Pipe support and hanger integration optimization method and system based on process engine

Abstract

The application provides a pipeline support and hanger integrated optimization method and system based on a process engine, and the method comprises the following steps: S1, creating or opening at least one optimization project; S2, constructing an optimization process; S3, setting optimization targets and constraint conditions, selecting target parameters to be optimized, and setting hyperparameters of an optimization algorithm; S4, starting automatic execution of the optimization process; S5, in the optimization process, asynchronous signal slot mechanism is used to realize decoupled communication between the optimization algorithm and the interface, and a multi-target dynamic curve diagram is used to draw the convergence trend of each optimization target in real time; S6, after the optimization is completed, multiple candidate schemes are selected, compared with the initial arrangement, the technical rationality of each scheme is comprehensively evaluated, and the optimal scheme is selected. Through the construction of a visual modeling and automatic execution mechanism based on the process engine, the traditional support and hanger optimization process which depends on manual repeated iteration is converted into automatic execution, and full-process closed-loop optimization of analysis, optimization and feedback is realized.

Description

Technical Field

[0001] This invention relates to the field of pipeline support and hanger arrangement in nuclear power plants, and particularly to a pipeline support and hanger integrated optimization method and system based on a process engine. Background Technology

[0002] In the nuclear power field, the piping system of a nuclear power plant is a critical channel for transporting media, and its safety and reliability directly affect the operational safety of the entire system. To effectively control the displacement and stress levels of pipelines under various operating conditions such as self-weight, thermal expansion, and earthquakes, it is essential to rationally design pipeline supports and hangers. The arrangement of supports and hangers not only affects the stress distribution of the piping system but also significantly impacts the load transfer to critical components such as connected equipment pipes, valves, and penetrations. Therefore, the optimized design of supports and hangers is a core aspect of pipeline mechanical analysis and engineering design.

[0003] In existing technologies, such as Figure 1 As shown, the traditional pipe support optimization process mainly relies on the experience of design engineers for manual iterative adjustments. Specifically, engineers conduct pipe stress analysis based on the initial layout. If the results show that the stress exceeds the limit or the equipment connection load exceeds the standard, they manually adjust the position, type, or quantity of the supports and hangers, and repeat the calculation and evaluation process until all data meets the specifications and limit requirements. In addition, engineers also need to comprehensively compare multiple feasible solutions to determine the final recommended solution.

[0004] However, this manual optimization method has the following significant problems: 1. Low optimization efficiency: Each adjustment of the supports and hangers requires rearranging and running the analysis software. The entire process relies entirely on manual operation, resulting in a long iteration cycle that is time-consuming and labor-intensive.

[0005] Second, it is difficult to balance economy and safety: Engineers often prioritize structural safety and focus on one or a few key indicators, which can easily lead to overly conservative design schemes, such as an excessive number of supports and hangers and a high proportion of non-standard parts. This increases material costs and makes it difficult to generate a comprehensive optimal solution that balances safety and economy.

[0006] Third, existing software lacks closed-loop integration capabilities: mainstream pipe mechanics analysis software (such as Pipestress and CAESAR II) only provides modeling and solving functions, without built-in optimization algorithms, and does not have the ability to automatically complete the optimization of pipe supports and hangers.

[0007] Therefore, there is an urgent need for a pipeline support optimization method and system that is based on process automation, supports multi-objective collaboration, has intelligent decision-making capabilities, and can be deeply integrated with third-party analysis software, in order to solve the problems of low efficiency and difficulty in global optimization of traditional manual optimization, thereby promoting the development of pipeline design towards intelligence and efficiency.

[0008] In view of this, the inventors of this application have designed a pipeline support and hanger integration optimization method and system based on a process engine in order to overcome the above-mentioned technical problems. Summary of the Invention

[0009] The technical problem to be solved by the present invention is to overcome the shortcomings of traditional manual optimization of pipe supports and hangers in the prior art, which is characterized by low efficiency and difficulty in global optimization, and to provide an integrated optimization method and system for pipe supports and hangers based on a process engine.

[0010] The present invention solves the above-mentioned technical problems through the following technical solution: A process engine-based integrated optimization method for pipe supports and hangers, characterized in that the method includes the following steps: S1. Based on the SQLite database, persistently store project information and create or open at least one optimized project; S2. Build and optimize processes using a visual process editor; The optimization process in step S2 includes: S 21 Users drag and drop functional nodes from the node library into the modeling space; S 22 After the nodes are placed, the user can connect the nodes by using the mouse; when the output port Socket is pressed and dragged to the input port Socket, the system automatically generates a permanent connection line. The step S 21 The functional nodes include analysis block nodes, optimization block nodes, and auxiliary function block nodes; S3. Set optimization objectives and constraints, extract key evaluation parameters from the mechanical analysis results file output by the third-party pipeline analysis software, select the target parameters to be optimized, set the hyperparameters of the optimization algorithm, and save the optimization configuration information in JSON format after configuration to support subsequent loading and reuse. S4. Initiate the automated execution of the optimization process; S5. During the optimization process, an asynchronous signal slot mechanism is used to achieve decoupled communication between the optimization algorithm and the interface. The multi-objective dynamic curve graph draws the convergence trend of each optimization objective in real time, and the user judges the convergence of the algorithm by the curve changes. S6. After optimization, select multiple candidate schemes, compare and analyze them with the initial layout, comprehensively evaluate the technical rationality of each scheme, and select the optimal scheme.

[0011] According to an embodiment of the present invention, the construction of the optimization process in step S2 includes: the user selects a typical optimization process template from the preset template library and generates a complete optimization process diagram with one click; after the user selects the template, the system automatically generates a complete process containing all necessary nodes and connections.

[0012] According to an embodiment of the present invention, the analysis block node is configured to: in the process construction stage, load the pipeline system model and mechanical analysis results before optimization in response to user operation, assist the user in identifying and defining the range of pipe segments to be optimized based on actual working conditions, and transmit the pipe segment information and support and hanger layout data within the optimization range to the optimization algorithm module; During the process execution phase, in response to the scheduling instructions of the process engine, the system automatically calls third-party pipeline mechanics analysis software to perform static or dynamic stress analysis on the current support and hanger layout scheme, automatically acquires mechanical evaluation data, and returns the data as optimization indicators to the downstream node for objective function verification and constraint condition evaluation.

