Job superposition risk quantitative early warning method and system fusing three-dimensional GIS and multi-dimensional dynamic data

Abstract

The invention discloses an operation superposition risk quantitative early warning method and system fusing three-dimensional GIS and multi-dimensional dynamic data, and belongs to the technical field of industrial internet and safety production. The method comprises the following steps: acquiring multi-source operation permission data including three-dimensional space coordinates, operation types and initial risk values; mapping an operation entity in the three-dimensional GIS model, and constructing a dynamic risk influence body (Risk Volume) which deforms along with environmental parameters in combination with real-time environmental field data (wind speed, wind direction, gas density and the like); detecting whether the risk influence bodies of different operation entities have a three-dimensional space intersection or not by using a hierarchical collision detection mechanism (bounding box coarse screening and grid actuarial); if the intersection exists, obtaining an interaction influence parameter based on a semantic interaction relationship of the job type, and calculating a superposition risk quantized value by using a preset composite risk assessment logic; and when the numerical value exceeds the limit, triggering early warning containing the three-dimensional positioning information. According to the method, the three-dimensional dynamic diffusion model and the spatial index algorithm based on the physical law are introduced, so that the problem that a planar GIS cannot identify the three-dimensional crossing operation risk and the bottleneck of mass data calculation performance is solved, and accurate quantification and visualization of the dynamic superposition risk of a complex industrial site are realized.

Description

Technical Field

[0001] This invention relates to the fields of industrial internet, safety production monitoring and geographic information technology, and in particular to a method and system for quantifying and warning of operational risks by integrating 3D GIS and multi-dimensional dynamic data. Background Technology

[0002] In high-risk industries such as petrochemicals and power energy, site work management (PTW) is the core of safe production. However, existing work management systems and risk warning technologies have the following significant shortcomings:

[0003] First, it lacks a three-dimensional spatial dimension. Existing systems mostly rely on two-dimensional plane maps for distance calculations, failing to identify the vertical overlap risks between "hot work at heights" and "work in confined spaces below." This planar risk control logic easily leads to missed detections, resulting in serious safety hazards on site.

[0004] Second, the risk model is static. Risk areas are usually simplified into fixed circular buffer zones, which cannot be dynamically adjusted in conjunction with real-time physical parameters such as wind speed, wind direction, and gas density. In windy conditions, the actual risk range downwind often exceeds the fixed circle, while upwind areas may be falsely reported, resulting in low accuracy and a lack of scientific rigor in early warnings.

[0005] Third, there is a conflict between computing power and real-time performance. As factories expand, the computational cost of pairwise comparisons between massive numbers of work points (such as tens of thousands of points) increases exponentially (O(N^2)). Traditional three-dimensional Boolean operations are extremely expensive and cannot meet the demand for second-level real-time early warning.

[0006] Fourth, there is a lack of visual, quantitative, and evidentiary methods. Existing systems mostly display conflicts in text or two-dimensional list formats, lacking intuitive, three-dimensional representations, making it difficult for managers to quickly understand the severity of risks. Furthermore, after an incident, there is a lack of ability to dynamically recreate the three-dimensional scene based on a timeline, leading to difficulties in incident evidence collection and liability tracing. Summary of the Invention

[0007] To address the aforementioned technical problems, this invention provides a method and system for quantifying and early warning of operational risks by integrating 3D GIS and multidimensional dynamic data.

[0008] The present invention adopts the following technical solution:

[0009] A method for quantifying and early warning of operational risks by integrating 3D GIS and multidimensional dynamic data, executed by a computer system, includes the following steps:

[0010] Acquire first work permit data and second work permit data, the data including the three-dimensional spatial coordinates of the work point, the work type identifier and the initial risk characterization value;

[0011] Based on the aforementioned three-dimensional spatial coordinates, the operation entity is mapped and generated in the three-dimensional GIS virtual scene;

[0012] Real-time environmental field data (including wind speed, wind direction, gas density, etc.) around the work site are acquired, and combined with the work type and initial risk characterization value, a dynamic risk impact body is constructed. The construction process uses a three-dimensional fluid diffusion algorithm or a geometric transformation algorithm to generate the risk impact body into an irregular geometric body that deforms with environmental parameters, such as a plume model or an inverted cone.

[0013] A hierarchical collision detection mechanism is used to detect whether there is spatial intersection of risk-affecting bodies: first, axis-aligned bounding boxes (AABB) or spatial indexes are used for initial screening, and then mesh-based precise three-dimensional Boolean operations are performed on potential conflict combinations.

