Danger source dynamic prediction method and system based on cellular automaton, and medium

Abstract

The invention discloses a dangerous source dynamic prediction method and system based on a cellular automaton and a medium. Relates to the technical field of building construction. The method is improved on the basis of a traditional hazard source monitoring technology, the dynamic evolution characteristics of a construction site hazard source are effectively captured by constructing a construction site hazard source CA model, the generation, diffusion and elimination process of the hazard source is simulated, and a theoretical basis is provided for identifying a high-risk area; dynamic evolution is carried out on different construction stages, corresponding management strategies are generated, and safety management is enhanced, so that spreading of dangerous sources can be effectively inhibited; according to the scheme, the cellular automaton theory is applied to the field of construction safety management, and a new hazard source dynamic analysis method is provided.

Description

Technical Field

[0001] The present invention relates to the technical field of building construction, and in particular to a cellular automaton-based dynamic prediction method, system and medium for predicting dangerous sources. Background Art

[0002] With the rapid development of my country's construction industry and the ever-expanding scale of construction projects, construction site safety management faces increasing challenges. Construction sites present a variety of potential hazards, such as falls from heights, impacts from objects, and mechanical injuries. The dynamic nature of these hazards increases the difficulty of safety management. Summary of the Invention

[0003] The technical problem to be solved by the present invention is that traditional monitoring of construction site hazards mainly relies on manual or machine vision detection, which makes it difficult to predict hazards and prevent them in advance; the purpose of the present invention is to provide a dynamic prediction method, system and medium for hazards based on cellular automata, and to improve the method on the basis of traditional hazard source monitoring technology. By constructing a CA model of construction site hazards, the dynamic evolution characteristics of construction site hazards can be effectively captured, and the generation, diffusion and elimination process of hazards can be simulated, providing a theoretical basis for identifying high-risk areas.

[0004] The present invention is achieved through the following technical solutions:

[0005] This solution provides a cellular automation-based dynamic prediction method for hazardous sources, including:

[0006] The construction site is evenly divided into multiple grids to construct a cellular space;

[0007] A CA model for construction site hazards is constructed based on the cellular space. The initial state and state transition rules of each cell in the cellular space are configured according to the actual hazard conditions at the construction site. The state transition rules are used to characterize the generation, spread, and elimination of hazards.

[0008] The current scene status of the construction site is obtained, and the CA model of the construction site hazard source is simulated and analyzed based on the simulation software to obtain the dynamic prediction results of the hazard source of the construction site under different scenarios.

[0009] A further optimization scheme is to evenly divide the construction site into multiple grids to construct a cellular space; including the following method:

[0010] The construction site is divided into m×n two-dimensional grids according to the plane. Each two-dimensional grid represents a cell, and all cells constitute the cellular plane space.

[0011] A further optimization scheme is to construct a CA model of construction site hazards based on cellular space; including the following methods:

[0012] The state set of each cell is configured as Q; Q = {0, 1, 2, 3}; where: Q = 0 indicates a safe state; Q = 1 indicates a potentially dangerous state; Q = 2 indicates a dangerous state; Q = 3 indicates a severely dangerous state;

[0013] Configuring the state transition rules for each cell includes:

[0014] Cells in a safe state can transition to a potentially dangerous state;

[0015] The cells in the safe state within the dangerous state cell domain may be transformed into dangerous state, potentially dangerous state or seriously dangerous state; the cells in the potentially dangerous state within the dangerous state cell domain may be transformed into dangerous state or seriously dangerous state;

[0016] Cells in a serious dangerous state can be transformed into a dangerous state, a potentially dangerous state, or a safe state;

[0017] Cells in a dangerous state can be transformed into a potentially dangerous state or a safe state;

[0018] Cells in a potentially dangerous state can be transformed into a safe state.