[0013] According to an embodiment of the present invention, the optimization block node is configured to receive optimization objectives and constraints from the parameter configuration interface, as well as optimization range definitions and mechanical evaluation data from the analysis block node, as input parameters for a multi-objective optimization problem; A multi-objective optimization algorithm is invoked to perform non-dominated sorting and crowding calculation on the initial population, generating the Pareto front solution set for the current generation; Based on the Pareto front solution set, selection, crossover, and mutation operations are performed to generate offspring populations. The parent and offspring populations are then merged and environmental selection is performed to form a new generation population. Determine whether the preset termination condition is met. If not, continue the iterative optimization process. Otherwise, output the final Pareto optimal solution set to the post-processing module for visualization and optimization analysis of the optimization scheme.

[0014] According to one embodiment of the present invention, the auxiliary function block node is configured to perform additional data calculations, including standard support and hanger ratio calculations and similarity calculations with the pipe support and hanger layout before optimization, so as to support the comprehensive evaluation and optimal decision of the optimization scheme.

[0015] According to an embodiment of the present invention, step S 22 The center line connection uses A Path planning algorithms enable automatic layout, wherein A The overall cost function of the path planning algorithm is defined as F = G + H + E; Where G represents the actual path cost from the starting point to the current search point; H represents the heuristically estimated cost based on the improved Manhattan distance, with offset correction based on the relative orientation of the source Socket and the target Socket; and E represents the turning penalty term, used to suppress high-frequency turning.

[0016] According to one embodiment of the present invention, the A The path planning algorithm process includes the following steps: S 221 Initialize the search data structure; S 222 Add the starting point to the open list; S 223 Calculate the difference between the starting point and the ending point; S 224 1. Determine if the open set is empty; if not, remove the point with the smallest F value as the current point from the open set and add it to the closed set, then proceed to step S. 225 If so, proceed to step S. 228 ; S 225 1. Determine if the current point is the destination; if yes, return the found path; otherwise, obtain the surrounding walkable points and proceed to step S. 228 ; S 226 1. Determine whether all surrounding points have been traversed. If not, return to step S. 224 If so, proceed to step S. 227 ; S 227 1. Determine if the point is in the open set; if not, set the parent node as the current point, calculate the G, H, and F values, add it to the open set, and proceed to step S. 228 If so, check if a shorter path has been found and proceed to step S. 228 ; S 228 1. Determine if the destination is in the open list; if yes, return the found path; otherwise, return the path not found.

[0017] According to one embodiment of the present invention, the mechanical analysis results in step S3 include: pipeline stress, flange weld stress, equipment and valve pipe load, valve mass center of gravity acceleration, and through-body load and displacement; the data structure is constructed in a hierarchical manner according to categories, and qualitative identification information characterizing the safety status of the parameters is generated based on the calculated values ​​and preset threshold results.

[0018] According to an embodiment of the present invention, the target parameters to be optimized in step S3 include: the maximum or minimum value of the target, and the maximum and minimum values ​​of the constraints.

[0019] According to an embodiment of the present invention, the hyperparameters of the optimization algorithm in step S3 include: population size, maximum number of iterations, crossover probability, and mutation probability.

[0020] According to an embodiment of the present invention, step S4 includes the following steps: S 41 Topology parsing is performed using a process engine; S 42 A depth-first search algorithm is used to perform topological sorting to determine the execution order of nodes; S 43 The system verifies the completeness and validity of the parameters of each node and executes each node sequentially. S 44 When the execution reaches the optimization block node, the multi-objective optimization algorithm is called to obtain the computational solution set; S 45 Based on the calculated solution set, selection, crossover, and mutation operations are performed to generate offspring populations, and a new generation population is formed through environmental selection; after iterating until the termination condition is met, the final optimal solution set is output.

[0021] According to an embodiment of the present invention, step S 44 The multi-objective optimization algorithm described herein is the NSGA-II multi-objective optimization algorithm.

[0022] According to an embodiment of the present invention, the invocation process of the NSGA-II multi-objective optimization algorithm includes the following steps: S 441 Initialize logging, read JSON configuration parameters, and obtain the interpreter path; S 442 Obtain the objective function parameters and constraints, and create an optimization configuration object; S 443 1. Determine if the optimization scope file exists; if so, proceed to step S. 444 If not, proceed to step S. 447 ; S 444 Read the optimization scope file, obtain the task type and optimization scope, and create a problem instance; obtain the checkpoint file path and create an evolutionary algorithm instance; S 445 , Call the evaluation method to perform evolutionary calculations; S 446 Perform data processing, output the results to a JSON file, return the results, and end the process; S 447 Error message will be displayed and the process will be terminated.

[0023] According to an embodiment of the present invention, the computation steps of the NSGA-II multi-objective optimization algorithm include: S 451 Load configuration from JSON file; S 452 1. Identify the problem type and analysis type; S 453 Initialize the population and call the pipeline software to calculate the target value; S 454 Based on the target value, divide the Pareto front and perform fast non-dominated sorting; S 455 Based on the Pareto front, the target values ​​of the proposed schemes are sorted and the crowding distance is calculated. S 456 Perform parent selection, crossover and mutation operations to produce offspring populations, calculate offspring target values, merge parent and offspring populations, perform fast non-dominated sorting, and select the best Pareto front. S 457 1. Determine whether the specified number of training generations has been reached, or whether the early stopping condition has been met; if not, return to step S. 456 If so, output the optimal solution.

[0024] This invention also provides a pipeline support and hanger integration optimization system based on a process engine, characterized in that the pipeline support and hanger integration optimization system based on a process engine adopts the pipeline support and hanger integration optimization method described above, and the optimization system includes: The GUI module is used to provide a graphical user interface. The process engine submodule is used to respond to user-built graphical processes and is responsible for process parsing and execution. The optimization algorithm module is used to call multi-objective optimization algorithms to generate the optimal solution set; The interface module provides an interface with pipeline finite element analysis software, configured to enable bidirectional data interaction with third-party pipeline mechanics analysis software via a standardized data interface, as well as Python script integration functionality. The database module is used for persistent storage and management of project information, process definitions, node parameters, and historical optimization data.

