A material storage control method and system for a welded pipe production line

Abstract

The present invention relates to the technical field of welded pipe stock management and control, and provides a welded pipe production line stock management and control method and system, which includes: obtaining multi-dimensional stock characteristic information and welding parameter weights corresponding to the multi-dimensional stock characteristic information, and obtaining process compatibility according to the multi-dimensional stock characteristic information and the welding parameter weights; obtaining multi-dimensional stock defect information, obtaining a comprehensive defect coefficient according to the multi-dimensional stock defect information, and obtaining a stock quality index according to the comprehensive defect coefficient; obtaining the latest delivery time of the welded pipe order, and obtaining the urgency of the welded pipe order delivery according to the current time and the latest delivery time of the welded pipe order; obtaining the order priority according to the process compatibility, the stock quality index and the urgency of the welded pipe order delivery, and scheduling the stock according to the order priority. The present invention comprehensively considers the multi-dimensional factors that may affect the overall production efficiency of the welded pipe production line, and can improve the intelligence level and overall production efficiency of the welded pipe production line.

Description

Technical Field

[0001] The present invention relates to the technical field of welded pipe material storage control, and in particular to a welded pipe production line material storage control method and system. Background Art

[0002] Welded steel pipe, also known as welded pipe, is a steel pipe made by welding steel plate or strip after curling. The blank used for welded steel pipe is steel plate or strip. Due to different welding processes, it is divided into furnace welded pipe, electric welded (resistance welded) pipe and automatic arc welded pipe. Due to different welding forms, it is divided into straight seam welded pipe and spiral welded pipe. Due to the shape of the end, it is divided into round welded pipe and special-shaped (square, flat, etc.) welded pipe. Common materials for welded pipes include: Q235A, Q235C, Q235B, 16Mn, 20#, Q345, L245, L290, X42, X46, X60, X80, 0Cr13, 1Cr17, 00Cr19Ni11, 1Cr18Ni9, 0Cr18Ni11Nb, etc.

[0003] Welded pipe materials are generally stored in warehouses or factories, and are controlled and scheduled according to actual orders. For existing welded pipe production lines, the process is mostly that the management first allocates according to the order situation, and then sends it to the welding workshop. After the welding technicians receive the order, they allocate the corresponding welded pipe materials for welding according to the actual order situation. This material scheduling method for welded pipe production lines relies on the personal experience of the management, and does not consider factors such as the compatibility between materials and processes and the urgency of orders. It is easy to cause confusion in the scheduling of welded pipe storage materials, which in turn leads to the problem of low overall production efficiency of the welded pipe production line, which needs to be improved. Summary of the invention

[0004] Based on this, in order to solve the problems existing in the prior art, the present invention provides a material storage control method and system for a welded pipe production line, which comprehensively considers the multi-dimensional factors that may affect the overall production efficiency of the welded pipe production line by obtaining process compatibility, material storage quality index and urgency of welded pipe order delivery, and can automatically obtain the priority level of material storage welding, thereby improving the degree of intelligence and production efficiency. The specific technical scheme is as follows:

[0005] A material storage control method for a welded pipe production line comprises the following steps:

[0006] Acquire multi-dimensional stock material characteristic information and welding parameter weights corresponding to the multi-dimensional stock material characteristic information, and acquire process compatibility according to the multi-dimensional stock material characteristic information and the welding parameter weights;

[0007] Obtain multi-dimensional stock material defect information, obtain a comprehensive defect coefficient based on the multi-dimensional stock material defect information, and obtain a stock material quality index based on the comprehensive defect coefficient;

[0008] Get the latest delivery time of the welded pipe order, and get the urgency of the welded pipe order delivery based on the current time and the latest delivery time of the welded pipe order;

[0009] The order priority is obtained based on the process compatibility, stock material quality index and the urgency of the welded pipe order delivery date, and the stock material is scheduled according to the order priority.

[0010] The material storage control method for the welded pipe production line obtains process compatibility, material storage quality index and the urgency of welded pipe order delivery, and obtains the priority of order acceptance according to the process compatibility, material storage quality index and the urgency of welded pipe order delivery, and then schedules the material storage, comprehensively considering the multi-dimensional factors that may affect the overall production efficiency of the welded pipe production line, and can improve the intelligence level and overall production efficiency of the welded pipe production line, and solves the problem that the prior art does not consider the compatibility between materials and processes and the urgency of orders, which easily leads to chaotic scheduling of welded pipe material storage, thereby causing low overall production efficiency of the welded pipe production line.

