Intelligent remote control method and system for special valve

Abstract

The application relates to the technical field of valve control, and discloses a remote control method and system for intelligent special valves. The method comprises the following steps: acquiring valve position distribution data and fluid initial flow parameter, simulating the influence of a plurality of valve opening combinations and generating a preliminary curvature distribution diagram; calculating a branch pipeline flow curvature value to determine a high curvature area; extracting a key valve combination influence factor; optimizing the curvature distribution diagram and acquiring a valve opening sequence with minimum overall curvature; judging whether a dynamic characteristic conflict exists; if the conflict exists, rearranging and generating a conflict-free control sequence by using a particle swarm optimization algorithm; simulating the sequence flow change to determine a final adjustment scheme, generating an energy loss prediction value, and outputting a cooperative control instruction if the energy loss prediction value is lower than a threshold value. Through simulation analysis and algorithm optimization, the application solves the problems of uneven curvature distribution and dynamic conflict in the traditional method, significantly improves the fluid flow efficiency and stability of the pipeline system, and reduces the operation energy consumption.

Description

TECHNICAL FIELD

[0001] The present application relates to the technical field of valve control, and in particular to a remote control method and system for intelligent special valves. BACKGROUND

[0002] In industrial production, complex pipeline systems need to be adjusted by multiple valves to meet process requirements. Fluid flow efficiency, system stability and energy consumption control are the core indicators to measure the advantages and disadvantages of valve control technology. The current industry urgently needs a control scheme that can balance the coordination accuracy of multiple valves and the operation efficiency of the system.

[0003] The existing valve control technology for complex pipeline systems has significant defects. On the one hand, traditional methods cannot accurately identify the contribution weight of key valves to fluid flow. The complexity of multiple valve opening combinations easily leads to uneven distribution of fluid flow curvature in branch pipelines. High curvature areas are prone to cause local turbulence, which greatly increases energy loss and cannot balance fluid distribution and energy efficiency by optimizing valve opening sequences. On the other hand, traditional methods lack comprehensive analysis of path dependence and time dependence. The dynamic characteristics of different valve opening sequences are prone to conflict, resulting in vibration, pressure surge or unbalanced flow distribution during system operation, reducing long-term operation stability, and even causing pipeline fatigue or equipment damage. Especially in complex pipeline networks with high load and multiple branches, the above problems are more prominent, which restricts the improvement of the overall operation efficiency of the pipeline system.

[0004] To overcome the above-mentioned defects, the present application provides a technical route for generating a curvature distribution map, determining a high curvature area, extracting an influence factor, optimizing a valve opening sequence, solving dynamic conflicts and verifying energy loss. The present application solves the problems of uneven curvature distribution, dynamic characteristic conflict and inaccurate identification of key valve contribution weight in traditional methods, improves the fluid flow efficiency and operation stability of the pipeline system, and reduces the energy consumption of the system operation, thereby providing an efficient and coordinated control scheme for complex pipeline systems. SUMMARY

[0005] The present application provides a remote control method and system for intelligent special valves, which solves the problems of uneven curvature distribution, dynamic characteristic conflict and inaccurate identification of key valve contribution weight in traditional methods, improves the fluid flow efficiency and operation stability of the pipeline system, and reduces the energy consumption of the system operation.

[0006] In a first aspect, the present application provides a remote control method for intelligent special valves, which comprises:

[0007] Step S101, acquiring valve position distribution data and fluid initial flow parameter, and simulating the influence of multiple valve opening combinations on fluid flow path, and generating a preliminary curvature distribution map according to the influence;

[0008] Step S102, according to the preliminary curvature distribution map, calculate the flow curvature value in each branch pipeline, determine the high curvature area;

[0009] Step S103, extract the influence factor of the key valve combination from the high curvature area;

[0010] Step S104, optimize the preliminary curvature distribution map according to the influence factor, generate the optimized curvature distribution map, and obtain the valve opening sequence with minimum overall curvature based on the optimized curvature distribution map;

[0011] Step S105, judge whether the valve opening sequence exists dynamic characteristic conflict;

[0012] Step S106, if the valve opening sequence exists dynamic characteristic conflict, use particle swarm optimization algorithm to rearrange the valve opening sequence, and generate conflict-free control sequence;

[0013] Step S107, simulate the real-time fluid flow change of the conflict-free control sequence, determine the final valve regulation scheme, generate energy loss prediction value, judge whether the energy loss prediction value is lower than the preset loss threshold, if yes, output multi-valve cooperative control instruction.

[0014] Optionally, the step S101 comprises:

[0015] Obtain the position distribution data of multiple valves and the initial fluid flow parameter;

[0016] According to the position distribution data and the initial fluid flow parameter, simulate the influence of different valve opening combinations on the fluid flow path;

[0017] According to the influence of different valve opening combinations on the fluid flow path, generate the preliminary curvature distribution map, and the preliminary curvature distribution map contains curvature peak value coordinates and intensity gradient distribution;

[0018] Determine the key nodes in the flow path based on the curvature peak value coordinates;

[0019] According to the intensity gradient distribution, extract the contribution weight of multiple valves to fluid flow.

[0020] Optionally, the step S102 comprises:

[0021] Use finite element analysis to process the preliminary curvature distribution map;

[0022] Based on the preliminary curvature distribution map processed by finite element analysis, calculate the flow curvature value in multiple branch pipelines;

[0023] The location and intensity of the high curvature region are determined based on multiple flow curvature values;

[0024] Valve-related weights are extracted from the intensity gradient distribution contained in the preliminary curvature distribution map.

[0025] Optionally, step S103 includes:

[0026] Based on the location and intensity data of the high curvature region, path dependency indicators are extracted;

[0027] Calculate the local attenuation rate of each segment of the fluid flow path corresponding to the path dependence index, and average the local attenuation rate of each segment to generate the intensity attenuation rate corresponding to the path dependence index.

[0028] Determine whether the intensity attenuation rate exceeds a preset attenuation rate threshold. If so, adjust the simulation parameters; and regenerate the curvature distribution map based on the adjusted simulation parameters.

[0029] The location and intensity data of high curvature regions are obtained from the regenerated curvature distribution map. The valve combination that controls the fluid flow in this region is associated with it. The contribution of the valve combination to the curvature change in the high curvature region is calculated. The correspondence between the contribution and the valve combination is integrated to generate the influence factor of the key valve combination.

[0030] Optionally, step S104 includes:

[0031] Based on the aforementioned influencing factors, the weight of each valve's effect on the fluid flow path is adjusted to generate the optimized curvature distribution map.

[0032] Based on the optimized curvature distribution map, the overall curvature value of multiple valve opening sequences is calculated.

[0033] By comparing the overall curvature values ​​of all valve opening sequences, the minimum overall curvature is selected, the valve opening sequence corresponding to the minimum overall curvature is determined, and it is used as the initial target valve opening sequence.

[0034] Determine whether the preliminary target valve opening sequence meets the preset sequence constraint conditions. If not, adjust the valve opening order, recalculate the overall curvature value of each valve opening sequence after adjustment, and generate a valve opening sequence that meets the preset sequence constraint conditions and has the smallest overall curvature value.

[0035] Optionally, step S105 includes:

[0036] Obtain the opening interval data in the valve opening sequence;

[0037] determining whether there is data in the opening interval data that is less than a preset opening interval constraint, if yes, determining that the valve opening sequence has a dynamic characteristic conflict, and if no, determining that the valve opening sequence does not have a dynamic characteristic conflict;

[0038] if it is determined that there is a dynamic characteristic conflict, extracting a conflict detection rule, analyzing a conflict reason according to the conflict detection rule, and generating a conflict detection report, wherein the conflict detection rule includes a corresponding relationship between an opening interval and a fluid curvature peak value, and an association rule between a valve combination and a pressure fluctuation, and the conflict detection report includes a conflict point and a corresponding valve combination.