[0025] The present invention also provides an electronic device, characterized in that the electronic device includes: a processor and a memory, the memory storing programs or instructions that can run on the processor, the programs or instructions being executed by the processor to implement the process engine-based pipe support and hanger integration optimization method as described above.

[0026] The present invention also provides a readable storage medium, characterized in that a program or instruction is stored on the readable storage medium, and when the program or instruction is executed by a processor, it implements the pipeline support and hanger integration optimization method based on the process engine as described above.

[0027] The positive and progressive effects of this invention are as follows: The pipe support and hanger integration optimization method and system based on a process engine, as described in this invention, has the following advantages: I. Achieving full-process automation and significantly improving optimization efficiency: This invention transforms the traditional support and hanger optimization process, which relies on repeated manual iterations, into automated execution by constructing a process engine-based visual modeling and automatic execution mechanism, realizing a closed-loop optimization process from "analysis → optimization → feedback".

[0028] II. Providing an intelligent human-machine collaborative interface to lower the barrier to entry: Through a tabbed multi-view architecture and templated process configuration, the learning cost for new engineers is significantly reduced, and the consistency of design quality is improved. Users can quickly start optimization tasks through "click-to-use" templates, quickly identify out-of-limit parameters through color indicators, and intuitively compare candidate solutions through Pareto's cutting-edge visualization, thereby standardizing the optimization process.

[0029] Third, it supports multi-objective collaborative optimization and effectively explores the global optimal solution: This invention can integrate multi-objective optimization algorithms such as NSGA-II, and can systematically weigh multiple conflicting objectives such as minimizing the number of supports and hangers, controlling the stress ratio of each working condition, and controlling the load of equipment pipe connection, providing a variety of optimization schemes, which are significantly better than manual adjustment that relies on experience. Attached Figure Description

[0030] The above and other features, properties and advantages of the present invention will become more apparent from the following description taken in conjunction with the accompanying drawings and embodiments, in which the same reference numerals always denote the same features, wherein: Figure 1 Optimize the flow chart for traditional pipelines.

[0031] Figure 2 This invention provides a formalized process for optimizing the layout design of pipe supports and hangers in the pipeline support and hanger integrated optimization method based on a process engine.

[0032] Figure 3 This diagram illustrates the relationship between the optimization process and stress analysis in the pipeline support and hanger integrated optimization method based on the process engine of this invention.

[0033] Figure 4 This is a flowchart of the pipe support and hanger integration optimization method based on a process engine according to the present invention.

[0034] Figure 5In the pipeline support and hanger integration optimization method based on process engine of this invention, A Flowchart of the path planning algorithm.

[0035] Figure 6 This is a flowchart illustrating the execution of the process engine algorithm in the pipeline support and hanger integrated optimization method based on the process engine of this invention.

[0036] Figure 7 This is a flowchart illustrating the overall operation of the NSGA-II multi-objective optimization algorithm in the pipeline support and hanger integrated optimization method based on the process engine of this invention.

[0037] Figure 8 This is a flowchart of the NSGA-II multi-objective optimization algorithm in the pipeline support and hanger integrated optimization method based on the process engine of this invention.

[0038] Figure 9 This is a module architecture diagram of the pipeline support and hanger integrated optimization system based on the process engine of this invention. Detailed Implementation

[0039] To make the above-mentioned objects, features and advantages of the present invention more apparent and understandable, the specific embodiments of the present invention will be described in detail below with reference to the accompanying drawings.

[0040] Embodiments of the invention will now be described in detail with reference to the accompanying drawings. Preferred embodiments of the invention will now be described in detail, examples of which are shown in the drawings. Wherever possible, the same reference numerals will be used in all the drawings to denote the same or similar parts.

[0041] Furthermore, although the terminology used in this invention is selected from commonly known and used terms, some terms mentioned in this specification may have been selected by the applicant in his or her judgment, and their detailed meanings are explained in the relevant sections of the description herein.

[0042] Furthermore, the invention should be understood not only through the actual terminology used, but also through the meaning implied by each term.

[0043] like Figures 1 to 8 As shown, this invention discloses an integrated optimization method for pipe supports and hangers based on a process engine, which includes the following steps: Step S1: Persistently store project information based on the SQLite database, and create or open at least one optimized project.

[0044] The stored project information includes metadata such as project name, creation time, and modification time. Users can launch the software, create or open at least one optimization project, and manage multiple optimization processes within the project directory.

[0045] Step S2: Build and optimize the process using a visual process editor.

[0046] The optimized process is constructed by the client using a visual process editor in the "Process" tab. The optimized process consists of multiple functional nodes and connecting lines.

[0047] For example, the following two methods can be used to quickly build an optimized process: Method 1: The construction of the optimization process in step S2 includes: the user selects a typical optimization process template from the preset template library and generates a complete optimization process diagram with one click; after the user selects a template, the system automatically generates a complete process containing all necessary nodes and connections.

[0048] Method 2: The optimization process in step S2 can preferably include: Step S 21 Users can drag and drop functional nodes from the node library into the modeling space.

[0049] The functional nodes may preferably include analysis block nodes, optimization block nodes, and auxiliary functional block nodes. Analysis block nodes are used to call third-party pipe mechanics analysis software to perform stress calculations. Optimization block nodes are used to execute optimization algorithms. Auxiliary functional block nodes are used to perform additional data calculations, such as the proportion of standard supports and hangers, and similarity assessment with the original layout.

[0050] The analysis block node is configured to: During the process construction phase, respond to user operations by loading the pre-optimization pipeline system model and mechanical analysis results, assisting the user in identifying and defining the pipe segment range to be optimized based on actual working conditions, and transmitting the pipe segment information and support / hanger layout data within the optimization range to the optimization algorithm module. During the process execution phase, respond to the scheduling instructions of the process engine, automatically invoke third-party pipeline mechanical analysis software to perform static or dynamic stress analysis on the current support / hanger layout scheme, automatically acquire key mechanical evaluation data such as pipeline stress, flange weld stress, equipment and valve nozzle loads, acceleration at the valve's center of gravity, and through-body loads and displacements, and return this data as optimization indicators to downstream nodes for objective function verification and constraint condition evaluation.

