Metallurgy pipeline design method and device and program product

Abstract

The invention provides a design method and device of a metallurgical pipeline and a program product, path planning design of the metallurgical pipeline is carried out in a pre-constructed three-dimensional metallurgical factory scene, the three-dimensional metallurgical factory scene is constructed through an imported building structure model, and the three-dimensional metallurgical factory scene comprises arrangement information of equipment and structural members of a building. The design method comprises the following steps: S1, acquiring a starting point position, an end point position, a prohibited area, process parameters and constraint conditions of a metallurgical pipeline input by a user; and S2, according to the starting point position, the terminal point position, the forbidden area, the process parameters and the constraint conditions, performing path search in the three-dimensional metallurgical factory scene through a predetermined path planning algorithm and a pre-trained pipeline layout prediction model to obtain a metallurgical pipeline planning path. By means of the technical scheme, the design efficiency can be improved, the construction cost is reduced, and the building safety is ensured.

Description

Technical Field

[0001] The present invention relates to the technical field of metallurgical pipeline design, and particularly to a design method, device and program product for metallurgical pipelines. Background Art

[0002] In the field of modern engineering design, traditional pipeline design methods rely on manual path planning, resulting in low efficiency and frequent errors. At the same time, it is difficult to comprehensively consider complex process parameter constraints. Collision detection lags often lead to high rework costs, and the optimization process lacks systematicness and repeatability. Although the introduction of existing building information model technology provides functions such as collision detection, it can only passively identify problems and lacks the ability to actively avoid and automatically optimize pipeline layouts. It is difficult to fully reflect key process parameter constraints in the model, and excellent design experience is difficult to be digitalized and reused. In addition, although the application of existing intelligent algorithms in pipeline design shows potential for solving the above problems, there are still challenges in dealing with multi-objective optimization problems and intelligent avoidance of pipeline design. Summary of the Invention

[0003] Embodiments of the present invention provide a design method, device and program product for metallurgical pipelines to improve design efficiency, reduce construction costs, and ensure building safety.

[0004] To achieve the above object, on the one hand, a design method for metallurgical pipelines is provided, which performs path planning design of metallurgical pipelines in a pre-constructed three-dimensional metallurgical plant scene. The three-dimensional metallurgical plant scene is constructed by importing a building structure model, and the three-dimensional metallurgical plant scene includes: layout information of equipment and structural members of buildings. The design method includes:

[0005] S1, obtaining the starting position, ending position, prohibited area, process parameters and constraint conditions of the metallurgical pipeline input by the user; wherein, the constraint conditions include: spatial constraints, process constraints and / or economic constraints;

[0006] S2, according to the starting position, the ending position, the prohibited area, the process parameters and the constraint conditions, performing path search in the three-dimensional metallurgical plant scene through a predetermined path planning algorithm and a pre-trained pipeline layout prediction model to obtain a metallurgical pipeline planned path, including:

[0007] An initial path of the metallurgical pipeline planned by the pipeline layout prediction model, and an initial geometric model of the metallurgical pipeline is constructed by the initial path of the metallurgical pipeline; a predetermined bounding box collision detection algorithm is used to check whether the initial geometric model of the metallurgical pipeline collides with the equipment or the structural member. If a collision occurs, an avoidance strategy is determined according to the process parameters and the relative position to obtain the metallurgical pipeline planned path;

[0008] In addition, if there are multiple constraint conditions and there are conflicting constraint conditions, a predetermined multi-objective optimization algorithm is used to optimize the planned path of the metallurgical pipeline.

[0009] Preferably, in the design method of the metallurgical pipeline, in step S1, the spatial constraints include: the minimum net distance between the metallurgical pipeline and other pipelines except the metallurgical pipeline in the three-dimensional metallurgical plant scene, the turning radius limit of the metallurgical pipeline, and the installation space between the metallurgical pipeline and the structural members;

[0010] The process constraints include: the flow rate of the metallurgical pipeline, the pressure of the metallurgical pipeline, the temperature compensation requirements, the exhaust and drainage requirements, and the vibration control requirements;

[0011] The economic constraints include: material cost, construction difficulty, and maintenance cost.

[0012] Preferably, in the design method of the metallurgical pipeline, in step S1, the process parameters include: pipeline specification parameters, fluid process parameters, and installation technical requirements related to the process parameters; among them, the pipeline specification parameters include: pipe diameter, wall thickness, material type, and insulation layer thickness;

[0013] The fluid process parameters include: medium type, working temperature, working pressure, and design flow rate;

[0014] The installation technical requirements include: minimum slope, support and hanger spacing, and maintenance space size.