[0039] Optionally, the step S106 comprises:

[0040] initializing particle swarm parameters, mapping each particle to a set of valve opening sequences to be optimized, and setting an iteration number, a learning factor, and an inertia weight parameter;

[0041] integrating permutation priority to construct a fitness function, and the fitness function aims to minimize a dynamic characteristic conflict;

[0042] performing a mutation operation on a valve opening sequence corresponding to an optimal particle in the group, and screening a sequence after the mutation and not having a dynamic characteristic conflict as a first candidate sequence containing a sequence mutation path;

[0043] calculating a dynamic adjustment vector of the first candidate sequence, performing weighted summation on each component of the dynamic adjustment vector to generate a sequence stability index, and screening a first candidate sequence with the stability index lower than a preset stability threshold value as a second candidate sequence;

[0044] selecting, from the second candidate sequence, a sequence with a difference between an overall curvature and a determined minimum overall curvature less than a preset difference threshold value as the conflict-free control sequence.

[0045] Optionally, in the step S107, determining a final valve regulation scheme comprises:

[0046] simulating real-time fluid flow changes in a pipeline when each valve is opened under the conflict-free control sequence;

[0047] adjusting opening duration and opening amplitude of the valves in the conflict-free control sequence according to a simulation result to form a preliminary valve regulation scheme;

[0048] verifying whether the preliminary valve regulation scheme meets a turbulent flow control requirement of a high curvature region, if not, continuously adjusting the opening duration and the opening amplitude of the valves until a final valve regulation scheme that meets the turbulent flow control requirement is generated.

[0049] Optionally, the step S107 outputs the multi-valve coordinated control instruction, including:

[0050] extracting pipeline geometric parameters corresponding to the final valve adjustment scheme to generate geometric constraint conditions;

[0051] calculating an energy loss prediction value of the pipeline system under the final valve adjustment scheme based on the geometric constraint conditions;

[0052] determining whether the energy loss prediction value is lower than a preset loss threshold, and if so, generating a preliminary multi-valve coordinated control instruction based on the final valve adjustment scheme;

[0053] verifying whether the preliminary multi-valve coordinated control instruction can optimize the local turbulence of the micro-branch, ensuring that the instruction execution effect meets the pipeline system stability requirements, and finally outputting the multi-valve coordinated control instruction.

[0054] In a second aspect, the present application provides a remote control system for intelligent special valves, which comprises:

[0055] a curvature distribution generation module for acquiring valve position distribution data and fluid initial flow parameter, simulating the influence of multi-valve opening combination on fluid flow path, and generating a preliminary curvature distribution map;

[0056] a high-curvature-area determination module for calculating flow curvature values of each branch pipeline according to the preliminary curvature distribution map and determining a high-curvature-area;

[0057] an influence factor extraction module for extracting influence factors of key valve combinations from the high-curvature-area;

[0058] a valve opening sequence acquisition module for generating a curvature distribution map according to the influence factors and acquiring a valve opening sequence with minimum overall curvature;

[0059] a dynamic characteristic conflict determination module for determining whether the valve opening sequence has dynamic characteristic conflict;

[0060] a sequence rearrangement module for rearranging the valve opening sequence using a particle swarm optimization algorithm to generate a conflict-free control sequence if there is conflict;

[0061] a valve adjustment scheme generation module for simulating real-time fluid flow changes of the conflict-free control sequence, determining a final valve adjustment scheme, generating an energy loss prediction value, and outputting a multi-valve coordinated control instruction if the value is lower than a threshold.

[0062] The application provides a remote control method and system for intelligent special valves, solves the problems of uneven curvature distribution, dynamic characteristic conflict and inaccurate identification of key valve contribution weight in traditional methods, improves the fluid flow efficiency and operation stability of the pipeline system, and reduces the system operation energy consumption. Compared with the prior art, the beneficial effects of the technical scheme of the application are at least as follows:

[0063] First, by generating a curvature distribution graph, determining a high curvature area and extracting a key valve combination influence factor, the contribution weight of the key node and valve is accurately identified, solving the problem that the traditional method cannot accurately identify the influence weight of the key valve.

[0064] Second, by optimizing the curvature distribution graph to obtain the valve opening sequence with the minimum overall curvature, the uneven fluid flow curvature distribution is improved to reduce local turbulence, solving the problem of increased energy loss due to uneven curvature in the traditional method.

[0065] Third, by judging the dynamic characteristic conflict and using an optimization algorithm to rearrange the sequence, the problems of system vibration, pressure mutation and the like are avoided, solving the problem of insufficient stability caused by the lack of path and time sequence dependence analysis in the traditional method.

[0066] Fourth, by simulating the conflict-free sequence, generating an energy loss prediction value and outputting a cooperative control instruction, the micro-branch local turbulence is optimized to further reduce energy consumption and improve flow efficiency, solving the problem that the traditional method cannot balance fluid distribution and energy efficiency. BRIEF DESCRIPTION OF DRAWINGS

[0067] In order to more clearly illustrate the technical solutions of the embodiments of the application, the following will briefly introduce the drawings needed to be used in the embodiment description. Obviously, the drawings in the following description are some embodiments of the application, and other drawings can be obtained by those skilled in the art without creative labor based on these drawings.

[0068] Figure 1 It is a flowchart of the remote control method for intelligent special valves in the application;

[0069] Figure 2 It is a structural schematic diagram of the remote control system for intelligent special valves in the application. DETAILED DESCRIPTION

[0070] The embodiments of the present application provide a remote control method and system for intelligent special valves. The terms "first", "second", "third", "fourth" and the like (if any) in the specification and claims of the present application and the above-described drawings are used to distinguish similar objects, and do not have to be used to describe a particular order or sequence. It should be understood that the data thus used can be interchanged under appropriate circumstances, so that the embodiments described herein can be implemented in an order other than that illustrated or described herein. In addition, the terms "comprise" or "have" and any variations thereof are intended to cover non-exclusive inclusion, for example, a process, method, system, product or device comprising a series of steps or units does not have to be limited to only those steps or units clearly listed, but can include other steps or units not clearly listed or inherent to these processes, methods, products or devices.

[0071] For ease of understanding, the specific processes of the embodiments of the present application are described below. Please refer to Figure 1 One embodiment of the remote control method for intelligent special valves in the embodiments of the present application includes the following steps.

[0072] In step S101, the position distribution data of the valves and the initial flow parameter of the fluid are obtained, and the influence of the opening combination of the multiple valves on the fluid flow path is simulated, and a preliminary curvature distribution map is generated according to the influence.

[0073] In one specific embodiment, the step S101 can specifically include the following steps.

[0074] The position distribution data of the multiple valves and the initial flow parameter of the fluid are obtained;

[0075] According to the position distribution data and the initial flow parameter of the fluid, the influence of different valve opening combinations on the fluid flow path is simulated;

[0076] According to the influence of the different valve opening combinations on the fluid flow path, the preliminary curvature distribution map is generated, and the preliminary curvature distribution map contains curvature peak coordinates and intensity gradient distribution;

[0077] Based on the curvature peak coordinates, the key nodes in the flow path are determined;

[0078] According to the intensity gradient distribution, the contribution weights of the multiple valves to the fluid flow are extracted.

[0079] Specifically, the position distribution data of multiple valves and the fluid initial flow parameter are acquired, the position distribution data is acquired through a three-dimensional coordinate model of a pipeline system, and contains the spatial coordinates of each valve in the pipeline network and the identification information of the connected branch pipeline; the fluid initial flow parameter is collected through the flow sensor and pressure sensor deployed on the pipeline, and covers real-time data such as fluid flow rate and pipeline internal pressure. When simulating the influence of different valve opening combinations on the fluid flow path according to the position distribution data and the fluid initial flow parameter, first, the pipeline network topology graph is constructed according to the position distribution data, the coordinates of each valve are associated with the corresponding branch pipeline, and the connection structure of the pipeline network is determined; then, the finite volume method is used as the fluid flow simulation algorithm, the flow rate and pressure values in the fluid initial flow parameter are used as the simulation input boundary conditions, the valve opening combinations are enumerated, including single valve full opening, multiple valve different opening degree combinations (such as half opening, 1 / 3 opening) and the like, the numerical calculation of the flow state of the fluid in the pipeline under each combination is performed, and the velocity vector, flow path deflection angle and other data of the fluid in each branch pipeline are obtained, and the change degree of the flow path caused by different valve opening combinations is quantified through these data.