[0019] A further optimization solution is that the state transition rule further includes:

[0020] If there is a dangerous state cell or a severely dangerous state cell in the neighborhood of the current safe state cell, the current safe state cell changes to a potentially dangerous state with a probability of P1; otherwise, the current safe state cell changes to a potentially dangerous state with a probability of P0;

[0021] If the number of cells in dangerous state and cells in severe dangerous state in the neighborhood of the current potentially dangerous state cell is greater than or equal to the first domain threshold K1, then the current potentially dangerous state cell changes to a dangerous state; otherwise, the current potentially dangerous state cell remains in a potentially dangerous state with probability P2 and changes to a safe state with probability (1-P2);

[0022] If the number of cells in the serious danger state in the neighborhood of the current dangerous state cell is greater than or equal to the second domain threshold K2, the current dangerous state cell changes to the serious danger state; otherwise, the current dangerous state cell remains in the dangerous state with probability P3 and changes to the potential danger state with probability (1-P3);

[0023] All cells in the severe danger state remain in the severe danger state with probability P4 and change to the dangerous state with probability (1-P4).

[0024] The further optimization solution is: P4>P3>P2>P1>P0.

[0025] The further optimization plan is to simulate and analyze the CA model of construction site hazards based on NetLogo software.

[0026] A further optimization scheme is to simulate and analyze the CA model of construction site hazards based on simulation software to obtain dynamic prediction results of hazard sources in different scenarios at the construction site; including the following methods:

[0027] For normal construction scenarios, high-risk construction phase scenarios, and enhanced safety measures scenarios, initial hazard probabilities and conversion parameters are configured for simulation to obtain dynamic prediction results of hazard sources under different scenarios; the conversion parameters include: probability P0, probability P1, probability P2, probability P3, probability P4, first domain threshold K1, and second domain threshold K2;

[0028] The configuration method of the conversion parameters includes:

[0029] In normal construction scenarios, high-risk construction phase scenarios, and enhanced safety measures scenarios:

[0030] The initial hazard probability is the highest in the high-risk construction phase scenario, and the lowest in the normal construction scenario;

[0031] The probability P0 of the normal construction scenario is the largest, and the probability P0 of the enhanced safety measures scenario is the smallest;

[0032] The probability P1, probability P2, probability P3 and probability P4 of the high-risk construction phase scenario are the largest, and the probability P1, probability P2, probability P3 and probability P4 of the enhanced safety measures scenario are the smallest;

[0033] The first field threshold K1 and the second field threshold K2 of the enhanced safety measures scenario are the largest; the first field threshold K1 and the second field threshold K2 of the high-risk construction phase scenario are the smallest.

[0034] Further optimization scheme also includes the following methods:

[0035] With T as the cycle, the hazard source dense area and hazard source common area are identified according to the number of cells in dangerous state and cells in severe dangerous state, and the first management strategy and the second management strategy are generated respectively.

[0036] This solution provides a cellular automaton-based dynamic prediction system for dangerous sources, which is used to implement the above-mentioned cellular automaton-based dynamic prediction method for dangerous sources; the system includes:

[0037] The first building module is used to evenly divide the construction site into multiple grids to construct a cellular space;

[0038] The second module is used to construct a CA model of construction site hazards based on the cellular space: the initial state and state transition rules of each cell in the cellular space are configured according to the actual hazard conditions at the construction site; the state transition rules are used to characterize the generation, spread and elimination of hazard sources;

[0039] The simulation module is used to obtain the current scene status of the construction site, simulate and analyze the CA model of the construction site hazard source based on the simulation software, and obtain the dynamic prediction results of the hazard source under different scenes of the construction site.

[0040] This solution also provides a computer-readable medium on which a computer program is stored. The computer program is executed by a processor to implement the above-mentioned cellular automation-based dynamic prediction method for dangerous sources.