[0015] Preferably, in the design method of the metallurgical pipeline, in step S2, the path search is performed in the three-dimensional metallurgical plant scene through a predetermined path planning algorithm and a pre-trained pipeline layout prediction model, and the obtained planned path of the metallurgical pipeline includes:

[0016] S21, performing grid processing on the three-dimensional metallurgical plant scene to obtain a plurality of three-dimensional grids, using the three-dimensional grids as nodes, and using the node where the starting point is located as the current parent node;

[0017] S22, among the multiple nodes adjacent to the current parent node, obtaining adjacent nodes that are not within the prohibited area and meet the constraint conditions;

[0018] In step S2, the initial path of the metallurgical pipeline planned by the pipeline layout prediction model, and an initial geometric model of the metallurgical pipeline is constructed through the initial path of the metallurgical pipeline, including:

[0019] S23. Plan the initial path of the metallurgical pipeline between the current parent node and the adjacent node through the pipeline layout prediction model, and construct an initial geometric model of the metallurgical pipeline through the initial path of the metallurgical pipeline;

[0020] In step S2, use a predetermined bounding box collision detection algorithm to check whether the initial geometric model of the metallurgical pipeline collides with the equipment or the structural member. If a collision occurs, determine an avoidance strategy according to the process parameters and the relative position. The obtained planned path of the metallurgical pipeline includes:

[0021] S24. Calculate the first volume of the initial geometric model of the metallurgical pipeline, and calculate the second volume of the equipment and / or the structural member between the current parent node and the adjacent node; use the bounding box collision detection algorithm to determine whether there is a spatial collision between the first volume and the second volume. If so, go to step S25; if not, go to step S26;

[0022] S25. Determine an avoidance distance according to the process parameters in combination with the distance between the initial geometric model of the metallurgical pipeline and the equipment or the structural member, and determine an avoidance direction according to the position of the adjacent node relative to the prohibited area; the initial path of the metallurgical pipeline obtains the planned path of the metallurgical pipeline between the current parent node and the adjacent node by increasing the value of the avoidance distance in the avoidance direction;

[0023] In step S2, if there are multiple constraint conditions and there are conflicting constraint conditions, use a predetermined multi-objective optimization algorithm to optimize the planned path of the metallurgical pipeline, including:

[0024] S26. If there are multiple constraint conditions and there are conflicting constraint conditions, then design and optimize the planned path of the metallurgical pipeline according to the constraint conditions through the multi-objective optimization algorithm;

[0025] S27. Calculate the heuristic value of the adjacent node through a predetermined heuristic function according to the straight-line distance from the adjacent node to the end position and the collision risk, and calculate the total cost of the adjacent node according to the heuristic value. Select the adjacent node with the smallest total cost value as the current parent node;

[0026] The collision risk is the risk that the geometric model of the metallurgical pipeline collides with all structural members and / or equipment on the straight-line distance, and the collision risk is pre-calculated according to the historical metallurgical pipeline path design scheme and the construction difficulty of the metallurgical pipeline path;

[0027] S28. Determine whether the current parent node is the end position. If so, output the planned path of the metallurgical pipeline; if not, return to step S22.

[0028] Preferably, for the design method of the metallurgical pipeline, in step S23, planning the initial path of the metallurgical pipeline from the current parent node to the adjacent node through the pipeline layout prediction model includes:

[0029] Generating the initial path of the metallurgical pipeline between the current parent node and the adjacent node according to the pipeline layout prediction model in combination with a predetermined genetic algorithm and B-spline curve, and confirming that the generated metallurgical pipeline path meets the process parameters; wherein, the pipeline layout prediction model adopts an improved graph neural network structure and is trained through a predetermined historical metallurgical pipeline path design scheme. The pipeline layout prediction model includes: a node feature extraction module, an edge feature extraction module, a global feature aggregation module, and a multi-head attention module. Among them, through the node feature extraction module, the edge feature extraction module, and the global feature aggregation module, feature extraction and iterative update are performed on the input node information, and then the multi-head attention module is used to capture the patterns and dependencies in the node information.

[0030] Preferably, for the design method of the metallurgical pipeline, the types of constraint conditions include: hard constraints or soft constraints, where the hard constraints are constraint conditions that must be satisfied, and the soft constraints represent constraint conditions that can be optimized and adjusted;

[0031] The processing of the soft constraints is as follows: converting the soft constraints into mathematical expressions and using them as components of the objective function of the multi-objective optimization algorithm, including:

[0032] Converting the constraint conditions corresponding to the soft constraints into a part of the objective function through a predetermined penalty function, where:

[0033] F(x) = f(x) + Σwi * pi(x)

[0034] F(x) represents the objective function, f(x) represents the original objective function, pi(x) represents the penalty term for violating the constraints, and wi represents a predetermined penalty weight.

[0035] Preferably, for the design method of the metallurgical pipeline, the penalty weight is adjusted according to the following formula:

[0036] w(t + 1) = w(t) * (1 + α * v(t))

[0037] w(t + 1) represents the penalty weight updated at time t + 1, w(t) represents the penalty weight at time t, α represents a predetermined learning rate, and v(t) represents the degree of constraint violation at time t.