[0051] The optimization block node is configured to receive optimization objectives and constraints from the parameter configuration interface, as well as optimization range definitions and mechanical evaluation data from the analysis block node, as input parameters for the multi-objective optimization problem.

[0052] The NSGA-II multi-objective optimization algorithm is invoked to perform non-dominated sorting and crowding calculation on the initial population, generating the Pareto front solution set for the current generation. Based on the Pareto front solution set, selection, crossover, and mutation operations are performed to generate the offspring population. The parent and offspring populations are then merged and environmental selection is performed to form a new generation population.

[0053] Determine whether the preset termination condition is met. If not, continue the iterative optimization process. Otherwise, output the final Pareto optimal solution set to the post-processing module for visualization and optimization analysis of the optimization scheme.

[0054] The auxiliary function block node is configured to perform additional data calculations, including standard support and hanger ratio calculations and similarity calculations with the pipe support and hanger layout before optimization, to support the comprehensive evaluation and optimal decision-making of the optimization scheme.

[0055] The visual process editor provides a structured representation of functional nodes, each node including a unique identifier, type identifier, location coordinates, parameter configuration information, and a set of input and output sockets.

[0056] The visual process editor supports drag-and-drop node placement, responds to user input events, and creates node instances and initializes their positions after verifying the validity of the drag source and target positions.

[0057] Step S 22 After the nodes are placed, the user can connect the nodes by operating the mouse; when the output port Socket is pressed and dragged to the input port Socket, the system automatically generates a permanent connection line.

[0058] The visual process editor supports the creation of interactive connection lines. Temporary connection lines can be initiated via mouse operations, and permanent connections are generated after successful socket docking between both parties. Simultaneously, the visual process editor also provides view control functions, supporting interface zooming via the mouse wheel and canvas panning via middle mouse button dragging, to accommodate the browsing and editing needs of large flowcharts.

[0059] In this embodiment, the connection line is used to define the data transmission relationship between nodes. The connection line is represented by an Edge data structure, where each Edge includes a unique identifier, a source Socket, a target Socket, and a path vertex sequence, used to define a directed data path between two Sockets.

[0060] The wire connection preferably uses A. The path planning algorithm enables automatic layout. It performs path search based on a gridded canvas space, transforming the wiring path of the connecting line into the problem of finding the optimal path from the source socket to the target socket.

[0061] The A The overall cost function of the path planning algorithm is defined as F = G + H + E.

[0062] Where G represents the actual path cost from the starting point to the current search point; H represents the heuristically estimated cost based on the improved Manhattan distance, with offset correction according to the relative orientation of the source Socket and the target Socket; E represents the turning penalty term, used to suppress high-frequency turning. Turning on straight segments, at the starting point, and at the end point is not included in the additional cost; path points turning along the target direction are given lower penalty values, while path points turning in non-aligned directions are given higher penalty values.

[0063] The algorithm manages candidate and visited nodes during the search process by maintaining open and closed sets, enabling obstacle avoidance path generation that bypasses already placed nodes, resulting in aesthetically pleasing and non-intersecting connection paths. The path planning results drive the automatic updating of the geometry of the connecting lines, generating smooth, low-turn, and visually appealing wiring effects. Users can zoom in and out using the mouse wheel and pan the canvas by dragging with the middle mouse button, facilitating editing operations in complex workflows.

[0064] Of course, the connection layout algorithm described in this application uses A The path planning algorithm is only an example; it can also be implemented using other path planning methods such as Dijkstra or RRT. The specific cost function can be adjusted, and all of these fall within the technical scope of this invention.

[0065] like Figure 5 As shown, A The path planning algorithm process includes the following steps: Step S 221 Initialize the search data structure.

[0066] Step S 222 Add the starting point to the open list.

[0067] Step S 223 Calculate the difference between the starting point and the ending point.

[0068] Step S 224 1. Determine if the open set is empty; if not, remove the point with the smallest F value as the current point from the open set and add it to the closed set, then proceed to step S. 225 If so, proceed to step S. 228 .

[0069] Step S 2251. Determine if the current point is the destination; if yes, return the found path; otherwise, obtain the surrounding walkable points and proceed to step S. 228 .

[0070] Step S 226 1. Determine whether all surrounding points have been traversed. If not, return to step S. 224 If so, proceed to step S. 227 .

[0071] Step S 227 1. Determine if the point is in the open set; if not, set the parent node as the current point, calculate the G, H, and F values, add it to the open set, and proceed to step S. 228 If so, check if a shorter path has been found and proceed to step S. 228 .

[0072] Step S 228 1. Determine if the destination is in the open list; if yes, return the found path; otherwise, return the path not found.

[0073] Step S3: Set optimization objectives and constraints. Extract key evaluation parameters from the mechanical analysis results file output by the third-party pipeline analysis software, select the target parameters to be optimized, set the hyperparameters of the optimization algorithm, and save the optimization configuration information in JSON format after configuration to support subsequent loading and reuse.

[0074] For example, in the "Parameters" tab, the user clicks the "Data Browser" button, adds the object to be optimized in the pop-up window, and then configures the optimization objectives and constraints in the table. The system extracts key evaluation parameters from the mechanical analysis results file output by third-party pipeline analysis software and preferably displays them in a tree structure.

[0075] In response to the optimization objectives and constraints set by the client in the "Parameters" tab, the mechanical analysis results preferably include pipe stress, flange weld stress, equipment and valve nozzle loads, valve center of gravity acceleration, and through-body loads and displacements. The optimization objectives and constraints are interchangeable.

[0076] The "Parameters" tab includes a parameterized configuration interface, a parameter status visualization module, a parameter selection module, a parameter usage configuration module, and a configuration data storage module. The parameterized configuration interface preferably includes a data browser configured to display the mechanical analysis results output by third-party pipeline analysis software in a tree structure. These mechanical analysis results include pipeline stress, equipment nozzle loads, valve acceleration, and penetration loads, with each item organized as an independent node in an expandable hierarchical structure.

[0077] The parameter status visualization module is configured to display the parameter values ​​in color based on the relationship between the parameter values ​​and preset thresholds. Red indicates that the value exceeds the limit, orange indicates that the value is close to the limit, and black indicates that the value is within the safe range, in order to help users quickly identify the parameter items that need to be optimized.