[0041] Compared with the prior art, the present invention has the following advantages and beneficial effects:

[0042] The present invention provides a cellular automaton-based dynamic prediction method, system and medium for hazardous sources; it improves the method on the basis of traditional hazardous source monitoring technology, and effectively captures the dynamic evolution characteristics of hazardous sources at the construction site by constructing a CA model of hazardous sources at the construction site, simulates the generation, spread and elimination process of hazardous sources, and provides a theoretical basis for identifying high-risk areas; it also dynamically evolves different construction stages and generates corresponding management strategies, and strengthening safety management can effectively suppress the spread of hazardous sources; this scheme applies cellular automaton theory to the field of construction safety management, and provides a new dynamic analysis method for hazardous sources. BRIEF DESCRIPTION OF THE DRAWINGS

[0043] In order to more clearly illustrate the technical solutions of the exemplary embodiments of the present invention, the following briefly introduces the drawings required for use in the examples. It should be understood that the following drawings only illustrate certain embodiments of the present invention and should not be considered as limiting the scope. A person of ordinary skill in the art can also derive other relevant drawings based on these drawings without inventive effort. In the drawings:

[0044] Figure 1 This is a flowchart of the hazard source dynamic prediction method based on cellular automata;

[0045] Figure 2 A schematic diagram of random hazard sources at the construction site is initially set (randomly generated);

[0046] Figure 3 Schematic diagram of the dynamic evolution process of the hazard source 1;

[0047] Figure 4 Schematic diagram of the dynamic evolution process 2 of the hazard source;

[0048] Figure 5 Schematic diagram of the final evolution. DETAILED DESCRIPTION

[0049] In order to make the objectives, technical solutions and advantages of the present invention more clearly understood, the present invention is further described in detail below in conjunction with examples and drawings. The exemplary embodiments of the present invention and their descriptions are only used to explain the present invention and are not intended to limit the present invention.

[0050] In view of this, this solution provides the following embodiments to solve the above technical problems:

[0051] Example 1

[0052] A cellular automaton (CA) is a dynamic model with discrete time, space, and state. It consists of a large number of units (called cells) with the same simple structure, distributed in a regular grid according to certain rules. Each cell continuously updates its own state based on the states of its neighbors and pre-set rules, thereby simulating the dynamic evolution of complex systems.

[0053] This embodiment provides a method for dynamic prediction of dangerous sources based on cellular automata, such as Figure 1 Shown, including:

[0054] Step 1: Divide the construction site into multiple grids to construct a cellular space. This step specifically includes the following methods:

[0055] The construction site is divided into an m×n two-dimensional grid according to the plane. Each two-dimensional grid represents a cell, and all cells constitute the cellular plane space S; S = {(i, j)|1≤i≤m, 1≤j≤n}; where (i, j) represents the cell in the i-th row and j-th column.

[0056] Step 2: Construct a CA model of construction site hazards based on the cellular space: configure the initial state and state transition rules of each cell in the cellular space according to the actual hazard conditions at the construction site; the state transition rules are used to characterize the generation, spread, and elimination of hazards;

[0057] This step specifically includes the following methods:

[0058] The state set of each cell is configured as Q; Q = {0, 1, 2, 3}; where: Q = 0 indicates a safe state; Q = 1 indicates a potentially dangerous state; Q = 2 indicates a dangerous state; Q = 3 indicates a severely dangerous state;

[0059] Configuring the state transition rules for each cell includes:

[0060] Generation of dangerous sources: cells in a safe state can be transformed into a potentially dangerous state;

[0061] Diffusion of dangerous sources: cells in a safe state within the dangerous state cell domain can be transformed into a dangerous state, a potentially dangerous state, or a seriously dangerous state; cells in a potentially dangerous state within the dangerous state cell domain can be transformed into a dangerous state or a seriously dangerous state; the domain can be a von Neumann neighborhood (four adjacent cells above, below, left, and right) or a Moore neighborhood (eight surrounding adjacent cells).