[0038] Preferably, for the design method of the metallurgical pipeline, after generating the planned path of the metallurgical pipeline, it further includes:

[0039] A geometric model of the metallurgical pipeline constructed according to the planned path of the metallurgical pipeline, and determining the pipeline specifications and pipeline materials of the metallurgical pipeline model through Bayesian optimization and a predetermined reinforcement learning method.

[0040] On the other hand, an embodiment of the present invention provides a design device for a metallurgical pipeline, which includes a memory and a processor. The memory stores at least one program, and the at least one program is executed by the processor to implement the design method of the metallurgical pipeline described in any one of the above.

[0041] In yet another aspect, an embodiment of the present invention provides a computer program product, including a computer program, wherein when the computer program is executed by a processor, it implements the design method of the metallurgical pipeline described in any one of the above.

[0042] The above technical solutions have the following technical effects:

[0043] In the embodiment of the present invention, by combining the starting position, ending position, prohibited area, process parameters and various constraint conditions input by the user, intelligent path search is carried out in the pre-constructed three-dimensional metallurgical plant scene using a path planning algorithm and a pipeline layout prediction model. At the same time, a bounding box collision detection algorithm is used to ensure that the newly designed metallurgical pipeline does not collide with existing equipment or structural members. When encountering conflicting constraint conditions, a multi-objective optimization algorithm is used to optimize and adjust the path, which not only improves the design efficiency and accuracy, but also effectively avoids potential spatial conflicts, reduces construction costs, and ensures the safety and economy of the design scheme, thus realizing an efficient, reliable and metallurgical pipeline path planning design that meets complex process requirements. BRIEF DESCRIPTION OF THE DRAWINGS

[0044] Figure 1 It is a schematic flowchart of the design method of the metallurgical pipeline according to an embodiment of the present invention;

[0045] Figure 2 It is a schematic structural diagram of the design device of the metallurgical pipeline according to another embodiment of the present invention. DETAILED DESCRIPTION OF THE EMBODIMENTS

[0046] To further illustrate the embodiments, the present invention provides accompanying drawings. These drawings are a part of the disclosure of the present invention, which are mainly used to illustrate the embodiments and can be used to explain the operating principle of the embodiments in combination with the relevant descriptions in the specification. With reference to these contents, those of ordinary skill in the art should be able to understand other possible implementation manners and the advantages of the present invention. The components in the figures are not drawn to scale, and similar component symbols are usually used to represent similar components.

[0047] The present invention will be further described in conjunction with the accompanying drawings and specific embodiments.

[0048] Embodiment 1:

[0049] In order to improve the design efficiency, reduce the construction cost, and ensure the building safety, an embodiment of the present invention provides a design method for metallurgical pipelines. Figure 1 It is a schematic flowchart of the design method for metallurgical pipelines according to an embodiment of the present invention. As Figure 1 shown, in a pre-constructed three-dimensional metallurgical plant scene, path planning design of metallurgical pipelines is carried out. The three-dimensional metallurgical plant scene is constructed by importing a building structure model. The three-dimensional metallurgical plant scene includes: layout information of equipment and structural members of buildings. The design method includes:

[0050] S1. Obtain the starting position, ending position, prohibited area, process parameters, and constraint conditions of the metallurgical pipeline input by the user. Among them, the constraint conditions include: spatial constraints, process constraints, and / or economic constraints;

[0051] S2. According to the starting position, ending position, prohibited area, process parameters, and constraint conditions, perform path search in the three-dimensional metallurgical plant scene through a predetermined path planning algorithm and a pre-trained pipeline layout prediction model to obtain the planned path of the metallurgical pipeline, including:

[0052] The initial path of the metallurgical pipeline planned by the pipeline layout prediction model, and construct the initial geometric model of the metallurgical pipeline through the initial path of the metallurgical pipeline. Use a predetermined bounding box collision detection algorithm to check whether the initial geometric model of the metallurgical pipeline collides with equipment or structural members. If a collision occurs, determine an avoidance strategy according to the process parameters and relative positions to obtain the planned path of the metallurgical pipeline;

[0053] And, if there are multiple constraint conditions and there are conflicting constraint conditions, use a predetermined multi-objective optimization algorithm to optimize the planned path of the metallurgical pipeline.

[0054] Embodiment 2:

[0055] Another embodiment of the present invention provides a design method for metallurgical pipelines. The design method performs path planning design of metallurgical pipelines in a pre-constructed three-dimensional metallurgical plant scene, including:

[0056] 1. Obtain the starting position, ending position, prohibited area, process parameters, and constraint conditions of the metallurgical pipeline input by the user.