[0014] If there is an intersection, the interaction impact parameters are obtained based on the preset risk interaction association model, and the superimposed risk quantification value is calculated in combination with the initial risk value. When the threshold is exceeded, an early warning containing three-dimensional positioning information is triggered.

[0015] Furthermore, this invention also provides a risk quantification and early warning system for operation overlay that integrates 3D GIS and multi-dimensional dynamic data, comprising: a data perception layer for collecting height information and environmental sensor data; a 3D computing engine for constructing a dynamic risk impact body and performing hierarchical spatial conflict detection; a risk quantification engine for calculating composite risk values ​​based on semantic interaction analysis or AI models; and an early warning interaction terminal for rendering the dynamic risk body and highlighting conflict areas in a digital twin interface.

[0016] The beneficial effects of this invention are as follows:

[0017] 1. Achieved precise risk control in three dimensions: By introducing three-dimensional GIS coordinates and entity mapping, it breaks through the limitations of traditional two-dimensional maps and can accurately identify the risks of three-dimensional cross-operations in the vertical direction, such as "hot work at height" and "ground leakage", effectively eliminating spatial blind spots.

[0018] 2. Improved accuracy of early warnings in dynamic environments: The system abandons the static circular risk zone and innovatively introduces a dynamic deformation model (Risk Volume) based on environmental fields (wind speed, wind direction, gravity, and gas density). In windy or gas leak scenarios, the risk-affected volume automatically extends downwind or deforms according to gravity, ensuring that the definition of the risk range conforms to physical laws and significantly reducing the false alarm rate.

[0019] 3. Solved the performance bottleneck of large-scale data computing: A hierarchical collision detection mechanism of "bounding box initial screening + grid fine calculation" and spatial indexing technology were adopted. When handling massive concurrent jobs, it can quickly eliminate the vast majority of unrelated entities and perform high-precision calculations only on a very small number of potential conflicts. Thus, while ensuring calculation accuracy, the system response time is controlled at the millisecond level, meeting the requirements of real-time monitoring.

[0020] 4. Achieved quantification and visual evidence collection of risks: Through a composite risk assessment logic, qualitative risk descriptions are transformed into quantitative superimposed risk entropy values, and combined with volumetric rendering technology, the shape and overlap of the risk body are intuitively displayed in a 3D interface. This visualized dynamic mask effect not only provides managers with intuitive decision-making basis, but also provides tamper-proof digital evidence for post-accident analysis. Attached Figure Description

[0021] Figure 1 This is the overall flowchart of the method of the present invention;

[0022] Figure 2 This is a schematic diagram illustrating the principle of how dynamic risk-affected bodies deform with wind speed;

[0023] Figure 3 This is a flowchart illustrating the hierarchical collision detection mechanism (bounding box and mesh).

[0024] Figure 4 This is a schematic diagram of the system architecture of the present invention. Detailed Implementation

[0025] The technical solutions of the present invention will be clearly and completely described below with reference to the accompanying drawings in the embodiments of the present invention.

[0026] Example 1: Risk Early Warning Process for Three-Dimensional Cross-Operations

[0027] This example demonstrates a complete early warning process using a chemical plant maintenance scenario as an example.

[0028] Step 1: Data Acquisition. The system retrieves two concurrent job data entries from the Job Permit (PTW) system:

[0029] Task A: Type is "Level 1 Hot Work (Welding)", spatial coordinates are (X=100, Y=200, Z=12), that is, located on the third floor of the pipe gallery, initial risk value R1=0.8.

[0030] Task B: Type is "Pipeline Open (containing benzene medium)", spatial coordinates are (X=105, Y=205, Z=0), that is, located on the ground, initial risk value R2=0.6.

[0031] Step 2: Environmental Perception and Entity Construction. The system reads data from the on-site micro-weather station, showing that the current wind is from the northwest with a speed of 6 m / s.

[0032] For task A, considering the influence of gravity and wind on spark splash, the system constructs an inverted cone tilted downwind (southeast) as its dynamic risk impact body.

[0033] For task B, considering that benzene vapor has a higher density than air and is affected by wind, a flat elliptical fan-shaped column extending downwind along the ground is constructed as its dynamic risk impact body.

[0034] Step 3: Spatial Detection. The 3D calculation engine detects that the bottom of the inverted cone and the sector cylinder have overlapping volumes in the spatial coordinate system.

[0035] Step 4: Quantitative Calculation. The system queries the risk interaction correlation model and identifies the interaction effect parameter K=5.0 between "open flame" and "flammable gas" (indicating extremely high risk).

[0036] Substituting into the compound risk formula, we get: R_total = 0.5 * sqrt(0.8^2 + 0.6^2) + 0.5 * (0.8 * 0.6 * 5.0) = 0.5 * 1.0 + 1.2 = 1.7.