[0062] Cells in a serious dangerous state can be transformed into a dangerous state, a potentially dangerous state, or a safe state;

[0063] Cells in a dangerous state can be transformed into a potentially dangerous state or a safe state;

[0064] Disappearance of danger sources: cells in a potentially dangerous state can be transformed into a safe state.

[0065] Specifically, the state transition rules also include:

[0066] If there is a dangerous state cell or a severely dangerous state cell in the neighborhood of the current safe state cell, the current safe state cell changes to a potentially dangerous state with a probability of P1; otherwise, the current safe state cell changes to a potentially dangerous state with a probability of P0;

[0067] If the number of cells in dangerous state and cells in severe dangerous state in the neighborhood of the current potentially dangerous state cell is greater than or equal to the first domain threshold K1, then the current potentially dangerous state cell changes to a dangerous state; otherwise, the current potentially dangerous state cell remains in a potentially dangerous state with probability P2 and changes to a safe state with probability (1-P2);

[0068] If the number of cells in the serious danger state in the neighborhood of the current dangerous state cell is greater than or equal to the second domain threshold K2, the current dangerous state cell changes to the serious danger state; otherwise, the current dangerous state cell remains in the dangerous state with probability P3 and changes to the potential danger state with probability (1-P3);

[0069] All cells in the severe danger state remain in the severe danger state with probability P4 and change to the dangerous state with probability (1-P4).

[0070] Among them, P4>P3>P2>P1>P0.

[0071] This embodiment adopts Moore's neighborhood, that is, considering the influence of the eight neighboring cells around each cell. The state transition function f is defined as follows:

[0072] S(t+1)(i,j)=f(S(t)(i,j),N(i,j),P)

[0073] Among them, S(t)(i,j) represents the state of cell (i,j) at time t, N(i,j) represents the neighborhood state set of (i,j), and P represents the transformation parameter set.

[0074] The specific state transition rules are as follows:

[0075] (1) If S(t)(i,j) = 0 (safe state): If there is a cell in a dangerous state or a severely dangerous state in the neighborhood, it will be transformed into a potentially dangerous state with probability P1; otherwise, it will be transformed into a potentially dangerous state with a very small probability P0.

[0076] (2) If S(t)(i,j) = 1 (potentially dangerous state): If the number of cells in the neighborhood in dangerous and severe dangerous states is ≥ k1, then the state changes to dangerous; otherwise, the state remains in the potentially dangerous state with probability P2 and changes to the safe state with probability (1-P2).

[0077] (3) If S(t)(i,j) = 2 (dangerous state): If the number of cells in the neighborhood in a serious dangerous state is ≥ k2, then the state changes to a serious dangerous state; otherwise, the state remains dangerous with probability P3 and changes to a potential dangerous state with probability (1-P3).

[0078] (4) If S(t)(i,j)=3 (severe dangerous state): maintain the severe dangerous state with probability P4 and turn to dangerous state with probability (1-P4).

[0079] Step 3: Obtain the current scene status of the construction site, simulate and analyze the CA model of the construction site hazard source based on simulation software, and obtain dynamic prediction results of the hazard source under different scenes of the construction site.

[0080] Conduct simulation analysis on the CA model of construction site hazards based on NetLogo software;

[0081] For normal construction scenarios, high-risk construction phase scenarios, and enhanced safety measures scenarios, initial hazard probabilities and conversion parameters are configured for simulation to obtain dynamic prediction results of hazard sources under different scenarios; the conversion parameters include: probability P0, probability P1, probability P2, probability P3, probability P4, first domain threshold K1, and second domain threshold K2;

[0082] In the specific prediction process, the time interval for cell state updates can be set to hours, days, or weeks according to actual needs.