[0078] The parameter selection module is configured to respond to user operations by selecting one or more parameters from the tree structure and adding them to the parameter configuration table.

[0079] The parameter usage configuration module is configured to support dynamic switching of the usage role of the same parameter. It can be configured to be used as "only as a constraint", "only as an objective function term", and "both as an objective function and a constraint". Minimizing the number of supports and hangers is the default optimization goal and does not require additional settings by the user.

[0080] The configuration data storage module is configured to save optimization configuration information in JSON format. This optimization configuration information includes optimization algorithm hyperparameters, optimization variable definitions, and optimization target definitions. The optimization variables include adjustable support / hanger positions, types, and stiffness parameters, used to generate candidate support / hanger layout schemes during the optimization process.

[0081] Furthermore, the analysis results are structured into a hierarchical data structure by category, and qualitative identification information representing the safety status of parameters is generated based on calculated values ​​and preset thresholds. For example, red indicates exceeding the limit, orange indicates approaching the limit, and black indicates within the safe range, allowing users to quickly identify areas requiring optimization.

[0082] Users select the target parameters to be optimized, preferably including maximizing or minimizing the target, and setting the maximum and minimum values ​​of the constraints. The same parameter can be flexibly configured as "only as an objective function term", "only as a constraint condition", or "both as an objective function and a constraint condition", realizing the dynamic reconstruction of the optimization problem.

[0083] Users also need to set the hyperparameters of the optimization algorithm in the main interface. For example, these preferably include population size, maximum number of iterations, crossover probability, and mutation probability. After configuration, the system saves the optimization configuration information in JSON format for subsequent loading and reuse.

[0084] In this embodiment, the main interface of the client adopts a tab-based multi-view architecture, specifically including: The functions of process modeling, parameter configuration, execution control, and result post-processing are each encapsulated into independent tabs; The tabs are displayed side-by-side at the top or side of the interface, allowing users to quickly navigate between functional modules by clicking to switch between them; During tab switching, maintain the synchronization of context data and consistency of interface state for the current optimization task to ensure the continuity of multi-step operations and the integrity of user experience.

[0085] Step S4: The user clicks the "Run" button to start the automated execution of the optimization process.

[0086] In response to the optimization task submission operation triggered by the client in the "Execute" tab, the optimization algorithm module is invoked to perform optimization calculations. This module integrates the NSGA-II multi-objective optimization algorithm and supports the integration of other optimization algorithms. The optimization progress information, including the current algebra and the objective function convergence curve, is displayed in the real-time monitoring area within the "Execute" tab.

[0087] like Figure 6 As shown, step S4 preferably includes the following steps: Step S 41 A process engine is used to perform topology parsing, read the process definition, and construct a node dependency graph. The graphical flowchart is then parsed to extract functional nodes and their connections, constructing a node dependency graph representing the data flow.

[0088] Step S 42 A depth-first search algorithm is used to perform topological sorting to determine the execution order of nodes.

[0089] The system checks for circular dependencies in the node dependency graph. If a circular dependency is detected, an error message is displayed prompting the user to modify the process. After confirming that there are no circular dependencies, a topological sort is performed on the node dependency graph to generate a linear execution sequence that satisfies the dependency constraints.

[0090] Step S 43 The system verifies the completeness and validity of the parameters for each node, including whether required parameters have been set, whether parameter values ​​are within a valid range, and whether the input socket is connected. After successful verification, each node is executed sequentially according to the topology sorting order.

[0091] Before execution, the integrity of the parameter configurations of each node and the validity of the input / output socket connections are verified. The corresponding functional modules of each node are called sequentially according to the described linear execution sequence, thereby achieving automated scheduling and execution of the optimized process.

[0092] Step S 44 When execution reaches an optimization block node, a multi-objective optimization algorithm is invoked to obtain the computational solution set. For example, ... Figure 7 and Figure 8 The system calls the NSGA-II multi-objective optimization algorithm to perform non-dominated sorting and crowding calculation on the initial population, generating the Pareto front solution set for the current generation.

[0093] Step S 45 Based on the calculated solution set, selection, crossover, and mutation operations are performed to generate a progeny population, and a new generation population is formed through environmental selection; after iterating until the termination condition is met, the final optimal solution set is output (e.g., ...). Figure 7 and Figure 8 The NSGA-II multi-objective optimization algorithm shown obtains the Pareto optimal solution set.

[0094] Preferably, step S 44 The multi-objective optimization algorithms described include the NSGA-II multi-objective optimization algorithm. Alternatively, ensemble particle swarm optimization or differential evolution algorithms can also be used. These multi-objective optimization algorithms can all achieve global search and Pareto front generation for support and hanger layout schemes.

[0095] When the system reaches the analysis block node, it uses the interface module to enable bidirectional data transfer between the third-party pipeline mechanics analysis software and the optimization algorithm, and parses the output results to extract key mechanical indicators for optimization evaluation.

[0096] like Figure 7 As shown, the calling process of the NSGA-II multi-objective optimization algorithm includes the following steps: Step S 441 Initialize logging, read JSON configuration parameters, and obtain the interpreter path.

[0097] Step S 442 Obtain the objective function parameters and constraints, and create an optimization configuration object.

[0098] Step S 443 1. Determine if the optimization scope file exists; if so, proceed to step S. 444 If not, proceed to step S. 447 .

[0099] Step S 444 Read the optimization scope file, obtain the task type and optimization scope, and create a problem instance; obtain the checkpoint file path and create an evolutionary algorithm instance.

[0100] Step S 445 Evolutionary computation is performed by calling an evaluation method (such as the function evolve).

[0101] Step S 446 Perform data processing, output the results to a JSON file, return the results, and end the process.

[0102] Step S 447 Error message will be displayed and the process will be terminated.

[0103] like Figure 8As shown, the computation steps of the NSGA-II multi-objective optimization algorithm include: Step S 451 Load configuration from JSON file.

[0104] Step S 452 1. Confirm the problem type and analysis type.

[0105] Step S 453 1. Initialize the population and call the pipeline software to calculate the target value.

[0106] Step S 454 Based on the target value, divide the Pareto front and perform fast non-dominated sorting.

[0107] Step S 455 Based on the Pareto front, the target values ​​of the schemes are sorted and the crowding distance is calculated.