[0080] A preliminary curvature distribution map is generated according to the influence of different valve opening combinations on the fluid flow path, the curvature values of the flow paths in each branch pipeline are calculated based on the velocity vector and path deflection angle data obtained through the above simulation, and the curvature value calculation formula is (wherein is the curvature value, is the included angle of the velocity directions of two adjacent points on the path, is the path length between the two points); the calculated curvature value is matched with the corresponding pipeline position coordinates, a preliminary curvature distribution map is generated by using the heat map drawing method, the curvature peak coordinates in the map are the position coordinates with the maximum curvature value, and the intensity gradient distribution is obtained by calculating the difference between the curvature values of adjacent positions, and reflects the spatial change rate of the curvature value in the pipeline network. The key nodes in the flow path are determined based on the curvature peak coordinates, and the pipeline position corresponding to the curvature peak coordinates is directly marked as a key node, and these nodes are the positions with the maximum bending degree in the fluid flow. The contribution weights of multiple valves to the fluid flow are extracted according to the intensity gradient distribution, first, the intensity gradient vector field is constructed, the gradient vector component corresponding to the position of each valve is calculated, the preliminary contribution weight is obtained by integrating the gradient vector components of all valves; then, the preliminary contribution weight is normalized so that the sum of the contribution weights of all valves is 1, the correlation between the normalized weight and the simulation results of different valve opening combinations is fitted through the least square method, and the weight value is corrected to improve the simulation accuracy.

[0081] In the process, the position distribution data corresponds to the pipeline network topology graph, the fluid initial flow parameter corresponds to the simulation algorithm input, different valve opening combinations correspond to different flow state simulation results, the simulation results correspond to the curvature value calculation, the curvature value and the position coordinate correspond to the preliminary curvature distribution map generation, the curvature peak value coordinate corresponds to the key node determination, the intensity gradient distribution corresponds to the valve contribution weight extraction, and the complete data corresponding chain is formed. By extracting the valve contribution weight, the influence proportion of each valve on the fluid flow is clear, which provides a basis for subsequent optimization of valve opening sequence; by generating the preliminary curvature distribution map, the high curvature area is located, which lays a foundation for subsequent reduction of local turbulence and energy loss.

[0082] In step S102, according to the preliminary curvature distribution map, the flow curvature values in each branch pipeline are calculated, and a high curvature area is determined.

[0083] In a specific embodiment, the step S102 can specifically include the following steps:

[0084] The preliminary curvature distribution map is processed by using finite element analysis;

[0085] Based on the preliminary curvature distribution map processed by finite element analysis, the flow curvature values in a plurality of branch pipelines are calculated;

[0086] According to a plurality of flow curvature values, the position and intensity of a high curvature area are determined;

[0087] From the intensity gradient distribution contained in the preliminary curvature distribution map, a valve associated weight is extracted.

[0088] Specifically, when the preliminary curvature distribution map is processed by using finite element analysis, the preliminary curvature distribution map is first imported into the pipeline system grid model constructed by the finite element analysis software. The grid model is divided into units according to the actual size of the pipeline, and each unit corresponds to a section of the branch pipeline in the pipeline system. Then, the curvature data in the preliminary curvature distribution map is mapped to each grid unit, and the interpolation algorithm of the finite element analysis software is used to discretize the curvature data, so that each grid unit corresponds to a discretized curvature initial value. At the same time, combined with the material parameters, fluid viscosity coefficient and other basic data of the pipeline system, boundary conditions are provided for subsequent flow curvature value calculation.

[0089] Based on the preliminary curvature distribution map processed by finite element analysis, the flow curvature values in a plurality of branch pipelines are calculated, taking the discretized curvature initial value of each grid unit as the basis, calling the fluid dynamics calculation module in the finite element analysis, inputting the fluid initial flow parameter (such as flow rate V, pressure P), and solving the Navier-Stokes equation (u · ∇u) = -∇p + μ∇2u + f by using the finite element analysis software. wherein, is the fluid density, , viscosity of fluid power, The velocity field distribution of the fluid in each grid cell is obtained by calculating the flow curvature value of the grid cell corresponding to each branch pipe according to the velocity field distribution The calculation formula is The flow curvature value of the branch pipe is obtained by integrating and averaging the flow curvature values of all grid cells on the same branch pipe. The position and intensity of the high curvature region are determined according to a plurality of flow curvature values. The average value of the flow curvature values of all branch pipes is calculated, the high curvature threshold is set to 1.2 times the average value, the branch pipes with flow curvature values greater than the threshold are marked as to-be-analyzed pipes, and the spatial coordinates of these to-be-analyzed pipes in the pipe system are located. The to-be-analyzed pipe segments with continuous coordinates are defined as high curvature regions, and the intensity of the high curvature region is represented by the maximum value of the flow curvature values of all grid cells in the region.

[0090] The valve correlation weight is extracted from the intensity gradient distribution contained in the preliminary curvature distribution map. The gradient vector of the intensity gradient distribution is calculated to obtain the gradient vector of each position, and the coordinates of each valve in the pipe system are obtained. The gradient vector module value at each valve coordinate is calculated, and the gradient vector module values of all valves are summed. The correlation weight of a single valve is the gradient vector module value at each valve coordinate divided by the sum of the gradient vector module values of all valves.

[0091] In this process, the preliminary curvature distribution map corresponds to the input data of finite element analysis, the finite element processing result corresponds to the basic data of flow curvature value calculation, the flow curvature value corresponds to the positioning basis of the high curvature region, and the intensity gradient distribution corresponds to the extraction source of the valve correlation weight. By calculating the flow curvature value and determining the high curvature region, the turbulent flow risk point can be accurately located. The extraction of the valve correlation weight can clearly associate the valve with the curvature distribution, provide data support for subsequent optimization of the valve opening sequence, and reduce energy loss caused by improper valve control.

[0092] Step S103, extracting an influence factor of a key valve combination from the high curvature region.

[0093] In a specific embodiment, the step S103 can specifically include the following steps:

[0094] According to the position and intensity data of the high curvature region, a path dependence index is extracted;

[0095] The local attenuation rate of each segment of the fluid flow path corresponding to the path dependence index is calculated, and the average value of the local attenuation rates of each segment is calculated to generate the intensity attenuation rate corresponding to the path dependence index;

[0096] determining whether the intensity decay rate exceeds a preset decay rate threshold, and if so, adjusting simulation parameters; and regenerating the curvature distribution map according to the adjusted simulation parameters;

[0097] obtaining position and intensity data of the high-curvature region from the regenerated curvature distribution map, associating a valve combination that controls fluid flow in the region, calculating a contribution of the valve combination to a change in curvature of the high-curvature region, integrating the contribution and the corresponding relationship of the valve combination, and generating an influence factor of a key valve combination.

[0098] Specifically, when extracting the path dependence index according to the position and intensity data of the high-curvature region, first, the position data and intensity data of the high-curvature region are obtained from the high-curvature region determination unit, and then all possible fluid flow paths from the pipeline inlet to the high-curvature region are traced based on the pipeline network topology map, and the number n of branch pipelines and the total length L of the path on each path are counted. At the same time, the average intensity value of the high-curvature region is calculated, and the calculation method is average intensity value = sum of high-curvature sampling intensity data / total number of sampling points in the high-curvature region, and then the formula is used to calculate the path dependence index of each path , and finally the maximum I value in all paths is taken as the path dependence index corresponding to the high-curvature region.

[0099] When calculating the intensity decay rate corresponding to the path dependence index, the fluid flow path obtained by tracing is divided into several segments according to the branch pipeline nodes, each segment corresponds to a continuous branch pipeline, the length of each segment is recorded, and the curvature intensity value at the starting end and the curvature intensity value at the end of each segment are obtained through sensors or simulation data, and then the formula: local decay rate = (curvature intensity value at the starting end - curvature intensity value at the end) / (curvature intensity value at the starting end x length of the corresponding segment) is used to calculate the local decay rate of each segment, and the arithmetic mean of the local decay rates of all segments is taken to obtain the intensity decay rate corresponding to the path dependence index.