[0083] The configuration method of the conversion parameters includes:

[0084] In normal construction scenarios, high-risk construction phase scenarios, and enhanced safety measures scenarios:

[0085] The initial hazard probability is the highest in the high-risk construction phase scenario, and the lowest in the normal construction scenario;

[0086] The probability P0 of the normal construction scenario is the largest, and the probability P0 of the enhanced safety measures scenario is the smallest;

[0087] The probability P1, probability P2, probability P3 and probability P4 of the high-risk construction phase scenario are the largest, and the probability P1, probability P2, probability P3 and probability P4 of the enhanced safety measures scenario are the smallest;

[0088] The first field threshold K1 and the second field threshold K2 of the enhanced safety measures scenario are the largest; the first field threshold K1 and the second field threshold K2 of the high-risk construction phase scenario are the smallest.

[0089] Step 4: With T as the period, identify the hazard source dense area and hazard source common area according to the number of cells in dangerous state and cells in severe dangerous state, and generate the first management strategy and the second management strategy respectively.

[0090] Example 2

[0091] This embodiment provides a cellular automaton-based dynamic prediction system for dangerous sources, which is used to implement the cellular automaton-based dynamic prediction method for dangerous sources described in Example 1. The system includes:

[0092] The first building module is used to evenly divide the construction site into multiple grids to construct a cellular space;

[0093] The second building block is used to construct a CA model of construction site hazards based on the cellular space: the initial state and state transition rules of each cell in the cellular space are configured according to the actual hazard conditions at the construction site; the state transition rules are used to characterize the generation, spread, and elimination of hazard sources;

[0094] The simulation module is used to obtain the current scene status of the construction site, simulate and analyze the CA model of the construction site hazard source based on the simulation software, and obtain the dynamic prediction results of the hazard source under different scenes of the construction site.

[0095] Example 3

[0096] This embodiment provides a computer-readable medium having a computer program stored thereon. The computer program is executed by a processor to implement the hazard source dynamic prediction method based on cellular automation as described in Example 1. Specifically, the following steps are performed:

[0097] Step 1: Divide the construction site evenly into multiple grids to construct a cellular space;

[0098] Step 2: Construct a CA model of construction site hazards based on the cellular space: configure the state of each cell in the cellular space and the state transition rules according to the actual situation of the construction site;

[0099] Step 3: Obtain the current scene status of the construction site, simulate and analyze the CA model of the construction site hazard source based on simulation software, and obtain dynamic prediction results of the hazard source under different scenes of the construction site.

[0100] Example 4

[0101] In order to verify the effectiveness of the CA model for construction site hazards, this embodiment selected an actual large-scale construction site as a case study object. The site covers an area of ​​about 40,000 square meters, including the main building construction area, material storage area, mechanical operation area, etc. The construction site is divided into a 50×50 grid, and each cell represents an area of ​​4 square meters; using the Moore Neighborhood model, each cell is affected by the state of the 8 adjacent cells around it, forming a spatial diffusion effect of the hazard source. The neighborhood action radius can be dynamically adjusted according to the characteristics of the construction area;

[0102] The core mechanism includes: a. If more than 4 (including 4) of the 8 adjacent cells around a cell are in a potential hazard source state, the cell is marked as a hazard source state; b. If less than 4 of the 8 adjacent cells around a cell are in a potential hazard source state, the cell is marked as a safe state; the specific evolution process is as follows Figure 2-5 This embodiment sets up three different simulation scenarios to analyze the dynamic evolution of hazard sources under different circumstances:

[0103] Scenario 1, normal construction status: initial danger probability: 5%; conversion parameters: p0 = 0.001, p1 = 0.1, p2 = 0.7, p3 = 0.8, p4 = 0.9, k1 = 3, k2 = 2;

[0104] Scenario 2, high-risk construction phase: initial hazard probability: 15%; conversion parameters: p0 = 0.002, p1 = 0.2, p2 = 0.8, p3 = 0.9, p4 = 0.95, k1 = 2, k2 = 1;

[0105] Scenario 3: Enhanced safety measures are taken: Initial probability of danger: 10%; Conversion parameters: p0 = 0.0005, p1 = 0.05, p2 = 0.6, p3 = 0.7, p4 = 0.8, k1 = 4, k2 = 3;

[0106] For each scenario, 30 time steps (equivalent to one month of construction period) were simulated, and the state distribution at each time step was recorded.