[0108] Step S 456 Perform parent selection, crossover, and mutation operations to produce offspring populations, calculate offspring target values, merge parent and offspring populations, perform fast non-dominated sorting, and select the optimal Pareto front.

[0109] Step S 457 1. Determine whether the specified number of training generations has been reached, or whether the early stopping condition has been met; if not, return to step S. 456 If so, output the optimal solution.

[0110] Step S5: During the optimization process, the user observes the optimization progress through the real-time monitoring interface in the "Execute" tab. An asynchronous signal-slot mechanism is used to decouple the optimization algorithm from the interface, ensuring the interface responds to the flow field. A multi-objective dynamic curve graph plots the convergence trend of each optimization objective in real time, including curves for the mean, maximum, and minimum values. The user judges the algorithm's convergence by observing curve changes. The progress bar updates dynamically according to the ratio of the current algebra to the total number of algebras.

[0111] The control panel provides buttons such as "Run," "Stop," "Restart," and "Resume Operation," allowing users to flexibly control the optimization process. The system saves the optimization record every 5 generations. Users can stop optimization at any time, and the system retains the most recently saved calculation state. Users can resume the most recently saved calculation state from the stopped state to continue optimization, or they can clear the current results and start over.

[0112] Step S6: After optimization, select multiple candidate solutions (e.g., select multiple candidate solutions from the Pareto frontier), compare and analyze them with the initial layout, comprehensively evaluate the technical rationality of each solution, and select the optimal solution.

[0113] Responding to client query requests initiated in the "Post-processing" tab, it displays optimization results, supports Pareto's cutting-edge visualization, queries of historical generation results, and comparative analysis between multiple candidate solutions.

[0114] Users can view the analysis results on the "Post-processing" tab, and click on any solution to view its detailed parameter configuration and simultaneously display the optimization results in a bar chart.

[0115] In addition, users can preview the evolution of Pareto fronts throughout history using the historical algebra query function and observe the optimization search path. The system supports resuming the optimization process from any checkpoint.

[0116] As described above, in the pipeline support and hanger integration optimization method based on the process engine of this invention, the user interface can be organized in the form of tabs, sidebars, cards, or multi-window layouts. As long as the functional modules are divided and operated step-by-step, an equivalent implementation is constituted. The database can be SQLite, MySQL, PostgreSQL, or other persistent storage solutions.

[0117] like Figure 9 As shown, the present invention also provides a pipeline support and hanger integration optimization system based on a process engine, employing the pipeline support and hanger integration optimization method based on a process engine as described above. The optimization system includes: The GUI module is configured to provide a graphical user interface. The GUI module preferably includes multiple sub-modules: The first module, the Model / Template Library, manages process templates and provides the ability to quickly create processes. This module is configured to manage preset process templates and typical optimized models, allowing users to quickly call and reuse them.

[0118] II. The modeling space submodule provides a visual process design environment. The runtime monitoring module monitors the process execution status and optimization progress in real time. The post-processing module analyzes and visualizes the process execution results. The modeling space submodule is configured to provide a visual process design environment, supporting drag-and-drop placement of functional nodes, automatic layout and routing of connecting lines, and zoom and pan operations of the view.

[0119] The third module, the monitoring module, is configured to display the execution status of optimization tasks in real time, including the convergence trend curve of the multi-objective function, the current algebra, and the optimization progress bar.

[0120] IV. The post-processing submodule is configured to analyze and visualize the optimization results, supporting Pareto frontier distribution presentation, comparative analysis between multiple optimization schemes, and backtracking query of historical generation results.

[0121] The process engine module is configured to respond to user-constructed graphical processes, responsible for process parsing and execution. The process engine module preferably includes an interpreter, an executor, and a node editor. The interpreter is responsible for parsing the process definition and handling the logical relationships between nodes and connections. The executor is responsible for executing the functions of each node according to the interpreted process definition. The node editor provides functions such as node creation configuration, socket connection management, and Edge path planning.

[0122] More preferably, the process engine module is configured to: when executing an analysis block node, call the interface module to complete pre- and post-processing interactions with the pipeline mechanics analysis software; when executing an optimization block node, schedule the optimization algorithm module to generate a new support and hanger layout scheme; and when executing an auxiliary function block node, calculate the standard support and hanger usage ratio or the similarity index with the original layout to support the comprehensive evaluation of the optimization scheme.

[0123] The optimization algorithm module is configured to invoke a multi-objective optimization algorithm to iteratively optimize the pipe support and hanger layout scheme and generate an optimal solution set (e.g., a Pareto optimal solution set). The optimization algorithm module preferably includes: NSGA-II optimization algorithm configuration for providing algorithm parameter settings; single-objective optimization support for scenarios with a single optimization objective; multi-objective optimization support for simultaneous optimization of multiple objective functions; and optimization progress monitoring for real-time monitoring of the optimization progress.

[0124] The interface module provides an interface with pipeline finite element analysis software. It is configured to enable bidirectional data interaction with third-party pipeline mechanics analysis software (such as Pipestress and CAESAR II) via a standardized data interface, and to integrate with Python scripts, supporting the import of pipeline models and the export of optimization results. Preferably, the interface module includes: a pipeline finite element analysis software interface, supporting data interaction with the software via Python scripts, and providing a script execution environment, parameter passing, and result parsing functions.

[0125] The database module is configured to persistently store and manage project information, process definitions, node parameters, and historical optimization data. This database module is implemented based on SQLite and preferably includes: project data storage for managing basic project information; process data storage for storing process definitions and configurations; and optimization result storage for saving and managing optimization calculation results.

[0126] More preferably, the database module is configured to include: multiple data tables for storing project information, process definitions, node configurations, connection relationships, optimization parameter input / output mappings, and visualization chart configurations. The data tables include Projects, Flows, Nodes, Edges, Configs, ParamLinkInput, ParamLinkOutput, and Tablecharts. The main data tables in each table have an `is_deleted` field to mark the logical deletion status of records, and deleted records are excluded by default in query operations, implementing a soft deletion function. Non-unique indexes are created on foreign key fields and frequently queried fields to support fast data retrieval and related queries.

