Intelligent matching method for oil and gas module pipeline materials

Abstract

The invention discloses an intelligent matching method for oil and gas module pipeline materials, and the method comprises the steps: building a stroke unit model of a pipeline, constructing a feature data set of a required pipeline, building a membership matrix of all required pipelines, and carrying out the comparison of the pipeline and an inventory pipeline through an intelligent neural network model, and calculating the geometric feature similarity and the pipeline material feature matching degree of the required pipeline and the inventory pipeline to obtain the matching condition of the required pipeline and the inventory pipeline, and performing pipeline delivery and pipeline purchase according to a comparison result. According to the method, the matching accuracy is improved through technologies such as three-dimensional modeling, and human errors are reduced; and resource utilization is optimized, material waste and overstock are avoided, and the working efficiency is improved.

Description

Technical Field

[0001] The invention relates to a material matching algorithm, and in particular to an intelligent matching method for oil and gas module pipeline materials. Background Art

[0002] The construction cycle of oil and gas processing industry projects is very tight and requires strict material management. At present, the matching of oil and gas module pipeline materials mainly relies on manual experience and traditional document records. Material selection is based on manual experience and lacks data analysis; arrival records are manually operated and then entered into the computer, which is inefficient and prone to errors; there is no intelligent control in the distribution link. The recorded information is simple, the system response is poor, and the information transmission is not smooth. Summary of the invention

[0003] The purpose of the present invention is to overcome the shortcomings of the existing technology and provide an intelligent matching method for oil and gas module pipeline materials. Through precise data modeling, intelligent feature analysis and efficient matching algorithm, accurate matching and optimized management of oil and gas module pipeline materials can be achieved to improve work efficiency.

[0004] In order to achieve the above object, the present invention provides the following technical solutions:

[0005] A method for intelligently matching oil and gas module pipeline materials of the present invention comprises the following steps:

[0006] Step 1: Create a 3D model of the pipeline of the oil and gas module to be constructed in the 3D modeling software;

[0007] Step 2: According to the pipeline 3D model of the oil and gas module to be constructed, the stroke method is used to establish the spatial connection relationship between the pipelines during construction, and the pipeline 3D model of the oil and gas module to be constructed is converted into a stroke unit model of the pipeline. The stroke unit model includes pipeline information, including: pipeline diameter, pipeline length, angle between pipelines, and spatial information between pipelines.

[0008] Step 3: Construct a feature data set of the required pipeline according to the geometric features of the pipeline;

[0009] Step 4: Establish a membership matrix D for all demand pipelines;

[0010] Step 5: Establish a pipeline management database for pipelines in the warehouse and set the minimum threshold Δx for the inventory of each pipeline; input the membership matrix D of the demand pipeline into the intelligent neural network model for comparing the demand pipeline and the inventory pipeline, calculate the geometric feature similarity and pipeline material feature matching degree of the demand pipeline and the inventory pipeline, obtain the matching situation of the required pipeline and the inventory pipeline, and output the matching result;

[0011] Step 6: According to the comparison result, if it is not available in the warehouse, directly remove the pipelines that are not in the warehouse pipeline database from the pipeline requirement list and mark them as out-of-stock pipelines for subsequent procurement. Then, sort the remaining pipelines in the material requirement list in ascending order of their numbers and proceed to the next step;

[0012] Step 7: Compare the required quantity of the first pipeline with the inventory quantity n of the pipelines. If the inventory pipeline quantity ≥ the required quantity, mark it as an out-of-warehouse pipeline on the pipeline requirement list and modify the current inventory quantity in the pipeline management database to the original inventory quantity minus the required inventory quantity; otherwise, issue all pipelines of this type from the pipeline management database and mark it as a pipeline with insufficient inventory on the pipeline requirement list;

[0013] Step 8: Determine the inventory status of other pipelines on the pipeline requirement list and make relevant marks in turn according to the method in Step 7 until all pipelines are matched;

[0014] Step 9: Compare the inventory quantity of each pipeline in the warehouse with the minimum threshold Δx in turn. If it is lower than the minimum threshold required for the warehouse inventory, send the list to the pipeline procurement department for pipeline procurement.

[0015] The beneficial effects of the present invention are as follows: By technologies such as 3D modeling, the matching accuracy is improved, and human errors are reduced; resource utilization is optimized, material waste and backlog are avoided, and dynamic management enhances flexibility and timeliness, ensuring the project progress. BRIEF DESCRIPTION OF THE DRAWINGS

[0016] Figure 1 It is a flowchart of an intelligent matching method for pipeline materials of an oil and gas module of the present invention. DETAILED DESCRIPTION OF THE EMBODIMENTS

[0017] The present invention will be described in detail below with reference to the drawings and specific embodiments.