[0037] Step 5: Warning Trigger. Since the calculated result of 1.7 far exceeds the preset high-risk threshold (e.g., 0.9), the system immediately triggers a red alarm and highlights the conflict area on the 3D map, indicating that "sparks from a high-altitude fire may fall into the leaking gas cloud below."

[0038] Example 2: Algorithm for Constructing Dynamic Risk Impact Entities

[0039] This embodiment details the deformation logic in claim 2 to ensure that the model conforms to physical laws.

[0040] 1. Plume Model: Applicable to the risk of toxic / flammable gas leaks.

[0041] Let the leak point be the origin (0,0,0), and the wind direction be the positive x-axis. The boundary of the risk body is defined by the Gaussian diffusion equation, and its cross-sectional radius r varies with distance x according to the functional relationship: r(x) = a * x^b, where a and b are constants related to atmospheric stability. As the wind speed increases, the plume model stretches in the x-axis direction and contracts in the y and z-axis directions, simulating the effect of gas being dispersed by the wind.

[0042] 2. Projected cylindrical model: suitable for situations involving falling objects or sparks from heights.

[0043] Let the height of the work point be H, and the ground projection point be P0. In a windless state, the risk body is a cylinder with P0 as the center and radius R. In a windy state (wind speed V), the centerline of the risk body shifts towards the wind direction as the height decreases, with a shift of D = f(V,H), eventually forming a tilted, twisted cylinder.

[0044] Example 3: Performance Optimization of Hierarchical Collision Detection Mechanism

[0045] This embodiment details the calculation strategy in claim 3, which is used to solve the problem of concurrent calculation of tens of thousands of job points.

[0046] Phase 1: Building the Spatial Index (Broad Phase).

[0047] The system initializes an octree structure, dividing the three-dimensional space of the entire factory area into hierarchical cubic nodes.

[0048] For each generated dynamic risk impact body, calculate its minimum bounding rectangle—the axis-aligned bounding box (AABB). Insert the AABB into the corresponding node of the octree.

[0049] The system traverses the octree to quickly filter out job pairs with overlapping AABB values, forming a "potential conflict list". For jobs with non-overlapping AABB values ​​(which constitute the vast majority), subsequent calculations are skipped.

[0050] Phase Two: Narrow Phase (Precise Calculation)

[0051] For each pair of jobs in the "Potential Conflict List", load its fine triangular mesh model (Mesh).

[0052] The Separating Axis Theorem (SAT) algorithm is used to detect whether two convex polyhedral meshes intersect. If they intersect, Boolean operations are used to calculate the volume V_intersect of the intersection.

[0053] This mechanism ensures 100% detection accuracy while reducing computational complexity from O(N^2) to close to O(N), thus meeting real-time requirements.

[0054] Example 4: Visual Rendering and Evidence Collection

[0055] In response to the warning signal, the system performs the following rendering operations in the 3D GIS interface:

[0056] 1. Draw a semi-transparent outer shell for the risk body, with the color mapped according to the individual risk value (e.g., yellow represents moderate risk, and red represents high risk).

[0057] 2. For the calculated intersection volume, volume rendering technology is used to render it as a bright pulse flashing state (e.g., dark red, frequency 2Hz) to intuitively show the core area of ​​risk superposition.

[0058] 3. Generate a floating Billboard label at the center of the intersection area, displaying the text: "3D Cross Conflict: Hot Work + Leak, Overlay Value 1.7".

[0059] This obvious visual feature not only facilitates real-time monitoring but also has significant value in post-incident evidence collection. When reviewing historical data, the wind-distorted light mask and the red conflict core area on the screen can clearly demonstrate the physical mechanism of the accident and management oversights.

[0060] The above description is only a preferred embodiment of the present invention and is not intended to limit the present invention. Any modifications, equivalent substitutions, improvements, etc., made within the spirit and principles of the present invention should be included within the protection scope of the present invention.