[0107] By running the model and visualizing the results, we found the following:

[0108] 1. Spatial distribution of hazards: Across all scenarios, hazards were observed to be concentrated in certain areas, forming "hazard clusters." This is consistent with actual construction site conditions, where high-risk work areas (such as overhead work and crane operating areas) are often home to concentrations of hazards.

[0109] 2. Temporal evolution of hazard levels: Scenario 1 (normal construction): The number of hazard sources increases slightly in the early stages and then stabilizes; the number of cells in severe hazard states remains low. Scenario 2 (high-risk construction phase): The number of hazard sources increases rapidly and tends to transition to severe hazard states; if no measures are taken, a safety accident may occur. Scenario 3 (enhanced safety measures): Despite the initial high hazard probability, the number of hazard sources decreases significantly over time, indicating the effectiveness of safety measures.

[0110] 3. Hazard spread and elimination dynamics: The model successfully captures the spread of hazards, whereby previously safe areas can become unsafe due to the impact of surrounding hazards. It also simulates the impact of safety measures on hazard elimination.

[0111] 4. Critical behavior: Under certain parameter settings, critical behavior of the system is observed, that is, small parameter changes may lead to significant changes in the overall safety status. This suggests that special attention should be paid to these critical points in actual management. In order to show the results more intuitively, a trend graph of the change of danger level over time is drawn. Initial state, such as Figure 2 As shown in , and the spatial distribution heat map of typical time steps, as shown in Figure 3 As shown, the spatial distribution heat maps of the three scenarios at the initial, mid-term and final time steps are as follows: Figure 4 and Figure 5 These results show that the CA model of construction site hazards can effectively simulate the dynamic evolution process of construction site hazards and provide a quantitative basis for safety management decisions.

[0112] Based on the aforementioned models and simulation results, the effects of different safety management strategies were further evaluated to provide decision support for actual construction safety management.

[0113] Simulation of different safety management measures Three typical safety management strategies were designed and simulated in the model: Strategy A: Routine safety inspection A comprehensive safety inspection is carried out every 5 time steps, and all cells in dangerous state (state 2) and serious dangerous state (state 3) are downgraded by one level. Strategy B: Strengthened management of key areas Identify areas with dense hazard sources and carry out continuous key management of these areas. At each time step, all unsafe state cells in the 10% area with the highest hazard source density are downgraded by one level. Strategy C: Dynamic response mechanism dynamically adjusts the intensity of safety measures according to the current overall risk level. When the proportion of cells in dangerous and serious dangerous states exceeds a certain threshold, the effectiveness of the safety measures is increased (that is, the probability of the state being converted to a safer state is increased).

[0114] Comparative analysis of strategy effects: We performed 100 Monte Carlo simulations for each strategy, with 30 time steps each simulation, and analyzed the following indicators:

[0115] 1. Average danger level: ∑(cell state value) / (total number of cells * number of time steps);

[0116] 2. Proportion of high-risk status: (number of cells in dangerous status + number of cells in severe dangerous status) / total number of cells;

[0117] 3. Safety improvement rate: The difference between the initial high-risk state ratio and the final high-risk state ratio is divided by the number of time steps. The results are shown in the following table:

[0118] Table 1 Comparative analysis of the effects of different security management strategies

[0119] Strategy Average risk level Final high-risk status ratio Safety improvement speed No strategy 0.95 18.3％ -0.02％ Strategy A 0.72 12.1％ 0.21％ Strategy B 0.58 8.7％ 0.32％ Strategy C 0.63 9.5％ 0.29％

[0120] All strategies showed significant improvements compared to the situation with no strategy. Strategy B (strengthening management of key areas) was the most effective in reducing the average level of danger and the proportion of high-risk states. Strategy C (dynamic response mechanism) performed well in terms of the speed of safety improvement and was able to quickly respond to sudden high-risk situations. Although Strategy A (routine safety inspections) was not as effective as B and C overall, it was simple to implement and suitable as a basic safety management measure.