[0018] As shown in FIG. 1, an intelligent matching method for pipeline materials of an oil and gas module of the present invention includes the following steps:

[0019] Step 1: Establish a 3D pipeline model of the oil and gas module to be constructed in 3D modeling software;

[0020] Step 2: Establish the spatial connection relationship between pipelines during construction by using the stroke method according to the 3D pipeline model of the oil and gas module to be constructed, and convert the 3D pipeline model of the oil and gas module to be constructed into a stroke unit model of the pipeline. The stroke unit model of the pipeline includes pipeline information, and the information includes: pipeline diameter, pipeline length, included angle between pipelines, and spatial information between pipelines. The specific steps can be as follows:

[0021] 201. Perform data preprocessing on the 3D pipeline model, remove irrelevant models (such as pipeline suspension bracket models) from the 3D pipeline model, and at the same time clean up noise data and delete duplicate data to ensure data accuracy;

[0022] 202. Determine an appropriate grid size according to the pipeline size range (such as a certain multiple of the minimum pipeline diameter), and divide the 3D pipeline model into basic stroke unit models using uniform division or an adaptive division rule based on the pipeline distribution density. In the software, establish an index for each stroke unit model, and then connect the stroke units that are adjacent in space position or functionally related according to the actual spatial layout relationship of the pipelines;

[0023] Step 3. Construct a feature dataset of the required pipelines according to the geometric features of the pipelines (including pipeline length, pipeline radius, etc.). The feature dataset F of the Nth required pipeline N is expressed as F N =[X 1 , X 2 , X 3 , where X 1 is the diameter of the Nth required pipeline, X 2 is the length of the Nth required pipeline, and X 3 is the vector expression of the Nth required pipeline in space (the included angle between two pipelines can be obtained through the expressions of two adjacent pipelines);

[0024] Step 4. Establish a membership matrix D for all required pipelines;

[0025] D = [F 1 , F 2 … F G … F N T

[0026] In the formula, D is the membership matrix of all required pipelines, F 1 , F 2 , … F G … F N are the feature datasets of the required pipelines, and T is the matrix transpose symbol. F G is the feature dataset of the Gth required pipeline.

[0027] Step 5. Establish a pipeline management database for the pipelines in the warehouse and set a minimum threshold Δx for each pipeline inventory; input the membership matrix D of the required pipelines into the comparison intelligent neural network model of the required pipelines and the inventory pipelines, calculate the geometric feature similarity and pipeline material feature matching degree between the required pipelines and the inventory pipelines, obtain the matching situation between the required pipelines and the inventory pipelines, and output the matching result. The specific steps of the comparison scheme are as follows;

[0028] ​501. Establish a pipeline management database for the pipelines in the warehouse. To better achieve the pipeline matching in the oil and gas module, the minimum threshold Δx of the inventory of each pipeline should be reasonably designed according to the importance and usage of each pipeline.

[0029] 502. Use TensorFlow software to establish a comparative intelligent neural network analysis model for whether the required pipelines and the inventory pipelines match.

[0030] 503. Input the membership matrix into the comparative intelligent neural network analysis model to calculate the geometric feature matching degree and pipeline material feature matching degree between the required pipelines and the inventory pipelines. The specific calculation formula can be referred to: Wang Shuai, Han Suo, Xiao Hongyu. Multi-step matching algorithm for pipeline data considering geometric and scene structure similarity [J]. Science of Surveying and Mapping, 2023, 48(08): 210-219. The following briefly introduces this method:

[0031] The formula for the geometric feature matching degree SI() is:

[0032]

[0033] In the formula: ω 1 , ω 2 and ω 3 are weight parameters and are positive numbers, ω 1 +ω 2 +ω 3 =1 (ω 1 The weight is 0.5, focusing on the matching degree related to the distance threshold; ω 2 The weight is 0.35, paying attention to the length ratio; ω 3 The weight is 0.15, considering the angle calculation.).; s 1 and s 2 are two pipelines stroke in the source matching set and the target matching set; H() is the MHD algorithm (the distance threshold is HT), used to calculate the distance similarity between two strokes; s 1 and s 2 The length ratio calculation method is LR(), reflecting the length matching degree; s 1 and s 2 The angle calculation method is A() (the angle threshold is AT), measuring the angle matching situation.

[0034] The pipeline material feature matching degree SIM() is:

[0035]

[0036] Where: a certain pipeline in the pipeline requirement list is S, the candidate pipeline matching S in the inventory is S', len(i) is the sum of the lengths of two mutually matching strokes in the i-th matching pair of the required pipeline and the inventory pipeline; SI(i) represents the geometric feature matching degree of the two mutually matching strokes in the i-th matching pair of the required pipeline and the inventory pipeline. n represents the number of strokes of S in the spatial scene where the two mutually matching strokes in the i-th matching pair are located; m represents the number of strokes of S' in the spatial scene where the two mutually matching strokes in the i-th matching pair are located.

[0037] 504. Determine the matching situation between the required pipeline and the inventory pipeline according to the calculated matching degree, and output the matching result.