[0127] The present invention also provides an electronic device, comprising: a processor and a memory, wherein the memory stores a program or instructions executable on the processor, the program or instructions being executed by the processor to implement the process engine-based pipe support and hanger integration optimization method as described above.

[0128] The present invention also provides a readable storage medium, characterized in that a program or instruction is stored on the readable storage medium, and when the program or instruction is executed by a processor, it implements the pipeline support and hanger integration optimization method based on the process engine as described above.

[0129] Based on the above description, the pipe support and hanger integrated optimization method and system of the present invention, based on a process engine, transforms the optimization of pipe supports and hangers from "manual experience-driven" to "process automation-driven" through the process engine, and can achieve the following functions: I. The collaborative mechanism between the visual process editor and the process engine enables the optimization of process modeling, parsing, and automatic execution; Second, the analysis of the auxiliary definition function of block nodes in the process construction stage for optimizing the scope reflects the intelligent guidance capability; Third, parameters can be dynamically configured as targets or constraints on the same interface, improving optimization flexibility; IV. Connecting cables use A The algorithm enables automatic obstacle avoidance layout, improving the efficiency of graphic editing; V. Deep integration with system and pipeline mechanics analysis software (Pipestress, CAESAR II) and the ability to write back optimization solutions.

[0130] For those skilled in the art, the above disclosure is merely illustrative and does not constitute a limitation of this application. Although not explicitly stated herein, those skilled in the art may make various modifications, improvements, and corrections to this application. Such modifications, improvements, and corrections are suggested in this application and therefore remain within the spirit and scope of the exemplary embodiments of this application.

[0131] Furthermore, this application uses specific terms to describe embodiments of the application. For example, "an embodiment," "one embodiment," and / or "some embodiments" refer to a particular feature, structure, or characteristic related to at least one embodiment of the application. Therefore, it should be emphasized and noted that "an embodiment," "one embodiment," or "an alternative embodiment" mentioned twice or more in different locations in this specification do not necessarily refer to the same embodiment. In addition, certain features, structures, or characteristics in one or more embodiments of the application can be appropriately combined.

[0132] Similarly, it should be noted that, in order to simplify the description of the embodiments disclosed in this application and thus aid in the understanding of one or more embodiments of the invention, the foregoing description of the embodiments of this application sometimes combines multiple features into a single embodiment, drawing, or description thereof. However, this disclosure method does not imply that the subject matter of this application requires more features than those mentioned in the claims. In fact, the embodiments have fewer features than all the features of a single embodiment disclosed above. Some embodiments use numbers describing the number of components or attributes; it should be understood that such numbers used in the description of embodiments are modified in some examples by the terms "approximately," "about," or "generally."

[0133] While specific embodiments of the present invention have been described above, those skilled in the art should understand that these are merely illustrative examples, and the scope of protection of the present invention is defined by the appended claims. Those skilled in the art can make various changes or modifications to these embodiments without departing from the principles and essence of the present invention, but all such changes and modifications fall within the scope of protection of the present invention.

Claims

1. A pipeline support and hanger integration optimization method based on a process engine, characterized in that, The process engine-based pipe support and hanger integration optimization method includes the following steps: S1. Based on the SQLite database, persistently store project information and create or open at least one optimized project; S2. Build and optimize processes using a visual process editor; The optimization process in step S2 includes: S 21 Users drag and drop functional nodes from the node library into the modeling space; S 22 After the nodes are placed, the user can connect the nodes by using the mouse; when the output port Socket is pressed and dragged to the input port Socket, the system automatically generates a permanent connection line. The step S 21 The functional nodes include analysis block nodes, optimization block nodes, and auxiliary function block nodes; S3. Set optimization objectives and constraints, extract key evaluation parameters from the mechanical analysis results file output by the third-party pipeline analysis software, select the target parameters to be optimized, set the hyperparameters of the optimization algorithm, and save the optimization configuration information in JSON format after configuration to support subsequent loading and reuse. S4. Initiate the automated execution of the optimization process; S5. During the optimization process, an asynchronous signal slot mechanism is used to achieve decoupled communication between the optimization algorithm and the interface. The multi-objective dynamic curve graph draws the convergence trend of each optimization objective in real time, and the user judges the convergence of the algorithm by the curve changes. S6. After optimization, select multiple candidate schemes, compare and analyze them with the initial layout, comprehensively evaluate the technical rationality of each scheme, and select the optimal scheme.

2. The pipe support and hanger integration optimization method based on process engine as described in claim 1, characterized in that, The construction of the optimization process in step S2 includes: the user selects a typical optimization process template from the preset template library and generates a complete optimization process diagram with one click; after the user selects a template, the system automatically generates a complete process containing all necessary nodes and connections.

3. The pipeline support and hanger integration optimization method based on process engine as described in claim 1, characterized in that, The analysis block node is configured to: in the process construction stage, respond to user operations to load the pipeline system model and mechanical analysis results before optimization, assist the user in identifying and defining the range of pipe sections to be optimized based on actual working conditions, and transmit the pipe section information and support and hanger layout data within the optimization range to the optimization algorithm module; During the process execution phase, in response to the scheduling instructions of the process engine, the system automatically calls third-party pipeline mechanics analysis software to perform static or dynamic stress analysis on the current support and hanger layout scheme, automatically acquires mechanical evaluation data, and returns the data as optimization indicators to the downstream node for objective function verification and constraint condition evaluation.

4. The pipeline support and hanger integration optimization method based on process engine as described in claim 1, characterized in that, The optimization block node is configured to receive optimization objectives and constraints from the parameter configuration interface, as well as optimization range definitions and mechanical evaluation data from the analysis block node, as input parameters for the multi-objective optimization problem. A multi-objective optimization algorithm is invoked to perform non-dominated sorting and crowding calculation on the initial population, generating the Pareto front solution set for the current generation; Based on the Pareto front solution set, selection, crossover, and mutation operations are performed to generate offspring populations. The parent and offspring populations are then merged and environmental selection is performed to form a new generation population. Determine whether the preset termination condition is met. If not, continue the iterative optimization process. Otherwise, output the final Pareto optimal solution set to the post-processing module for visualization and optimization analysis of the optimization scheme.