[0057] In a specific embodiment, the user inputs the spatial layout information and basic process parameters of the metallurgical pipeline through a graphical interface, wherein the pipe fittings and accessories of the metallurgical pipeline are selected according to the basic process parameters, and the corresponding technical requirements are considered in path planning; specifically, the user directly clicks or inputs three-dimensional coordinates in a pre-constructed three-dimensional metallurgical plant scene to define the starting position, end position, process parameters and constraints of the metallurgical pipeline; for metallurgical pipelines with complex process requirements, must-pass positions can also be set to ensure that the metallurgical pipeline passes through the specified position. At the same time, the user is allowed to define prohibited areas to prevent the metallurgical pipeline from crossing specific areas.

[0058] Preferably, the three-dimensional metallurgical plant scene is constructed by importing the building structure model, and the spatial index is established by parsing the imported building structure model data to provide basic data support for subsequent path planning and collision detection; wherein, the three-dimensional metallurgical plant scene includes: structural components such as walls, columns, beams, floor slabs of the building, and layout information of various equipment; the imported building structure model includes: IFC, RVT, DWG and other mainstream formats.

[0059] Preferably, the constraints include: space constraints, process constraints and / or economic constraints; specifically, the space constraints include: the minimum clearance between the metallurgical pipeline and other pipelines except the metallurgical pipeline in the three-dimensional metallurgical plant scene, the turning radius limit of the metallurgical pipeline and the installation space between the metallurgical pipeline and the structural components; the process constraints include: the flow rate of the metallurgical pipeline, the pressure of the metallurgical pipeline, temperature compensation requirements, exhaust and drainage requirements and vibration control requirements; the economic constraints include: material cost, construction difficulty and maintenance cost.

[0060] In a specific embodiment, in the metallurgical process, the pipeline needs to withstand the scouring and erosion of high-temperature molten steel or slag, so the temperature resistance of the pipeline needs to be constrained. For example, for the hot blast furnace pipeline in blast furnace ironmaking, it is necessary to ensure that it can withstand a high temperature of at least 1200°C without deformation or leakage.

[0061] In a specific embodiment, during the metallurgical process, the pipeline may be exposed to corrosive media, such as slag, acidic gas, etc., and the corrosion resistance of the pipeline needs to be constrained. For example, for the exhaust gas emission pipeline in the steelmaking process, it is necessary to ensure that it can resist the corrosion of acidic gas and ensure long-term stable operation;

[0062] The pressure and flow rate of the pipeline have an important impact on production efficiency and product quality. The pressure and flow rate of the pipeline need to be constrained. For example, for the cooling water pipeline of the blast furnace, its flow rate and pressure must be within a certain range to ensure the cooling effect and production safety.

[0063] During the operation of metallurgical equipment, vibration and noise may occur, which can damage pipelines or affect the working environment. Therefore, it is necessary to restrict the vibration and noise of pipelines. For example, for the gas dedusting pipelines of blast furnaces, it is necessary to ensure that they can withstand the equipment vibration without leakage or damage.

[0064] In a specific embodiment, the types of constraint conditions include: hard constraints or soft constraints. Among them, hard constraints are the constraint conditions that must be satisfied, and soft constraints represent the constraint conditions that can be optimized and adjusted.

[0065] The processing of hard constraints includes: eliminating infeasible solutions; for solutions that slightly violate certain constraint conditions, they are repaired through predefined heuristic rules or local search strategies to make them feasible; in evolutionary algorithms such as genetic algorithms, specially designed operators are introduced to ensure that the design solutions still satisfy all constraint conditions after crossover and mutation operations; the constraint conditions are sorted by priority.

[0066] The processing of soft constraints is: converting soft constraints into mathematical expressions and making them part of the objective function of the multi-objective optimization algorithm, including:

[0067] Converting the constraint conditions corresponding to soft constraints into a part of the objective function through a predefined penalty function, where:

[0068] F(x) = f(x) + Σwi * pi(x)

[0069] F(x) represents the objective function, f(x) represents the original objective function, pi(x) represents the penalty term for violating the constraint, and wi represents the predefined penalty weight.

[0070] Preferably, the penalty weight is adjusted according to the following formula:

[0071] w(t + 1) = w(t) * (1 + α * v(t))

[0072] w(t + 1) represents the updated penalty weight at time t + 1, w(t) represents the penalty weight at time t, α represents the predefined learning rate, and v(t) represents the degree of constraint violation at time t.

[0073] When no feasible solutions that satisfy all constraint conditions are found during consecutive predefined iterations of the optimization process, the constraint limits are temporarily relaxed to expand the search space and avoid getting stuck in local optima or being completely unable to find feasible solutions. After one or more feasible solutions are found, the previously relaxed constraint conditions are gradually restored to their original strictness to ensure that the final obtained design solution is as close as possible to the ideal state and meets all the original design requirements and standards.

[0074] Preferably, the process parameters include: pipeline specification parameters, fluid process parameters, and installation technical requirements related to the process parameters; among them, the pipeline specification parameters include: pipe diameter, wall thickness, material type, and insulation layer thickness; the fluid process parameters include: medium type, working temperature, working pressure, and design flow; the installation technical requirements include: minimum slope, support and hanger spacing, and maintenance space size.

