in the construction area, M A The number of historical inspections in the construction area; The inspection path adjustment submodule calls the inspection task allocation information based on the inspection priority of the construction area, constructs the initial inspection path of the inspection personnel, calculates the path cost between the inspection points, adjusts the order of the inspection points according to the urgency value of the inspection task, optimizes the inspection route, and obtains the inspection task path optimization information.

6. The municipal engineering intelligent safety management system according to claim 1 is characterized in that: The abnormal behavior monitoring module includes: The skeleton point coordinate extraction submodule calls the inspection task path optimization information, obtains the real-time monitoring data of the camera in the municipal construction area, extracts the skeleton point coordinate data of the construction personnel based on the construction personnel image data, and obtains the skeleton point coordinate information of the construction personnel; The abnormal behavior matching calculation submodule calculates the matching degree between the skeleton point coordinate data and the known abnormal behavior skeleton feature data based on the skeleton point coordinate information of the construction personnel, using the formula: Calculate the matching degree of abnormal behavior of construction workers; Among them, M ab represents the abnormal behavior matching degree, X ak represents the three-dimensional coordinates of the kth skeleton point of the construction worker, X bk The three-dimensional coordinates of the kth skeleton point representing known abnormal behavior, N B Represents the total number of key bone points, W k represents the matching weight of the k-th bone point, D max Represents the maximum bone point displacement reference value; The abnormal behavior notification submodule calls the abnormal behavior matching information and determines the abnormal behavior determination status based on the abnormal behavior determination threshold. If the matching degree meets the abnormal behavior determination status, the abnormal behavior category data is obtained, and the safety management personnel and construction personnel are notified immediately to obtain abnormal behavior notification information.

7. The municipal engineering intelligent safety management system according to claim 1 is characterized in that: The system further comprises: The high-risk area warning module obtains the abnormal behavior notification information, the construction personnel location information, extracts the high-risk area boundary data, calculates the intersection of the construction personnel's movement trajectory and the high-risk area, and determines whether the construction personnel have entered the high-risk area. If the construction personnel have entered the high-risk area, the construction personnel and safety management personnel are notified immediately to obtain high-risk area warning management information; The high-risk area early warning management information specifically refers to the high-risk area intrusion judgment value, personnel movement risk index and early warning level classification.

8. The municipal engineering intelligent safety management system according to claim 7 is characterized in that: The high-risk area early warning module includes: The high-risk area data extraction submodule calls the abnormal behavior notification information, obtains the construction personnel's location information, extracts the high-risk area boundary data, including the regional boundary data of the aerial work platform, the tunnel excavation face, and the underground pipe network inspection shaft, and establishes a high-risk area boundary data set; The construction worker trajectory calculation submodule extracts the construction worker movement trajectory based on the high-risk area boundary dataset and calculates the intersection of the construction worker movement trajectory and the high-risk area using the formula: Calculate the status value of construction workers entering high-risk areas; Among them, I represents the status value of construction workers entering high-risk areas, d m Represents the distance between the construction worker’s mth position and the nearest high-risk area boundary, d th Represents the distance threshold for determining high-risk areas, N p represents the number of points in the construction workers' moving trajectory, ∈ represents the smoothing factor; The early warning notification submodule calls the status value of the construction personnel entering the high-risk area, and determines whether the construction personnel have entered the high-risk area based on the high-risk area early warning threshold. If the construction personnel have entered the high-risk area, the safety management personnel and construction personnel are notified immediately to obtain the high-risk area early warning management information.

Images