5. The pipeline support and hanger integration optimization method based on process engine as described in claim 1, characterized in that, The auxiliary function block node is configured to perform additional data calculations, including standard support and hanger ratio calculations and similarity calculations with the pipe support and hanger layout before optimization, in order to support the comprehensive evaluation and optimal decision-making of the optimization scheme.

6. The pipeline support and hanger integration optimization method based on process engine as described in claim 1, characterized in that, The step S 22 The center line connection uses A Path planning algorithms enable automatic layout, wherein A The overall cost function of the path planning algorithm is defined as F = G + H + E; Where G represents the actual path cost from the starting point to the current search point; H represents the heuristically estimated cost based on the improved Manhattan distance, with offset correction based on the relative orientation of the source Socket and the target Socket; and E represents the turning penalty term, used to suppress high-frequency turning.

7. The pipeline support and hanger integration optimization method based on process engine as described in claim 6, characterized in that, The A The path planning algorithm process includes the following steps: S 221 Initialize the search data structure; S 222 Add the starting point to the open list; S 223 Calculate the difference between the starting point and the ending point; S 224 1. Determine if the open set is empty; if not, remove the point with the smallest F value as the current point from the open set and add it to the closed set, then proceed to step S. 225 If so, proceed to step S. 228 ; S 225 1. Determine if the current point is the destination; if yes, return the found path; otherwise, obtain the surrounding walkable points and proceed to step S. 228 ; S 226 1. Determine whether all surrounding points have been traversed. If not, return to step S. 224 If so, proceed to step S. 227 ; S 227 1. Determine if the point is in the open set; if not, set the parent node as the current point, calculate the G, H, and F values, add it to the open set, and proceed to step S. 228 If so, check if a shorter path has been found and proceed to step S. 228 ; S 228 1. Determine if the destination is in the open list; if yes, return the found path; otherwise, return the path not found.

8. The pipeline support and hanger integration optimization method based on process engine as described in claim 1, characterized in that, The mechanical analysis results in step S3 include: pipeline stress, flange weld stress, equipment and valve pipe load, valve mass center of gravity acceleration, and through-body load and displacement. The data are constructed into a hierarchical data structure according to categories, and qualitative identification information characterizing the safety status of the parameters is generated based on the calculated values ​​and preset thresholds.

9. The pipe support and hanger integration optimization method based on process engine as described in claim 1, characterized in that, The target parameters to be optimized in step S3 include: the maximum or minimum value of the target, and the maximum and minimum values ​​of the constraints.

10. The pipeline support and hanger integration optimization method based on process engine as described in claim 1, characterized in that, The hyperparameters of the optimization algorithm in step S3 include: population size, maximum number of iterations, crossover probability, and mutation probability.

11. The pipeline support and hanger integration optimization method based on process engine as described in claim 1, characterized in that, Step S4 includes the following steps: S 41 Topology parsing is performed using a process engine; S 42 A depth-first search algorithm is used to perform topological sorting to determine the execution order of nodes; S 43 The system verifies the completeness and validity of the parameters of each node and executes each node sequentially. S 44 When the execution reaches the optimization block node, the multi-objective optimization algorithm is called to obtain the computational solution set; S 45 Based on the calculated solution set, selection, crossover, and mutation operations are performed to generate offspring populations, and a new generation population is formed through environmental selection; after iterating until the termination condition is met, the final optimal solution set is output.

12. The pipeline support and hanger integration optimization method based on process engine as described in claim 11, characterized in that, The step S 44 The multi-objective optimization algorithm described herein is the NSGA-II multi-objective optimization algorithm.

13. The pipeline support and hanger integration optimization method based on process engine as described in claim 12, characterized in that, The invocation process of the NSGA-II multi-objective optimization algorithm includes the following steps: S 441 Initialize logging, read JSON configuration parameters, and obtain the interpreter path; S 442 Obtain the objective function and constraints, and create an optimization configuration object; S 443 1. Determine if the optimization scope file exists; if so, proceed to step S. 444 If not, proceed to step S. 447 ; S 444 Read the optimization scope file, obtain the task type and optimization scope, and create a problem instance; obtain the checkpoint file path and create an evolutionary algorithm instance; S 445 , Call the evaluation method to perform evolutionary calculations; S 446 Perform data processing, output the results to a JSON file, return the results, and end the process; S 447 Error message will be displayed and the process will be terminated.

14. The pipeline support and hanger integration optimization method based on process engine as described in claim 12, characterized in that, The computation steps of the NSGA-II multi-objective optimization algorithm include: S 451 Load configuration from JSON file; S 452 1. Identify the problem type and analysis type; S 453 Initialize the population and call the pipeline software to calculate the target value; S 454 Based on the target value, divide the Pareto front and perform fast non-dominated sorting; S 455 Based on the Pareto front, the target values ​​of the proposed schemes are sorted and the crowding distance is calculated. S 456 Perform parent selection, crossover and mutation operations to produce offspring populations, calculate offspring target values, merge parent and offspring populations, perform fast non-dominated sorting, and select the best Pareto front. S 457 1. Determine whether the specified number of training generations has been reached, or whether the early stopping condition has been met; if not, return to step S. 456 If so, output the optimal solution.

15. A pipe support and hanger integrated optimization system based on a process engine, characterized in that, The process engine-based pipe support and hanger integrated optimization system adopts the process engine-based pipe support and hanger integrated optimization method as described in any one of claims 1-14, and the optimization system includes: The GUI module is used to provide a graphical user interface. The process engine submodule is used to respond to user-built graphical processes and is responsible for process parsing and execution. The optimization algorithm module is used to call multi-objective optimization algorithms to generate the optimal solution set; The interface module provides an interface with pipeline finite element analysis software, configured to enable bidirectional data interaction with third-party pipeline mechanics analysis software via a standardized data interface, as well as Python script integration functionality. The database module is used for persistent storage and management of project information, process definitions, node parameters, and historical optimization data.

16. An electronic device, characterized in that, The electronic device includes a processor and a memory, the memory storing programs or instructions that can run on the processor, the programs or instructions being executed by the processor to implement the pipeline support and hanger integration optimization method based on any one of claims 1-14.

17. A readable storage medium, characterized in that, The program or instructions are stored on the readable storage medium, and when the program or instructions are executed by the processor, they implement the pipeline support and hanger integration optimization method based on any one of claims 1-14.

Images