[0075] 2. According to the starting position, ending position, prohibited area, process parameters, and constraint conditions, perform path search in the three-dimensional metallurgical plant scenario through a predetermined path planning algorithm and a pre-trained pipeline layout prediction model to obtain the metallurgical pipeline planning path, including:

[0076] The initial path of the metallurgical pipeline planned by the pipeline layout prediction model, and construct the initial geometric model of the metallurgical pipeline through the initial path of the metallurgical pipeline; use a predetermined bounding box collision detection algorithm to check whether the initial geometric model of the metallurgical pipeline collides with equipment or structural components. If a collision occurs, determine the avoidance strategy according to the process parameters and relative positions to obtain the metallurgical pipeline planning path;

[0077] And, if there are multiple constraint conditions and there are conflicting constraint conditions, use a predetermined multi-objective optimization algorithm to optimize the metallurgical pipeline planning path.

[0078] Preferably, the path planning algorithm is the A* algorithm; the multi-objective optimization algorithm is a multi-objective optimization algorithm based on Pareto optimality.

[0079] In a specific embodiment, the multi-objective optimization algorithm incorporates expert experience and optimizes the metallurgical pipeline planning path through a dynamic weight mechanism. Among them, for various different process conditions and environmental requirements in the metallurgical plant, dynamically adjusting the weights can make the optimization results better meet the needs of specific working conditions, dynamically adjust the weights of each objective function according to the design focus, seek the best balance between these objectives, and flexibly respond to these changing needs to ensure that the final planned path obtained is not only theoretically optimal but also feasible in practical applications; in addition, combining historical design experience as a reference, extracting the key parameters and successful patterns from it, and integrating them into the current design process can help new designs avoid common mistakes and improve the success rate. At the same time, taking into account the specific engineering background, including but not limited to site conditions, available resources, budget constraints, etc., makes the design scheme more targeted and feasible; through continuous adjustment and improvement, ensure that the final output planned path is not only technically advanced but also has high value in terms of economy and implementation possibility.

[0080] In a specific embodiment, in step 2, performing path search in the three-dimensional metallurgical plant scenario through a predetermined path planning algorithm and a pre-trained pipeline layout prediction model to obtain the metallurgical pipeline planning path includes:

[0081] 2.1. Mesh the three-dimensional metallurgical plant scene to obtain multiple three-dimensional meshes. Use the three-dimensional meshes as nodes, and use the node where the starting position is located as the current parent node.

[0082] 2.2. Among the multiple nodes adjacent to the current parent node, obtain the adjacent nodes that are not within the prohibited area and meet the constraint conditions.

[0083] 2.3. Plan the initial path of the metallurgical pipeline between the current parent node and the adjacent nodes through the pipeline layout prediction model, and construct the initial geometric model of the metallurgical pipeline through the initial path of the metallurgical pipeline;

[0084] Preferably, between the current parent node and the adjacent nodes, generate the initial path of the metallurgical pipeline according to the pipeline layout prediction model in combination with the predetermined genetic algorithm and B-spline curve, and confirm that the generated metallurgical pipeline path meets the process parameters; among them, the pipeline layout prediction model adopts an improved graph neural network structure and is trained through the predetermined historical metallurgical pipeline path design scheme. The pipeline layout prediction model includes: a node feature extraction module, an edge feature extraction module, a global feature aggregation module, and a multi-head attention module. Among them, through the node feature extraction module, the edge feature extraction module, and the global feature aggregation module, feature extraction and iterative update are performed on the input node information, and then the multi-head attention module is used to capture the complex patterns and dependencies in the node information.

[0085] 2.4. Calculate the first volume of the initial geometric model of the metallurgical pipeline, and calculate the second volume of the equipment and / or structural members between the current parent node and the adjacent nodes; use the bounding box collision detection algorithm to determine whether there is a spatial collision between the first volume and the second volume. If so, go to step 2.5; if not, go to step 2.6;

[0086] Preferably, the bounding box collision detection algorithm includes: AABB bounding box, OBB bounding box, and / or spherical bounding box; the AABB bounding box is used for fast detection of rough screening, the OBB bounding box is used for precise detection, and the spherical bounding box is used for detection of special components.

[0087] 2.5. Determine the avoidance distance according to the process parameters in combination with the distance between the initial geometric model of the metallurgical pipeline and the equipment or structural members, and determine the avoidance direction according to the position of the adjacent node relative to the prohibited area; the initial path of the metallurgical pipeline obtains the planned path of the metallurgical pipeline between the current parent node and the adjacent nodes by adding the value of the avoidance distance in the avoidance direction.

[0088] 2.6. If there are multiple constraint conditions and there are conflicting constraint conditions, design and optimize the planned path of the metallurgical pipeline according to the constraint conditions through the multi-objective optimization algorithm.