[0100] determining whether the intensity decay rate exceeds a preset decay rate threshold (the preset decay rate threshold is set according to the type of the pipeline system, for example, the preset decay rate threshold in an industrial chemical pipeline system is 0.06 ), and if the calculated intensity decay rate is greater than the preset decay rate threshold, adjusting simulation parameters, wherein the simulation parameters include the time step (adjusting from 0.1s to 0.05s) and the grid division density (reducing the grid cell size from 0.4m to 0.2m) in fluid flow simulation, inputting the adjusted simulation parameters into the preliminary curvature distribution map generation unit, and re-executing the multi-valve opening combination simulation process to generate a regenerated curvature distribution map containing curvature peak coordinates and intensity gradient distribution.

[0101] After obtaining the position data and intensity data of the high-curvature area from the regenerated curvature profile, the associated relationship of the valve and branch pipe in the pipe network topology is combined to locate the valve combination (such as the combination of valve 1 and valve 3, the combination of valve 2 and valve 4, etc.) that can control the fluid flow in the high-curvature area, and then the contribution degree of each valve combination to the curvature change of the high-curvature area is calculated: first, keep the state of other valves unchanged, close all valves in the valve combination, and re-simulate the intensity value of the high-curvature area, and then calculate the contribution degree C through the formula The contribution degree C is calculated, wherein is the intensity data of the high-curvature area obtained from the regenerated curvature profile, is the intensity value of the high-curvature area re-simulated, and the contribution degree C is associated with the corresponding valve combination, and the contribution degree threshold value (such as =0.35) is set, and the valve combination with C> is screened as the key valve combination, and the identification information (such as the valve number combination) of the key valve combination and the corresponding contribution degree C are integrated to form the influence factor of the key valve combination.

[0102] In this process, the position and intensity data of the high-curvature area directly correspond to the calculation input of the path-dependent index, the path-dependent index corresponds to the calculation basis of the intensity decay rate, the comparison result of the intensity decay rate and the preset threshold value determines whether the simulation parameters are adjusted, the adjusted simulation parameters correspond to the regenerated curvature profile, the regenerated curvature profile data correspond to the positioning and contribution degree calculation of the key valve combination, and the association relationship between the contribution degree and the key valve combination finally constitutes the influence factor, forming a complete data corresponding chain. By extracting the path-dependent index and the intensity decay rate, the quantitative analysis of the fluid flow path characteristics is realized, and the problem of insufficient control instruction pertinence caused by the lack of path-dependent analysis is solved; by calculating the contribution degree of the valve combination and screening the key valve combination, the association relationship between the key valve combination and the high-curvature area is clear, which provides accurate data support for subsequent optimization of the valve opening sequence and reduces the local turbulence and energy loss caused by improper valve control.

[0103] Step S104, optimizing the preliminary curvature profile according to the influence factor to generate an optimized curvature profile, and obtaining a valve opening sequence of minimum overall curvature based on the optimized curvature profile.

[0104] In a specific embodiment, the step S104 can specifically include the following steps:

[0105] adjusting the action weight of each valve on the fluid flow path based on the influence factor to generate the optimized curvature profile;

[0106] Based on the optimized curvature distribution map, the overall curvature value of multiple valve opening sequences is calculated.

[0107] By comparing the overall curvature values ​​of all valve opening sequences, the minimum overall curvature is selected, the valve opening sequence corresponding to the minimum overall curvature is determined, and it is used as the initial target valve opening sequence.

[0108] Determine whether the preliminary target valve opening sequence meets the preset sequence constraints. If not, adjust the valve opening order, recalculate the overall curvature value of each valve opening sequence after adjustment, and generate a valve opening sequence that meets the preset sequence constraints and has the smallest overall curvature value.

[0109] Specifically, when adjusting the influence weights of each valve on the fluid flow path based on the influence factors to generate the optimized curvature distribution map, the influence factors of key valve combinations are first obtained from the influence factor extraction unit. These influence factors include the key valve combination identifier and the corresponding contribution C. The contribution C is then converted into the influence weights of each valve. ω The conversion method is (in, For a single valve i The weight of the effect, For a single valve i Contribution component in key combinations (The sum of the contributions of all valves involved in the key combination). Then, the adjusted weights are... ω The input curvature distribution map optimization model, combined with the curvature peak coordinates and intensity gradient distribution from the initial curvature distribution map, corrects the curvature values ​​at each location in the original distribution map. The correction formula is as follows: ,in The optimized curvature value, The initial curvature values ​​are shown in the preliminary curvature distribution map. The average value of the valve action weight in the corresponding region is used to generate an optimized curvature distribution map that includes the optimized curvature peak coordinates and intensity gradient distribution.

[0110] When calculating the overall curvature value of multiple valve opening sequences based on the optimized curvature distribution map, first enumerate all possible valve opening sequences in the pipeline system (e.g., in a system with N valves, the number of opening sequences is N!; in practical applications, obviously unreasonable sequences can be filtered out). For each opening sequence, extract the curvature values ​​of all branch pipes corresponding to that sequence from the optimized curvature distribution map. ,in k The branch pipe number is combined with the length of each branch pipe. Through formula Calculate the overall curvature value of the opening sequence. ,in, is the sum of the curvature value and length product of each branch pipe ， is the sum of the length of all branch pipes, ensuring that the overall curvature value reflects the comprehensive influence of the sequence on the fluid flow of the entire pipe system.

[0111] Compare the overall curvature values of all valve opening sequences, select the smallest overall curvature value, and locate the valve opening sequence corresponding to the overall curvature value, which is the preliminary target valve opening sequence. Determine whether the preliminary target valve opening sequence meets the preset sequence constraint conditions, which include adjacent valve opening interval constraints (such as in industrial heating pipe systems, the interval between adjacent valve opening times must be greater than 2s) and valve opening sequence constraints (such as valves located in the main pipe must be opened before branch pipe valves); verify by calling the opening timestamp data and valve position association data of the sequence, if the constraints are not met, adjust the valve opening sequence, for example, adjust the opening time sequence of the valve that violates the interval constraint backward, or adjust the opening order of the main and branch valves; recalculate the overall curvature value of the sequence after adjustment, and repeat the adjustment and calculation process until a valve opening sequence that meets the preset sequence constraint conditions and has the smallest overall curvature value is generated.

[0112] In this process, the adjustment basis of the influence factor corresponding to the valve action weight, the generation basis of the valve action weight corresponding to the optimized curvature distribution map, the calculation source of the overall curvature value corresponding to the multiple valve opening sequences, the determination basis of the comparison result of the overall curvature value corresponding to the preliminary target valve opening sequence, the adjustment standard of the preset sequence constraint condition corresponding to the valve opening sequence, and the final valve opening sequence that meets the conditions corresponding to the adjusted sequence and the overall curvature value calculation result. By adjusting the valve action weight through the influence factor and optimizing the curvature distribution map, the uneven curvature distribution can be reduced; by calculating the overall curvature value and screening the sequence combined with the constraint condition, a valve opening sequence that takes into account low curvature and stability can be generated, reducing energy loss caused by local turbulence and improving system operation stability.

[0113] Step S105, determine whether the valve opening sequence has a dynamic characteristic conflict.

[0114] In a specific embodiment, the step S105 can specifically include the following steps:

[0115] Obtain the opening interval data in the valve opening sequence;

[0116] Determine whether there is data in the opening interval data that is less than the preset opening interval constraint, if yes, determine that the valve opening sequence has a dynamic characteristic conflict, if no, determine that the valve opening sequence does not have a dynamic characteristic conflict;

[0117] If a dynamic characteristic conflict is determined, a conflict detection rule is extracted, the cause of the conflict is analyzed based on the conflict detection rule, and a conflict detection report is generated. The conflict detection rule includes the correspondence between the opening interval and the peak value of the fluid curvature, and the association rule between the valve combination and the pressure fluctuation. The conflict detection report includes the conflict point and the corresponding valve combination.