Claims

1. A method for quantifying and early warning of operational risks by integrating 3D GIS and multidimensional dynamic data, characterized in that, This method is executed by a computer system and includes the following steps: Acquire first work permit data and second work permit data, the data including the three-dimensional spatial coordinates (X, Y, Z) of the work point, the work type identifier and the initial risk characterization value; Based on the aforementioned three-dimensional spatial coordinates, a first operational entity and a second operational entity are generated by mapping in the virtual scene of a three-dimensional geographic information system (3D GIS). Real-time environmental field data around the work site is obtained, and combined with the work type identifier and initial risk characterization value, a first dynamic risk volume and a second dynamic risk volume that dynamically deform with environmental parameters are constructed using a three-dimensional fluid diffusion algorithm or a geometric transformation algorithm, respectively. A hierarchical collision detection mechanism is adopted to detect in real time whether the first dynamic risk impact body and the second dynamic risk impact body have a three-dimensional spatial intersection; If there is a three-dimensional spatial intersection, the job type identifier of the first and second job permit data is extracted, and the interaction impact parameters between the two are obtained based on the preset risk interaction association model; The interaction impact parameters, the initial risk characterization value of the first work permit data, and the initial risk characterization value of the second work permit data are input into a preset composite risk assessment logic for processing, and the superimposed risk quantification value is output. The superimposed risk quantification value is compared with a preset dynamic grading threshold. When the value triggers the threshold condition, an early warning signal containing three-dimensional spatial positioning information is generated.

2. The method according to claim 1, characterized in that, The steps for constructing a dynamic risk impact body that deforms dynamically with environmental parameters include: The acquired real-time environmental field data includes at least wind speed vector, wind direction vector, gravity direction, and gas density parameters; Determine the basic risk radius or basic volume based on the initial risk characterization value; When the type of operation involves the risk of gas diffusion, the basic volume is stretched into a plume model, a fan-shaped cylinder, or a deflecting cone that extends downwind, based on the wind direction vector and gas density. When the type of work involves the risk of falling from height or sparks flying, the foundation volume is constructed as an inverted cone or an irregularly projected cylinder that extends downward and is deflected by the wind speed, based on the direction of gravity and the wind speed vector.

3. The method according to claim 1, characterized in that, The hierarchical collision detection mechanism specifically includes: Broad Phase: Construct an axis-aligned bounding box (AABB) or a directed bounding box (OBB) for each dynamic risk impact body and insert it into a spatial index tree structure, which is an octree or a hierarchical bounding body (BVH) structure. Use the spatial index tree structure to quickly eliminate combinations of operational entities that do not have bounding box overlap. The Narrow Phase: This phase focuses on combinations of work entities identified as potentially conflicting in the coarse screening phase. It retrieves their high-precision geometric mesh and performs vertex-based precise 3D Boolean intersection operations using the Separating Axis Theorem (SAT) or the GJK algorithm to determine whether there is substantial spatial overlap and to calculate the overlap volume.

4. The method according to claim 1, characterized in that, The risk interaction correlation model determines the interaction impact parameters using at least one of the following methods: A) Discrete coefficient matrix: Calls a pre-stored mapping table of job type pairs and fixed weight coefficients; B) Knowledge graph reasoning: Based on the job node relationships and edge weights defined in ontology, derive dynamic coefficients; C) Machine learning prediction: Input the operation feature vector into a neural network model pre-trained based on historical accident data, and output the predicted interaction weights.

5. The method according to claim 1, characterized in that, The composite risk assessment logic uses the following nonlinear gain algorithm to calculate the superimposed risk quantification value: R_total = alpha * sqrt(Sum(Ri^2)) + beta * (R1 * R2 * ... * Rn * K_interaction) Where R_total is the superimposed risk quantification value, alpha and beta are preset adjustment factors, Ri is the initial risk characterization value of the i-th operation, Sum represents the summation operation, sqrt represents the square root operation, and K_interaction is the interaction effect parameter.

6. The method according to claim 1, characterized in that, The method also includes a visualization rendering step: In response to the early warning signal, the dynamic risk impact body is rendered in the 3D GIS graphical user interface; For the three-dimensional spatial intersection region, volume rendering technology is used to dynamically map different color gradients or transparency based on the magnitude of the superimposed risk quantization value. A floating label is generated at the geometric center of the intersection area, displaying the specific job type that caused the conflict and a risk description.

7. A risk quantification and early warning system for operations that integrates 3D GIS and multidimensional dynamic data, characterized in that, include: The data sensing module is configured to collect work permit data and environmental sensor data that include altitude information; The 3D computing engine is configured to perform the following operations: construct a dynamic risk impact body that deforms with wind speed and gravity parameters; Maintain a spatial index tree and perform bounding box-based coarse screening and grid-based fine 3D Boolean operations to identify spatial conflicts; The risk quantification engine is configured to infer the interaction impact parameters between operations based on semantic interaction analysis or AI models, and calculate the composite risk value according to preset logic. The early warning interactive terminal is configured to highlight three-dimensional risk and conflict areas in the digital twin interface using volumetric rendering technology and push control strategies.

8. A computer-readable storage medium having a computer program stored thereon, characterized in that, When executed by a processor, the program implements the steps of the method according to any one of claims 1 to 6.

Images