[0089] 2.7. Calculate the heuristic value of the adjacent node based on the straight-line distance from the adjacent node to the end point and the collision risk through a predetermined heuristic function, and calculate the total cost of the adjacent node based on the heuristic value, and select the adjacent node with the smallest total cost as the current parent node;

[0090] The collision risk is the risk of collision between the metallurgical pipeline geometric model and all structural components and / or equipment within a straight-line distance. The collision risk is pre-calculated based on the historical metallurgical pipeline route design plan and the construction difficulty of the metallurgical pipeline route.

[0091] Preferably, the predetermined heuristic function is a hybrid heuristic function based on distance and collision risk.

[0092] 2.8. Determine whether the current parent node is the end point. If so, output the metallurgical pipeline planning path; if not, return to step 2.2.

[0093] Preferably, after the metallurgical pipeline planning path is generated, it also includes: a metallurgical pipeline geometric model constructed according to the metallurgical pipeline planning path, and determining the pipeline specifications and pipeline materials of the metallurgical pipeline model through Bayesian optimization and a predetermined reinforcement learning method; wherein the metallurgical pipeline geometric model includes: accessory models such as pipe fittings, flanges, supports and hangers, and auxiliary structure models such as insulation layers and anti-corrosion layers; the metallurgical pipeline geometric model is obtained by rendering with real materials, supports dynamic roaming browsing, and provides interactive functions such as sectioning and measurement.

[0094] Preferably, after the metallurgical pipeline planning path is generated, the metallurgical pipeline planning path is simulated based on fluid dynamics and structural mechanics, and the simulation results are fed back to the multi-objective optimization algorithm to form a closed-loop optimization mechanism and continuously improve the design scheme; among them, the fluid dynamics simulation uses CFD technology to analyze the flow field distribution in the pipeline and calculate the pressure loss, so as to evaluate the local resistance system number and verify the flow stability; the structural mechanics simulation includes: static analysis, modal analysis, thermal stress analysis and fatigue analysis to ensure the safety of the pipeline structure.

[0095] Preferably, when outputting the planned path of the metallurgical pipeline, design drawings, technical documents, and analysis and evaluation reports that meet industry standards are generated simultaneously; among them, the design drawings include: pipeline plan layout drawings, pipeline system drawings, pipeline node details drawings, and material detail table drawings. The pipeline plan layout drawings mark the flow direction, elevation, slope and other information of the pipeline. The pipeline system diagram shows the system composition and control relationship of the pipeline. The pipeline node details include special connections of the pipeline, supports and hangers and other structural details. The material detail table drawings list the specifications and quantities of all pipes and fittings.

[0096] The technical documents include: design specifications, construction guides, material procurement lists, and inspection and acceptance criteria. Among them, the design specifications include: detailed design basis, technical parameters, and special requirements; the construction guides include: construction procedures, quality requirements, and precautions; the material procurement lists include: detailed material specifications and quantity information; the inspection and acceptance criteria stipulate the specific requirements for construction quality inspection.

[0097] The analysis and evaluation reports include: collision detection reports, hydraulic calculation reports, stress analysis reports, and construction feasibility reports. Among them, the collision detection reports include: the positions, types, and handling suggestions of all collision points; the hydraulic calculation reports include: calculation results such as pressure loss and flow velocity distribution; the stress analysis reports are used to evaluate the stress state of the pipeline under various working conditions; the construction feasibility reports are used to analyze the construction difficulty and potential risks.

[0098] In summary, the method of the embodiment of the present invention adopts a multi-level path planning strategy, decomposing the complex pipeline layout problem into three levels: macro planning, local optimization, and fine adjustment; at the macro level, an improved A* algorithm is used for preliminary path search, simplifying the calculation complexity through spatial rasterization; at the local optimization level, a genetic algorithm is used to optimize the path, and a B-spline curve is used to achieve smooth transition; at the fine adjustment level, a local search algorithm is used to handle pipe fitting layout and collision avoidance.

[0099] Embodiment 3:

[0100] The present invention also provides a design device for metallurgical pipelines, as Figure 2 shown. The device includes a processor 201, a memory 202, a bus 203, and a computer program stored in the memory 202 and operable on the processor 201. The processor 201 includes one or more processing cores. The memory 202 is connected to the processor 201 through the bus 203. The memory 202 is used to store program instructions. When the processor 201 executes the computer program, it implements the steps in the above method embodiment of Embodiment 1 of the present invention.

[0101] Furthermore, as an executable solution, the design device for metallurgical pipelines can be a computer unit, and this computer unit can be computing devices such as a desktop computer, a notebook, a palm computer, and a cloud server. The computer unit may include, but is not limited to, a processor and a memory. Those skilled in the art can understand that the above composition structure of the computer unit is only an example of the computer unit, and does not constitute a limitation on the computer unit. It may include more or fewer components than the above, or combine some components, or different components. For example, the computer unit may further include input and output devices, network access devices, a bus, etc., and the embodiments of the present invention do not make limitations in this regard.