[0118] Specifically, when acquiring the opening interval data in the valve opening sequence, the determined valve opening sequence is first retrieved from the valve opening sequence acquisition unit. This sequence contains the opening timestamp data of each valve. ,in i This refers to the valve serial number; the difference in opening time between two adjacent valves is calculated. This yields the time intervals between the opening of all adjacent valves. These time intervals are then integrated to form opening interval data. For example, in an industrial chemical pipeline system, the valve opening sequence might be valve 1, valve 3, and valve 2, with corresponding opening timestamps of [missing information]. , and Then the interval data is enabled. When determining whether there is any data in the opening interval data that is less than the preset opening interval constraint, first set the preset opening interval constraint according to the fluid characteristics and operating requirements of the pipeline system. For example, in a pipeline system transporting viscous fluid, the preset opening interval constraint is set to 2 seconds. Then, each data in the opening interval data... Compare each one with the preset opening interval constraint; if any one exists... If the preset opening interval constraint is not met, it is determined that there is a dynamic characteristic conflict in the valve opening sequence; if all If all values ​​are greater than or equal to the preset opening interval constraint, then it is determined that there is no dynamic characteristic conflict in the valve opening sequence. For example, in the opening interval data mentioned above... If the preset constraint is less than 2s, it can be determined that the sequence has dynamic characteristic conflicts.

[0119] If a dynamic characteristic conflict is determined, conflict detection rules are extracted from the system's pre-stored rule base. These conflict detection rules include the correspondence between opening intervals and fluid curvature peak values, and the association rules between valve combinations and pressure fluctuations. The correspondence between opening intervals and fluid curvature peak values ​​is constructed using historical simulation data, such as recording different opening intervals. Peak curvature in the lower high curvature region ,form Mapping table; the association rule between valve combinations and pressure fluctuations is obtained by statistically analyzing the pressure fluctuation values ​​in the pipeline when different valve combinations are opened. (Unit: MPa), setting The valve combination corresponding to 0.1 MPa is a pressure fluctuation sensitive combination. According to the conflict detection rule, the conflict reason is analyzed first according to the corresponding relationship between the opening interval and the fluid curvature peak value, and the conflict sequence is queried corresponding to , if exceeds the preset curvature peak threshold value (such as > 5 ), it is determined that one of the conflict reasons is that the small opening interval leads to the high curvature region curvature peak value exceeding the standard; and in combination with the valve combination and the pressure fluctuation correlation rule, it is judged whether the valve combination in the conflict sequence belongs to the pressure fluctuation sensitive combination, if it belongs to and the historical data shows that the pressure fluctuation value in the pipeline when the combination is opened the average value is 0.15 MPa, the conflict reason is supplemented that the sensitive valve combination opening causes the pressure fluctuation risk. When generating the conflict detection report, the time point corresponding to the conflict opening interval data (such as ) of the conflict is taken as the conflict point, the specific time (such as 1.5s, 3s) of the conflict point is recorded, and the valve combination (such as valve 1 and valve 3, valve 3 and valve 2) involved in the conflict point is listed, and the conflict reason (curvature peak value exceeding the standard, pressure fluctuation risk) analyzed is included in the report, forming a conflict detection report containing the conflict point and the corresponding valve combination.

[0120] In this process, the timestamp data of the valve opening sequence corresponds to the calculation source of the opening interval data, the comparison result of the opening interval data and the preset opening interval constraint corresponds to the determination basis of the dynamic characteristic conflict, the conflict detection rule corresponds to the analysis standard of the conflict reason, the conflict point, the valve combination and the conflict reason correspond to the core content of the conflict detection report, forming a complete data corresponding chain. By extracting the opening interval data and comparing it with the preset constraint, the conflict sequence can be accurately positioned; in combination with the conflict detection rule to analyze the reason and generate the report, the specific position and influencing factors of the conflict can be clearly determined, solving the problem that the traditional method cannot accurately identify the conflict root cause, providing targeted basis for subsequent sequence rearrangement, and reducing the system instability risk caused by dynamic conflict.

[0121] Step S106, if the valve opening sequence has a dynamic characteristic conflict, a particle swarm optimization algorithm is used to rearrange the valve opening sequence to generate a conflict-free control sequence.

[0122] In a specific embodiment, the step S106 can specifically include the following steps:

[0123] Initialize particle swarm parameters, map each particle to a set of valve opening sequences to be optimized, and set the number of iterations, learning factor, and inertia weight parameter;

[0124] The fitness function is constructed by incorporating the arrangement priority, and the goal of the fitness function is to minimize the dynamic characteristic conflict;

[0125] The valve opening sequence corresponding to the optimal particle in the population is subjected to mutation operation, and a sequence after mutation without dynamic characteristic conflict is selected as a first candidate sequence of the sequence mutation path;

[0126] The dynamic adjustment vector of the first candidate sequence is calculated, the weighted sum of each component of the dynamic adjustment vector is generated to generate a sequence stability index, and the first candidate sequence with the stability index lower than a preset stability threshold is defined as a second candidate sequence;

[0127] From the second candidate sequence, a sequence with a difference between the overall curvature and the determined minimum overall curvature less than a preset difference threshold is selected as the conflict-free control sequence.

[0128] Specifically, when initializing the particle swarm parameters, the particle swarm size (such as 50 particles in an industrial pipeline system) is first set, the position vector of each particle is mapped to a set of valve opening sequences to be optimized, for example, in a system including 6 valves, the particle position vector (2, 4, 1, 5, 3, 6) corresponds to the valve opening sequence of valve 2, valve 4, valve 1, valve 5, valve 3 and valve 6; At the same time, the iteration number (such as 100 times), the learning factor and (both set to 2.0), and the inertia weight W (the initial value is set to 0.9 and linearly decreases to 0.4 with the iteration number) are set. These parameters are verified by historical control data of the pipeline system to ensure the search efficiency and convergence of the particle swarm.

[0129] When constructing the fitness function by incorporating the arrangement priority, the arrangement priority is set according to the valve correlation weight, and the priority of the main pipeline valve is higher than that of the branch pipeline valve (for example, the priority coefficient of the main valve is set to 1.2, and the priority coefficient of the branch valve is set to 0.8). The goal of the fitness function is to minimize the dynamic characteristic conflict, and the function expression is wherein, is the number of dynamic characteristic conflicts in the sequence (one conflict is recorded if the opening interval is less than the preset opening interval constraint), H is the matching degree of the valve opening sequence in the sequence and the arrangement priority (the matching degree H is calculated by counting the proportion of the number of high-priority valves opened first), and are weight coefficients (both set to 0.5), and the smaller the fitness function value, the fewer the sequence conflicts and the higher the priority matching degree.

[0130] When the valve opening sequence corresponding to the optimal particle in the group is mutated, first calculate the fitness value of each particle by the fitness function, select the particle with the smallest fitness value as the current optimal particle in the group, and extract the valve opening sequence corresponding to the particle; the mutation operation adopts the adjacent position exchange strategy, randomly selects the positions of two adjacent valves in the sequence to exchange (such as sequence (2, 4, 1, 5, 3, 6) is mutated to (2, 1, 4, 5, 3, 6)), checks whether the sequence has a dynamic characteristic conflict after each mutation (i.e. judges whether all adjacent opening intervals meet the preset constraint), and selects the sequence without conflict after mutation as the first candidate sequence containing the sequence mutation path. For example, in an industrial heating pipeline system, the preset opening interval constraint is 2s, and the adjacent opening intervals of the sequence after mutation

[0131] When calculating the dynamic adjustment vector of the first candidate sequence, the dynamic adjustment vector component is the difference value of the fluid flow change caused by the opening of adjacent valves in the sequence is the real-time flow in the pipeline after the opening of the ith valve, unit: cubic meters / hour), and the vector expression is ; the weighted sum of the components of the dynamic adjustment vector generates a sequence stability index S, and the weight is the valve contribution weight , and the calculation formula is . Set a preset stability threshold (such as 5 cubic meters / hour in an industrial chemical pipeline system), select the first candidate sequence with S . Set a preset difference threshold (such as 10% of the determined minimum overall curvature ), select the sequence with as the conflict-free control sequence, for example = 8 , the preset difference threshold = 0.8 , the second candidate sequence has = 8.5 , which meets the condition and is included in the final conflict-free control sequence.