[0038] Step Six: According to the comparison result, if there is no such pipeline in the warehouse, directly remove the pipeline that does not exist in the warehouse pipeline database from the pipeline requirement list and mark it as a shortage pipeline for subsequent procurement. Then, sort the remaining pipelines in the material requirement list in ascending order of their numbers, and proceed to the next step;

[0039] Step Seven: Compare the required quantity of the first pipeline with the inventory quantity n of the pipeline. If the inventory pipeline quantity ≥ the required quantity, mark it as an out-of-warehouse pipeline on the pipeline requirement list, and modify the current inventory quantity to the original inventory quantity minus the required inventory quantity in the pipeline management database; otherwise, issue all pipelines of this type from the pipeline management database and mark it as a pipeline with insufficient inventory on the pipeline requirement list;

[0040] Step Eight: Determine the inventory situation of other pipelines on the pipeline requirement list and make relevant marks in turn according to the method in Step Seven until all pipelines are matched;

[0041] Step Nine: Compare the inventory quantity of each pipeline in the warehouse with the minimum threshold Δx in turn. If it is lower than the minimum threshold required for the warehouse inventory, send the list to the pipeline procurement department for pipeline procurement.

[0042] The above-described embodiments are only descriptions of the preferred embodiments of the present invention, and do not limit the scope of the present invention. Without departing from the design spirit of the present invention, various deformations and improvements made by those of ordinary skill in the art to the technical solutions of the present invention shall fall within the protection scope determined by the claims of the present invention.

Claims

1. A method for intelligent matching of oil and gas module pipeline materials, characterized in that The following steps are involved: Step 1: Create a 3D model of the pipeline of the oil and gas module to be constructed in the 3D modeling software; Step 2: According to the pipeline 3D model of the oil and gas module to be constructed, the stroke method is used to establish the spatial connection relationship between the pipelines during construction, and the pipeline 3D model of the oil and gas module to be constructed is converted into a stroke unit model of the pipeline. The stroke unit model includes pipeline information, including: pipeline diameter, pipeline length, angle between pipelines, and spatial information between pipelines. Step 3: Construct a feature data set of the required pipeline according to the geometric features of the pipeline; Step 4: Establish a membership matrix D for all demand pipelines; Step 5: Establish a pipeline management database for pipelines in the warehouse and set the minimum threshold Δx for the inventory of each pipeline; input the membership matrix D of the demand pipeline into the intelligent neural network model for comparing the demand pipeline and the inventory pipeline, calculate the geometric feature similarity and pipeline material feature matching degree of the demand pipeline and the inventory pipeline, obtain the matching situation of the required pipeline and the inventory pipeline, and output the matching result; Step 6: According to the comparison results, if there is no pipeline in the warehouse, the pipeline that is not in the warehouse pipeline database will be directly removed from the pipeline demand list and marked as a pipeline in short supply for subsequent purchase. Then, the remaining pipelines in the material demand list are sorted in ascending order of numbers, and the next step is executed; Step 7: Compare the required number of the first pipeline with the number of pipelines in stock n. If the number of pipelines in stock is ≥ the required number, mark it as a pipeline that has been shipped out on the pipeline demand sheet, and modify the current inventory quantity in the pipeline management database to the original inventory quantity minus the required inventory quantity; otherwise, all pipelines of this type are shipped out in the pipeline management database, and marked as pipelines with insufficient inventory on the pipeline demand sheet; Step 8: Determine the inventory status of other pipelines on the pipeline demand list in accordance with the method in step 7 and make relevant marks until all pipelines are matched; Step nine, compare the inventory quantity of each pipeline in the warehouse with the minimum threshold Δx in turn. If it is lower than the minimum threshold required for the warehouse inventory, send the list to the pipeline procurement department for pipeline procurement.

2. The intelligent matching method for oil and gas module pipeline materials according to claim 1 is characterized in that: The step 2 specifically includes the following steps:

201. Perform data preprocessing on the pipeline 3D model, remove irrelevant models in the pipeline 3D model, clean up noise data, and delete duplicate data; 202. Determine the appropriate grid size according to the pipeline size range, and divide the pipeline 3D model into basic stroke unit models by uniform division or adaptive division rules based on pipeline distribution density. In the software, create an index for each stroke unit model, and then connect the stroke units that are adjacent in space or related in function according to the actual spatial layout relationship of the pipeline.

3. The intelligent matching method for oil and gas module pipeline materials according to claim 1 or 2, characterized in that: The specific process in step 5 is as follows:

501. Establish a pipeline management database for pipelines in the warehouse and set a minimum threshold Δx for the inventory of each pipeline; 502. Use Tensorflow software to establish a comparative intelligent neural network analysis model for whether the demand pipeline matches the inventory pipeline; 503. Input the membership matrix into the comparative intelligent neural network analysis model to calculate the geometric feature matching degree and pipeline material feature matching degree of the demand pipeline and the inventory pipeline; 504. According to the calculated matching degree, determine the matching status of the required pipeline and the inventory pipeline, and output the matching result.

Images