[0102] Further, as an executable solution, the so-called processor may be a Central Processing Unit (CPU), or it may also be other general-purpose processors, Digital Signal Processors (DSPs), Application Specific Integrated Circuits (ASICs), Field-Programmable Gate Arrays (FPGAs), or other programmable logic devices, discrete gate or transistor logic devices, discrete hardware components, etc. The general-purpose processor may be a microprocessor or the processor may also be any conventional processor, etc. The processor is the control center of the computer unit, and connects various parts of the entire computer unit using various interfaces and lines.

[0103] The memory can be used to store the computer program and / or modules. The processor realizes various functions of the computer unit by running or executing the computer program and / or modules stored in the memory, and by calling the data stored in the memory. The memory may mainly include a program storage area and a data storage area. Among them, the program storage area can store an operating system and application programs required for at least one function; the data storage area can store data created according to the use of the mobile phone, etc. In addition, the memory can include high-speed random access memory, and can also include non-volatile memory, such as hard disks, memory, plug-in hard disks, Smart Media Cards (SMCs), Secure Digital (SD) cards, Flash Cards, at least one magnetic disk storage device, flash memory device, or other volatile solid-state storage devices.

[0104] Embodiment 4:

[0105] The present invention also provides a computer program product, including a computer program, and when the computer program is executed by a processor, the steps of the method described above are realized.

[0106] Although the present invention is specifically shown and introduced in combination with preferred implementation modes, those skilled in the art should understand that various changes can be made to the present invention in terms of form and details without departing from the spirit and scope of the present invention defined by the appended claims, and all of them fall within the protection scope of the present invention.

Claims

1. A design method for metallurgical pipelines, which conducts path planning and design of metallurgical pipelines in a pre-constructed three-dimensional metallurgical plant scenario. The three-dimensional metallurgical plant scenario is constructed by importing a building structure model, and the three-dimensional metallurgical plant scenario includes: The layout information of the equipment and the structural members of the building, characterized by including: S1. Obtain the starting position, ending position, prohibited area, process parameters, and constraint conditions of the metallurgical pipeline input by the user; wherein, the constraint conditions include: spatial constraints, process constraints, and / or economic constraints; S2. According to the starting position, the ending position, the prohibited area, the process parameters, and the constraint conditions, perform path search in the three-dimensional metallurgical plant scene through a predetermined path planning algorithm and a pre-trained pipeline layout prediction model to obtain the metallurgical pipeline planning path, including: The initial path of the metallurgical pipeline planned by the pipeline layout prediction model, and construct an initial geometric model of the metallurgical pipeline through the initial path of the metallurgical pipeline; use a predetermined bounding box collision detection algorithm to check whether the initial geometric model of the metallurgical pipeline collides with the equipment or the structural member. If a collision occurs, determine an avoidance strategy according to the process parameters and the relative position to obtain the metallurgical pipeline planning path; And, if there are multiple constraint conditions and there are conflicting constraint conditions, use a predetermined multi-objective optimization algorithm to optimize the metallurgical pipeline planning path.

2. The design method of the metallurgical pipeline according to claim 1, characterized in that In step S1, the spatial constraints include: the minimum net distance between the metallurgical pipeline and other pipelines except the metallurgical pipeline in the three-dimensional metallurgical plant scene, the turning radius limit of the metallurgical pipeline, and the installation space between the metallurgical pipeline and the structural member; The process constraints include: the flow rate of the metallurgical pipeline, the pressure of the metallurgical pipeline, the temperature compensation requirements, the exhaust and drainage requirements, and the vibration control requirements; The economic constraints include: material cost, construction difficulty, and maintenance cost.

3. The design method of the metallurgical pipeline according to claim 1, characterized in that In step S1, the process parameters include: pipeline specification parameters, fluid process parameters, and installation technical requirements related to the process parameters; wherein, the pipeline specification parameters include: pipe diameter, wall thickness, material type, and insulation layer thickness; The fluid process parameters include: medium type, working temperature, working pressure, and design flow rate; The installation technical requirements include: minimum slope, support and hanger spacing, and maintenance space size.