[0132] ​​The particle swarm parameters correspond to the mapping basis of the particle position and the valve opening sequence, the arrangement priority and the conflict index correspond to the construction basis of the fitness function, the group optimal particle corresponds to the initial sequence of the mutation operation, the dynamic adjustment vector component corresponds to the calculation source of the stability index, and the overall curvature difference corresponds to the selection standard of the conflict-free control sequence, forming a complete data chain. Through the particle swarm optimization algorithm and the mutation operation, a conflict-free sequence can be efficiently generated; combined with the stability index and the overall curvature selection, the problem of "dynamic conflict leading to system vibration or pressure fluctuation" is solved, ensuring that the sequence takes into account stability and low energy loss, and improving the reliability of the pipeline system operation.

[0133] In step S107, the real-time fluid flow change of the conflict-free control sequence is simulated, the final valve adjustment scheme is determined, the energy loss prediction value is generated, and it is judged whether the energy loss prediction value is lower than the preset loss threshold. If yes, the multi-valve cooperative control instruction is output.

[0134] In a specific embodiment, in step S107, determining the final valve adjustment scheme can specifically include the following steps:

[0135] Simulating the real-time fluid flow change in the pipeline when each valve is opened under the conflict-free control sequence;

[0136] According to the simulation results, the opening duration and opening amplitude of the valves in the conflict-free control sequence are adjusted to form a preliminary valve adjustment scheme;

[0137] It is verified whether the preliminary valve adjustment scheme meets the turbulent flow control requirements of the high-curvature region. If not, the valve opening duration and opening amplitude are continuously adjusted until the final valve adjustment scheme that meets the turbulent flow control requirements is generated.

[0138] Specifically, when simulating the real-time fluid flow change in the pipeline when each valve is opened under the conflict-free control sequence, the conflict-free control sequence is first obtained from the sequence rearrangement unit, which includes the valve opening sequence and the corresponding opening timestamp. The sequence is imported into the fluid dynamics simulation model, and the model input parameters also include basic parameters such as the pipe diameter (e.g. 0.5m in industrial pipeline systems), the fluid density (e.g. 1000 kg / m 3 ), the fluid viscosity (e.g. 0.001 Pa·s for water), etc. The fluid flow in the pipeline is numerically simulated by the finite volume method. The time step is set to 0.05s, and the fluid flow value of each section of the pipeline is calculated in each time step (unit: cubic meters / hour), generating a real-time fluid flow change curve, with the horizontal axis representing time (unit: s) and the vertical axis representing flow value, and recording the flow mutation value corresponding to each valve opening time (That is, the difference between the flow rate after opening and the flow rate before opening), for example, when valve 1 is opened, the flow rate increases from 50 cubic meters per hour to 70 cubic meters per hour. =20 cubic meters / hour.

[0139] When adjusting the valve opening duration and opening amplitude in the conflict-free control sequence based on simulation results to form a preliminary valve control scheme, first analyze the real-time fluid flow rate change curve. If the flow rate changes abruptly after a certain valve is opened... If the flow fluctuation exceeds the preset threshold (e.g., 15 cubic meters per hour in an industrial heating pipeline system), the valve opening range is adjusted. The opening range is expressed as a percentage of valve opening (e.g., from 100% fully open to 80% opening). The simulation is then repeated and monitored. until <15 cubic meters / hour. If the flow rate fluctuates continuously during the opening of a certain valve (e.g., the flow rate change exceeds 8 cubic meters / hour within 5 seconds), the opening duration is adjusted, such as extending the original opening duration from 3 seconds to 5 seconds, so that the flow rate rises gradually. The adjusted opening duration of each valve (e.g., valve 1 opening duration 5 seconds, valve 2 opening duration 4 seconds) and opening range (e.g., valve 1 opening degree 80%, valve 2 opening degree 90%) are recorded. Combined with the opening sequence of the original conflict-free control sequence, a preliminary valve regulation scheme is formed.

[0140] When verifying whether the preliminary valve regulation scheme meets the turbulence control requirements of the high curvature region, first obtain the position coordinates of the high curvature region and the corresponding turbulence intensity threshold from the high curvature region determination unit (e.g., set as [value] in an industrial chemical pipeline system). The preliminary valve regulation scheme is input into the fluid dynamics simulation model to re-simulate the fluid flow state inside the pipe and calculate the turbulence intensity in the high curvature region. The calculation method is as follows ,in These represent the root mean square velocities of the fluid in the x, y, and z directions within the high curvature region. If the simulated turbulence intensity... If the turbulence intensity is ≤ the turbulence intensity threshold, then the preliminary valve adjustment scheme is deemed to meet the turbulence control requirements; if .... If the turbulence intensity threshold is exceeded, the valve opening duration and opening amplitude are adjusted further. For example, for a valve combination controlling fluid flow in a high curvature region, the opening amplitude is increased from 90% to 75%, and the opening duration is extended from 4s to 6s. The turbulence intensity is then re-simulated and recalculated. Repeat the adjustment and verification process until the simulated turbulence intensity is obtained. <Turbulence intensity threshold, generate the final valve regulation scheme that meets the turbulence control requirements.

[0141] In the process, the conflict-free control sequence corresponds to the initial sequence input of the fluid flow simulation, the real-time fluid flow curve corresponds to the adjustment basis of the valve opening duration and opening amplitude, the adjusted opening parameters correspond to the generation basis of the preliminary valve adjustment scheme, the turbulence intensity calculation result of the high curvature area corresponds to the verification standard of the scheme, the verification result corresponds to the determination basis of the final valve adjustment scheme, and the complete data corresponds to the chain. By simulating the real-time flow change and adjusting the valve opening parameters, the turbulence intensity caused by the flow mutation can be reduced; by verifying the scheme in combination with the turbulence control requirements of the high curvature area, the problem that the traditional method cannot control the local turbulence specifically is solved, the energy loss caused by turbulence is reduced, the problem of "vibration or pressure fluctuation during system operation" is alleviated, and the stability of the pipeline system is improved.

[0142] In a specific embodiment, the step S107 of outputting the multi-valve cooperative control instruction can specifically include the following steps:

[0143] Extracting the pipeline geometric parameters corresponding to the final valve adjustment scheme to generate geometric constraints;

[0144] Based on the geometric constraints, calculating the energy loss prediction value of the pipeline system under the final valve adjustment scheme;

[0145] Judging whether the energy loss prediction value is lower than a preset loss threshold, if yes, generating a preliminary multi-valve cooperative control instruction based on the final valve adjustment scheme;

[0146] Verifying whether the preliminary multi-valve cooperative control instruction can optimize the local turbulence of the micro branch, ensuring that the instruction execution effect meets the stability requirements of the pipeline system, and finally outputting the multi-valve cooperative control instruction.

[0147] Specifically, when extracting the pipeline geometric parameters corresponding to the final valve adjustment scheme to generate geometric constraints, the position identifiers of the valves are first obtained from the final valve adjustment scheme, the three-dimensional model of the pipeline system is associated, and the geometric parameters of the branch pipelines where the valves are located are extracted, including the pipeline inner diameter d (the main pipeline d is set to 0.5 m and the micro branch pipeline d is set to 0.2 m in the industrial chemical pipeline scene), the pipeline length L (the main pipeline single segment L is set to 100 m and the micro branch pipeline L is set to 20 m), the pipeline bending angle (such as a 90° elbow or a 45° elbow), and the pipeline roughness ε (the industrial steel pipe ε is set to 0.045 mm). These parameters are sorted according to the pipeline segmentation, and the geometric boundaries of each segment of the pipeline are clearly defined, for example, "main pipeline segment 1: d = 0.5 m, L = 100 m, straight pipe; micro branch pipeline segment 1: d = 0.2 m, L = 20 m, =90°”, which serves as a geometric constraint to limit the spatial range and physical boundary of subsequent energy loss calculations.