[0121] Based on the above analysis, a multi-strategy combination should be adopted: Strategy B should be the primary management approach, supplemented by the dynamic response mechanism of Strategy C, while maintaining regular and comprehensive inspections of Strategy A. Refined regional management should be implemented: construction areas should be further subdivided, with targeted management parameters established for different types of work areas (such as high-altitude work areas and mechanical operation areas). Data-driven dynamic adjustments should be implemented: actual construction data should be continuously collected and model parameters should be regularly calibrated to make model predictions more accurate. Early warning mechanisms should be established: multi-level early warning thresholds should be established. When the risk level predicted by the model exceeds a certain threshold, the corresponding emergency response measures will be automatically triggered.

[0122] This solution effectively captures the dynamic evolution of construction site hazards through a CA model, simulating their generation, spread, and elimination, demonstrating the emergent behavior of complex systems. Hazards cluster spatially, forming "hazard clusters." This is consistent with actual construction site conditions and provides a theoretical basis for identifying high-risk areas. This solution also dynamically evolves different construction phases and generates corresponding management strategies. Strengthening safety management can effectively curb the spread of hazards. This solution applies cellular automaton theory to construction safety management, providing a new method for dynamic hazard analysis.

[0123] The specific implementation methods described above further illustrate the objectives, technical solutions and beneficial effects of the present invention in detail. It should be understood that the above description is only a specific implementation method of the present invention and is not intended to limit the scope of protection of the present invention. Any modifications, equivalent substitutions, improvements, etc. made within the spirit and principles of the present invention should be included in the scope of protection of the present invention.

Claims

1. A cellular automaton-based dynamic prediction method for hazardous sources, characterized by: include: The construction site is evenly divided into multiple grids to construct a cellular space; A CA model for construction site hazards is constructed based on the cellular space. The initial state and state transition rules of each cell in the cellular space are configured according to the actual hazard conditions at the construction site. The state transition rules are used to characterize the generation, spread, and elimination of hazards. The current scene status of the construction site is obtained, and the CA model of the construction site hazard source is simulated and analyzed based on the simulation software to obtain the dynamic prediction results of the hazard source under different scenes of the construction site.

2. The method for dynamic prediction of dangerous sources based on cellular automata according to claim 1, characterized in that: The construction site is evenly divided into multiple grids to construct a cellular space; Includes methods: The construction site is divided into m×n two-dimensional grids according to the plane. Each two-dimensional grid represents a cell, and all cells constitute the cellular plane space.

3. The method for dynamic prediction of dangerous sources based on cellular automata according to claim 1, characterized in that: The CA model of construction site hazards is constructed based on cellular space; including method: The state set of each cell is configured as Q; Q = {0, 1, 2, 3}; where: Q = 0 indicates a safe state; Q = 1 indicates a potentially dangerous state; Q = 2 indicates a dangerous state; Q = 3 indicates a severely dangerous state; Configuring the state transition rules for each cell includes: Cells in a safe state can transition to a potentially dangerous state; The cells in the safe state within the dangerous state cell domain may be transformed into dangerous state, potentially dangerous state or seriously dangerous state; the cells in the potentially dangerous state within the dangerous state cell domain may be transformed into dangerous state or seriously dangerous state; Cells in a serious dangerous state can be transformed into a dangerous state, a potentially dangerous state, or a safe state; Cells in a dangerous state can be transformed into a potentially dangerous state or a safe state; Cells in a potentially dangerous state can be transformed into a safe state.