4. The design method of the metallurgical pipeline according to claim 1, characterized in that The path search in the three-dimensional metallurgical plant scene through a predetermined path planning algorithm and a pre-trained pipeline layout prediction model in step S2 to obtain the metallurgical pipeline planning path includes: S21. Perform grid processing on the three-dimensional metallurgical plant scene to obtain a plurality of three-dimensional grids, use the three-dimensional grids as nodes, and use the node where the starting position is located as the current parent node; S22. Among the multiple nodes adjacent to the current parent node, obtain adjacent nodes that are not in the prohibited area and meet the constraint conditions; The initial path of the metallurgical pipeline planned by the pipeline layout prediction model in step S2, and construct an initial geometric model of the metallurgical pipeline through the initial path of the metallurgical pipeline, including: S23. Plan the initial path of the metallurgical pipeline between the current parent node and the adjacent node through the pipeline layout prediction model, and construct an initial geometric model of the metallurgical pipeline through the initial path of the metallurgical pipeline; In step S2, a predetermined bounding box collision detection algorithm is used to check whether the initial geometric model of the metallurgical pipeline collides with the equipment or the structural member. If a collision occurs, an avoidance strategy is determined based on the process parameters and the relative position, and obtaining the planned path of the metallurgical pipeline includes: S24, calculating the first volume of the initial geometric model of the metallurgical pipeline, and calculating the second volume of the equipment and / or the structural member between the current parent node and the adjacent node; judging whether there is a spatial collision between the first volume and the second volume through the bounding box collision detection algorithm. If so, go to step S25; if not, go to step S26; S25, determining an avoidance distance according to the process parameters in combination with the distance between the initial geometric model of the metallurgical pipeline and the equipment or the structural member, and determining an avoidance direction according to the position of the adjacent node relative to the prohibited area; the initial path of the metallurgical pipeline is obtained by adding the value of the avoidance distance in the avoidance direction to obtain the planned path of the metallurgical pipeline between the current parent node and the adjacent node; In step S2, if there are multiple constraint conditions and there are conflicting constraint conditions, using a predetermined multi-objective optimization algorithm to optimize the planned path of the metallurgical pipeline includes: S26, if there are multiple constraint conditions and there are conflicting constraint conditions, then design and optimize the planned path of the metallurgical pipeline according to the constraint conditions through the multi-objective optimization algorithm; S27, calculating the heuristic value of the adjacent node through a predetermined heuristic function according to the straight-line distance from the adjacent node to the end position and the collision risk, and calculating the total cost of the adjacent node according to the heuristic value, and selecting the adjacent node with the smallest total cost value as the current parent node; The collision risk is the risk that the geometric model of the metallurgical pipeline collides with all structural members and / or equipment on the straight-line distance, and the collision risk is pre-calculated according to the historical metallurgical pipeline path design scheme and the construction difficulty of the metallurgical pipeline path; S28, judging whether the current parent node is the end position. If so, output the planned path of the metallurgical pipeline; if not, return to step S22.

5. The design method of the metallurgical pipeline according to claim 4, characterized in that, In step S23, planning the initial path of the metallurgical pipeline from the current parent node to the adjacent node through the pipeline layout prediction model includes: Generate an initial path of the metallurgical pipeline between the current parent node and the adjacent node by combining a predetermined genetic algorithm and a B-spline curve according to the pipeline layout prediction model, and confirm that the generated metallurgical pipeline path meets the process parameters; wherein, the pipeline layout prediction model adopts an improved graph neural network structure and is trained through a predetermined historical metallurgical pipeline path design scheme. The pipeline layout prediction model includes: a node feature extraction module, an edge feature extraction module, a global feature aggregation module, and a multi-head attention module. Among them, through the node feature extraction module, the edge feature extraction module, and the global feature aggregation module, feature extraction and iterative update are performed on the input node information, and then the multi-head attention module is used to capture the patterns and dependencies in the node information.

6. The design method of the metallurgical pipeline according to claim 1, characterized in that, The types of constraint conditions include: hard constraints or soft constraints, where the hard constraints are the constraint conditions that must be satisfied, and the soft constraints represent the constraint conditions that can be optimized and adjusted; The processing of the soft constraints is: convert the soft constraints into mathematical expressions and use them as part of the objective function of the multi-objective optimization algorithm, including: Convert the constraint conditions corresponding to the soft constraints into a part of the objective function through a predetermined penalty function, where: F(x) = f(x) + Σwi * pi(x) F(x) represents the objective function, f(x) represents the original objective function, pi(x) represents the penalty term for violating the constraints, and wi represents the predetermined penalty weight.

7. The design method of the metallurgical pipeline according to claim 6, characterized in that, The penalty weight is adjusted according to the following formula, w(t + 1) = w(t) * (1 + α * v(t)) w(t + 1) represents the penalty weight updated at time t + 1, w(t) represents the penalty weight at time t, α represents the predetermined learning rate, and v(t) represents the degree of constraint violation at time t.

8. The design method of the metallurgical pipeline according to claim 1, characterized in that, After generating the planned path of the metallurgical pipeline, it further includes: Construct a geometric model of the metallurgical pipeline according to the planned path of the metallurgical pipeline, and determine the pipeline specifications and pipeline materials of the metallurgical pipeline model through Bayesian optimization and a predetermined reinforcement learning method.

9. A design device for a metallurgical pipeline, characterized in that, It includes a memory and a processor. The memory stores at least one segment of program, and the at least one segment of program is executed by the processor to implement the design method of the metallurgical pipeline as described in any one of claims 1 to 8.

10. A computer program product, comprising a computer program, characterized in that, When the computer program is executed by the processor, it implements the design method of the metallurgical pipeline as described in any one of claims 1 to 8.

Images