[0148] When calculating the predicted energy loss of the pipeline system under the final valve control scheme based on geometric constraints, the energy loss includes friction loss and local energy loss. Friction loss is calculated using the Darcy-Weisbach formula, which is as follows: ,in The friction coefficient is determined using the Colbrook formula. Iterative solution, Let Reynolds number be 1. ( For fluid density, industrial water Set as ; v The average flow velocity of the fluid in the pipe is calculated by the ratio of the simulated flow rate Q to the pipe cross-sectional area S corresponding to the final valve regulation scheme. μ The fluid dynamic viscosity is 20°C for water. μ (Set as 0.001 Pa·s); g is the acceleration due to gravity, taken as 9.8 m / s². 2 Local energy loss according to calculate, This is the local resistance coefficient, determined based on the valve opening range, such as when the valve is fully open. =0.17, at 50% opening =0.8, while also taking into account the local resistance at the pipe bend (90° bend). =0.75, 45° bend =0.3). Total energy loss prediction. (Unit: meter), then convert it into a power loss form. (Unit: Watt), as the final predicted energy loss value.

[0149] When determining whether the predicted energy loss value is lower than the preset loss threshold, the preset loss value is set according to the design energy consumption standard of the pipeline system. For example, in an industrial heating pipeline system, the preset loss value is set to 10000W. If the calculated value is lower than the preset loss threshold, the preset loss value is lower than the preset loss threshold. For valves with a power rating of ≤10000W, preliminary multi-valve coordinated control instructions are generated based on the final valve regulation scheme. The instructions include the opening sequence, opening time, opening duration, and opening amplitude of each valve. For example, "Valve 1 (main pipeline): opening time 0s, opening duration 5s, opening amplitude 100%; Valve 2 (micro branch): opening time 5s, opening duration 3s, opening amplitude 60%; Valve 3 (micro branch): opening time 8s, opening duration 4s, opening amplitude 55%", ensuring that the instruction parameters completely correspond to the final valve regulation scheme.

[0150] To verify whether the preliminary multi-valve coordinated control instruction can optimize the local turbulence of the micro branch, the preliminary instruction is input into a fluid dynamics simulation model. The model needs to load the micro branch pipeline parameters in the geometric constraint condition, simulate the fluid flow state in the micro branch during the execution of the instruction, and calculate the local flow intensity , in the form of , where are the root mean square of the fluctuating velocity of the fluid in the x, y, and z directions in the high-curvature region. Set the turbulence control standard for the micro branch, such as ≤ If the simulation result meets the standard, it is determined that the instruction can optimize the local turbulence and the execution effect meets the stability requirements of the pipeline system; if > The opening amplitude or opening duration of the micro branch corresponding valve in the final valve adjustment scheme needs to be adjusted, such as reducing the opening amplitude of valve 2 from 60% to 50%, recalculating the energy loss prediction value and verifying the turbulence optimization effect, until the requirements are met, and output the multi-valve coordinated control instruction.

[0151] In this process, the final valve adjustment scheme corresponds to the extraction of the pipeline geometric parameters, the geometric constraint condition corresponds to the boundary basis for energy loss calculation, the comparison result of the energy loss prediction value and the preset loss threshold corresponds to the generation of the preliminary instruction or not, the simulation result of the preliminary instruction corresponds to the verification standard of the turbulence optimization effect, and the verification result corresponds to the output basis of the final instruction, forming a complete data chain. By accurately calculating the friction loss and local loss and comparing with the preset value, it can be ensured that the energy consumption of the scheme meets the requirements; the turbulence intensity verification optimizes the instruction effect for the problem of "micro branch local turbulence causing system instability", solves the defect of traditional methods ignoring the flow state of micro branch, and at the same time alleviates the problem of "low fluid flow efficiency", and improves the overall operation performance of the pipeline system.

[0152] Please refer to Figure 2 , the intelligent remote control system 400 for special valves in the embodiments of the present application will be described below, including:

[0153] The curvature distribution generation module 201 is configured to obtain valve position distribution data and fluid initial flow parameter, simulate the influence of multi-valve opening combination on fluid flow path, and generate a preliminary curvature distribution map;

[0154] The high-curvature region determination module 202 is configured to calculate the flow curvature value of each branch pipeline according to the preliminary curvature distribution map, and determine the high-curvature region;

[0155] The influence factor extraction module 203 is configured to extract the influence factor of the key valve combination from the high-curvature region;

[0156] The valve opening sequence acquisition module 204 is configured to generate a curvature distribution diagram according to the influence factor, and acquire a valve opening sequence with minimum overall curvature.

[0157] The dynamic characteristic conflict judgment module 205 is configured to judge whether the valve opening sequence has a dynamic characteristic conflict.

[0158] The sequence rearrangement module 206 is configured to rearrange the valve opening sequence by using a particle swarm optimization algorithm if there is a conflict, and generate a conflict-free control sequence.

[0159] The valve adjustment scheme generation module 207 is configured to simulate real-time fluid flow changes of the conflict-free control sequence, determine a final valve adjustment scheme, generate an energy loss prediction value, and output a multi-valve cooperative control instruction if the energy loss prediction value is lower than a threshold value.

[0160] Through the cooperation of the above components, the system builds a closed-loop intelligent control system of “data acquisition-simulation analysis-conflict detection-sequence optimization-scheme verification-instruction output”, realizes the cooperative control of multiple valves in a complex pipeline system, and improves the fluid flow efficiency and stability, and reduces the operating energy consumption. Among them: the curvature distribution generation module 201 acquires valve position distribution data and fluid initial flow parameters, generates a preliminary curvature distribution diagram through simulation of multiple valve opening combinations, provides basic data for subsequent analysis, and solves the problem that the traditional method cannot quantify the influence of valve opening combinations on the flow path; the high curvature area determination module 202 calculates the branch pipeline flow curvature value and determines the high curvature area based on the preliminary curvature distribution diagram, clearly defines the key disturbance area in fluid flow, and solves the problem of local turbulence caused by high curvature; the influence factor extraction module 203 extracts the influence factor of the key valve combination from the high curvature area, accurately identifies the valve combination that plays a leading role in the curvature distribution, and makes up for the defect that the traditional method cannot clearly define the contribution weight of the key valve; the valve opening sequence acquisition module 204 optimizes the curvature distribution diagram according to the influence factor and acquires a valve opening sequence with minimum overall curvature, reduces the curvature distribution unevenness from the source, and reduces energy loss; the dynamic characteristic conflict judgment module 205 judges whether the valve opening sequence has a dynamic characteristic conflict, and identifies potential hazards that may cause system vibration or pressure fluctuation in advance; the sequence rearrangement module 206 rearranges the sequence with conflict by using a particle swarm optimization algorithm, generates a conflict-free control sequence, and solves the problem of system instability caused by dynamic characteristic conflict; the valve adjustment scheme generation module 207 simulates real-time fluid flow changes of the conflict-free control sequence, determines a final valve adjustment scheme, generates an energy loss prediction value, and verifies the local turbulence optimization effect of the micro branch, finally outputs a cooperative control instruction, forms a complete closed loop, ensures that the control scheme takes into account low energy consumption, high stability and high efficient flow, and provides an efficient cooperative control solution for a complex pipeline system.

[0161] The above examples are only used to illustrate the technical solutions of the present application, but not to limit the same; although the present application has been described in detail with reference to the foregoing examples, it should be understood by those of ordinary skill in the art that: it can still modify the technical solutions recorded in the foregoing examples, or make equivalent replacement for part of the technical features; and these modifications or replacements do not make the essence of the corresponding technical solutions deviate from the spirit and scope of the technical solutions of the embodiments of the present application.