4. The method for dynamic prediction of dangerous sources based on cellular automata according to claim 3 is characterized in that: The state transition rules also include: If there is a dangerous state cell or a severely dangerous state cell in the neighborhood of the current safe state cell, the current safe state cell changes to a potentially dangerous state with a probability of P1; otherwise, the current safe state cell changes to a potentially dangerous state with a probability of P0; If the number of cells in dangerous state and cells in severe dangerous state in the neighborhood of the current potentially dangerous state cell is greater than or equal to the first domain threshold K1, then the current potentially dangerous state cell changes to a dangerous state; otherwise, the current potentially dangerous state cell remains in a potentially dangerous state with probability P2 and changes to a safe state with probability (1-P2); If the number of cells in the serious danger state in the neighborhood of the current dangerous state cell is greater than or equal to the second domain threshold K2, the current dangerous state cell changes to the serious danger state; otherwise, the current dangerous state cell remains in the dangerous state with probability P3 and changes to the potential danger state with probability (1-P3); All cells in the severe danger state remain in the severe danger state with probability P4 and change to the dangerous state with probability (1-P4).

5. The method for dynamic prediction of dangerous sources based on cellular automata according to claim 4, characterized in that: P4>P3>P2>P1>P0.

6. The method for dynamic prediction of dangerous sources based on cellular automata according to claim 1, characterized in that: The CA model of construction site hazards is simulated and analyzed based on NetLogo software.

7. The method for dynamic prediction of dangerous sources based on cellular automata according to claim 1, characterized in that: The method is to simulate and analyze the CA model of construction site hazards based on simulation software to obtain dynamic prediction results of hazard sources in different scenarios of the construction site; the method includes: for normal construction scenarios, high-risk construction phase scenarios and enhanced safety measures scenarios, initial hazard probabilities and conversion parameters are configured for simulation respectively to obtain dynamic prediction results of hazard sources in different scenarios; the conversion parameters include: probability P0, probability P1, probability P2, probability P3, probability P4, first domain threshold K1 and second domain threshold K2; The configuration method of the conversion parameters includes: In normal construction scenarios, high-risk construction phase scenarios, and enhanced safety measures scenarios: The initial hazard probability is the highest in the high-risk construction phase scenario, and the lowest in the normal construction scenario; The probability P0 of the normal construction scenario is the largest, and the probability P0 of the enhanced safety measures scenario is the smallest; The probability P1, probability P2, probability P3 and probability P4 of the high-risk construction phase scenario are the largest, and the probability P1, probability P2, probability P3 and probability P4 of the enhanced safety measures scenario are the smallest; The first field threshold K1 and the second field threshold K2 of the enhanced safety measures scenario are the largest; the first field threshold K1 and the second field threshold K2 of the high-risk construction phase scenario are the smallest.

8. The method for dynamic prediction of dangerous sources based on cellular automata according to claim 1, characterized in that: Also includes methods: With T as the cycle, the hazard source dense area and hazard source common area are identified according to the number of cells in dangerous state and cells in severe dangerous state, and the first management strategy and the second management strategy are generated respectively.

9. The cellular automation-based dynamic prediction system for dangerous sources is characterized by: A method for dynamically predicting dangerous sources based on cellular automata according to any one of claims 1 to 8; the system comprises: The first building module is used to evenly divide the construction site into multiple grids to construct a cellular space; The second module is used to construct a CA model of construction site hazards based on the cellular space: the initial state and state transition rules of each cell in the cellular space are configured according to the actual hazard conditions at the construction site; the state transition rules are used to characterize the generation, spread and elimination of hazard sources; The simulation module is used to obtain the current scene status of the construction site, simulate and analyze the CA model of the construction site hazard source based on the simulation software, and obtain the dynamic prediction results of the hazard source under different scenes of the construction site.

10. A computer-readable medium having a computer program stored thereon, characterized in that: The computer program is executed by a processor to implement the cellular automation-based dynamic prediction method for dangerous sources as described in any one of claims 1 to 8.

Images