Claims

1. A remote control method for intelligent special valves, characterized in that, The method includes: Step S101: Obtain valve position distribution data and initial fluid flow parameters, and simulate the impact of multiple valve opening combinations on the fluid flow path, and generate a preliminary curvature distribution map according to the impact. Step S102: Based on the preliminary curvature distribution diagram, calculate the flow curvature value in each branch pipe and determine the high curvature region; Step S103: Extract the influence factors of key valve combinations from the high curvature region; Step S104: Optimize the preliminary curvature distribution map according to the influencing factors to generate an optimized curvature distribution map, and obtain the valve opening sequence with minimum overall curvature based on the optimized curvature distribution map; Step S105: Determine whether there is a dynamic characteristic conflict in the valve opening sequence; Step S106: If there is a dynamic characteristic conflict in the valve opening sequence, the particle swarm optimization algorithm is used to rearrange the valve opening sequence to generate a conflict-free control sequence. Step S107: Simulate the real-time fluid flow change of the conflict-free control sequence, determine the final valve adjustment scheme, generate an energy loss prediction value, and determine whether the energy loss prediction value is lower than the preset loss threshold. If so, output a multi-valve collaborative control command.

2. The remote control method for intelligent special valves according to claim 1, characterized in that, Step S101 includes: Acquire the location distribution data of multiple valves and the initial fluid flow parameters; Based on the location distribution data and the initial fluid flow parameters, the influence of different valve opening combinations on the fluid flow path is simulated. Based on the influence of different valve opening combinations on the fluid flow path, a preliminary curvature distribution map is generated, which includes the curvature peak coordinates and intensity gradient distribution. Key nodes in the flow path are determined based on the curvature peak coordinates. Based on the intensity gradient distribution, the contribution weights of multiple valves to fluid flow are extracted.

3. The remote control method for intelligent special valves according to claim 1, characterized in that, Step S102 includes: The preliminary curvature distribution map was processed using finite element analysis; Based on the preliminary curvature distribution diagram after finite element analysis, the flow curvature values ​​in multiple branch pipes are calculated. The location and intensity of the high curvature region are determined based on multiple flow curvature values; Valve-related weights are extracted from the intensity gradient distribution contained in the preliminary curvature distribution map.

4. The remote control method for intelligent special valves according to claim 1, characterized in that, Step S103 includes: Based on the location and intensity data of the high curvature region, path dependency indicators are extracted; Calculate the local attenuation rate of each segment of the fluid flow path corresponding to the path dependence index, and average the local attenuation rate of each segment to generate the intensity attenuation rate corresponding to the path dependence index. Determine whether the intensity attenuation rate exceeds a preset attenuation rate threshold. If so, adjust the simulation parameters; and regenerate the curvature distribution map based on the adjusted simulation parameters. The location and intensity data of high curvature regions are obtained from the regenerated curvature distribution map. The valve combination that controls the fluid flow in this region is associated with it. The contribution of the valve combination to the curvature change in the high curvature region is calculated. The correspondence between the contribution and the valve combination is integrated to generate the influence factor of the key valve combination.

5. The remote control method for intelligent special valves according to claim 1, characterized in that, Step S104 includes: Based on the aforementioned influencing factors, the weight of each valve's effect on the fluid flow path is adjusted to generate the optimized curvature distribution map. Based on the optimized curvature distribution map, the overall curvature value of multiple valve opening sequences is calculated. By comparing the overall curvature values ​​of all valve opening sequences, the minimum overall curvature is selected, the valve opening sequence corresponding to the minimum overall curvature is determined, and it is used as the initial target valve opening sequence. Determine whether the preliminary target valve opening sequence meets the preset sequence constraints. If not, adjust the valve opening order, recalculate the overall curvature value of each valve opening sequence after adjustment, and generate a valve opening sequence that meets the preset sequence constraints and has the smallest overall curvature value.

6. The remote control method for intelligent special valves according to claim 1, characterized in that, Step S105 includes: Obtain the opening interval data in the valve opening sequence; Determine whether there is any data in the opening interval data that is less than the preset opening interval constraint. If so, determine that there is a dynamic characteristic conflict in the valve opening sequence. If not, determine that there is no dynamic characteristic conflict in the valve opening sequence. If a dynamic characteristic conflict is determined, a conflict detection rule is extracted, the cause of the conflict is analyzed based on the conflict detection rule, and a conflict detection report is generated. The conflict detection rule includes the correspondence between the opening interval and the peak value of the fluid curvature, and the association rule between the valve combination and the pressure fluctuation. The conflict detection report includes the conflict point and the corresponding valve combination.

7. The remote control method for intelligent special valves according to claim 1, characterized in that, Step S106 includes: Initialize the particle swarm parameters, map each particle to a set of valve opening sequences to be optimized, and set the number of iterations, learning factor, and inertia weight parameters. A fitness function is constructed by incorporating permutation priority, and the objective of the fitness function is to minimize dynamic characteristic conflicts. Mutation operation is performed on the valve opening sequence corresponding to the best particle in the swarm, and the sequence that does not show dynamic characteristic conflict after mutation is selected as the first candidate sequence containing the sequence mutation path. Calculate the dynamic adjustment vector of the first candidate sequence, generate a sequence stability index by weighted summation of each component of the dynamic adjustment vector, and select the first candidate sequence whose stability index is lower than a preset stability threshold, which is defined as the second candidate sequence. From the second candidate sequences, the sequence in which the difference between the overall curvature and the determined minimum overall curvature is less than a preset difference threshold is selected as the conflict-free control sequence.

8. The remote control method for intelligent special valves according to claim 1, characterized in that, In step S107, determining the final valve adjustment scheme includes: Simulate the real-time fluid flow rate changes in the pipeline when each valve is opened under the conflict-free control sequence; Based on the simulation results, adjust the valve opening duration and opening amplitude in the conflict-free control sequence to form a preliminary valve regulation scheme; Verify whether the preliminary valve adjustment scheme meets the turbulence control requirements in the high curvature region. If not, continue to adjust the valve opening duration and opening amplitude until a final valve adjustment scheme that meets the turbulence control requirements is generated.

9. The remote control method for intelligent special valves according to claim 8, characterized in that, In step S107, the multi-valve coordinated control command is output, including: Extract the pipeline geometric parameters corresponding to the final valve adjustment scheme and generate geometric constraints; Based on the geometric constraints, the predicted energy loss of the pipeline system under the final valve regulation scheme is calculated; Determine whether the predicted energy loss value is lower than a preset loss threshold. If so, generate a preliminary multi-valve collaborative control command based on the final valve adjustment scheme. Verify whether the preliminary multi-valve coordinated control command can optimize the local turbulence of the micro-branch, ensure that the command execution effect meets the stability requirements of the pipeline system, and finally output the multi-valve coordinated control command.

10. A remote control system for intelligent special valves, characterized in that, For implementing the remote control method for intelligent special valves as described in any one of claims 1 to 9, the remote control system for intelligent special valves comprises: The curvature distribution generation module is used to acquire valve position distribution data and initial fluid flow parameters, simulate the impact of multiple valve opening combinations on the fluid flow path, and generate a preliminary curvature distribution map. The high curvature region determination module is used to calculate the flow curvature values ​​of each branch pipe based on the preliminary curvature distribution map, and to determine the high curvature region. The impact factor extraction module is used to extract the impact factors of key valve combinations from high curvature regions. The valve opening sequence acquisition module is used to optimize and generate a curvature distribution map based on influencing factors, and obtain the valve opening sequence with the minimum overall curvature. The dynamic characteristic conflict judgment module is used to determine whether there is a dynamic characteristic conflict in the valve opening sequence; If a conflict exists, the sequence rearrangement module uses a particle swarm optimization algorithm to rearrange the valve opening sequence and generate a conflict-free control sequence. The valve regulation scheme generation module is used to simulate real-time fluid flow changes in a conflict-free control sequence, determine the final valve regulation scheme, generate energy loss prediction values, and output multi-valve collaborative control commands if the values ​​are below a threshold.

Images