[0011] Preferably, the order priority levels include:

[0012] A prediction model for predicting the quality decay of stockpiles is constructed based on LSTM neural network;

[0013] Select the range of orders that need to be re-prioritized according to the preset strategy;

[0014] Continuously adjust the priority of the orders in the selected range, and obtain the quality attenuation of all the stock for each order in the selected range through the prediction model during each adjustment;

[0015] Calculate the total amount of mass decay for all orders in the selected range;

[0016] Taking the minimum total quality attenuation and meeting the delivery date of the welded pipe order as the final priority adjustment condition, the priority of the orders in the selected range is adjusted, and the priority of all orders is obtained according to the adjusted priority of the orders in the selected range.

[0017] Preferably, process compatibility ;in, The number of dimensions representing the stock characteristic information, Indicates The storage characteristic values ​​of the dimensions, Indicates The welding parameter weights corresponding to the stock characteristic values ​​in each dimension.

[0018] Preferably, the stock quality index ;

[0019] in, represents the comprehensive defect coefficient after normalization, Indicates the preset quality index adjustment coefficient, comprehensive defect coefficient , The number of dimensions representing the stock defect information, Indicates The storage defect value of the dimension, Indicates The weight coefficient corresponding to the storage defect value of each dimension, Indicates Exponential sensitivity factors of stock defects in dimensions.

[0020] Preferably, order priority ;

[0021] in, Indicates the urgency of the delivery date of the welded pipe order. They respectively represent the weight coefficients of process compatibility, stock material quality index and welded pipe order delivery urgency.

[0022] A material storage control system for a welded pipe production line, used to implement the material storage control method for a welded pipe production line, comprising:

[0023] A process compatibility acquisition module is used to obtain multi-dimensional stock material characteristic information and welding parameter weights corresponding to the multi-dimensional stock material characteristic information, and obtain process compatibility according to the multi-dimensional stock material characteristic information and the welding parameter weights;

[0024] A stock material quality index acquisition module is used to acquire multi-dimensional stock material defect information, acquire a comprehensive defect coefficient based on the multi-dimensional stock material defect information, and acquire a stock material quality index based on the comprehensive defect coefficient;

[0025] The delivery urgency acquisition module is used to obtain the latest delivery time of the welded pipe order, and obtain the delivery urgency of the welded pipe order based on the current time and the latest delivery time of the welded pipe order;

[0026] The management and control module is used to obtain the order priority level according to the process compatibility, the stock material quality index and the urgency of the welded pipe order delivery date, and to schedule the stock material according to the order priority level.

[0027] Preferably, the management and control module includes:

[0028] A prediction model building unit, used to build a prediction model for predicting the quality decay of stockpiles based on an LSTM neural network;

[0029] The order range selection unit is used to select the order range that needs to be re-prioritized according to the preset strategy;

[0030] An adjustment unit is used to continuously adjust the priority of the orders within the selected range, and obtain the quality attenuation of all the stored materials for each order within the selected range through a prediction model during each adjustment;

[0031] The total amount of attenuation calculation unit is used to calculate the total amount of mass attenuation for all orders within the selected range;

[0032] The priority adjustment unit is used to adjust the priority of the selected range of orders with the minimum total quality attenuation and meeting the delivery period of the welded pipe order as the final priority adjustment condition, and obtain the priority of all orders according to the adjusted priority of the selected range of orders.

[0033] Preferably, the process compatibility acquisition module is based on the formula Obtain process compatibility;

[0034] in, The number of dimensions representing the stock characteristic information, Indicates The storage characteristic values ​​of the dimensions, Indicates The welding parameter weights corresponding to the stock characteristic values ​​in each dimension.

[0035] Preferably, the stock quality index acquisition module is based on the formula Obtain stock quality index;

[0036] in, represents the comprehensive defect coefficient after normalization, Indicates the preset quality index adjustment coefficient, comprehensive defect coefficient , The number of dimensions representing the stock defect information, Indicates The storage defect value of the dimension, Indicates The weight coefficient corresponding to the storage defect value of each dimension, Indicates The storage defect value of the dimension, Indicates The weight coefficient corresponding to the storage defect value of each dimension, Indicates Exponential sensitivity factors of stock defects in dimensions.

[0037] Preferably, the control module is based on the formula Get order priority level;

[0038] in, Indicates the urgency of the delivery date of the welded pipe order. They respectively represent the weight coefficients of process compatibility, stock material quality index and welded pipe order delivery urgency. BRIEF DESCRIPTION OF THE DRAWINGS

[0039] The present invention can be further understood from the following description in conjunction with the accompanying drawings. The components in the figures are not necessarily drawn to scale, but the emphasis is placed on illustrating the principles of the embodiments. In different views, the same reference numerals designate corresponding parts.

[0040] Figure 1 It is a schematic diagram of the overall process of a material storage control method for a welded pipe production line in one embodiment of the present invention;

[0041] Figure 2 is a flow chart of a specific method for obtaining an index sensitivity factor in one embodiment of the present invention;

[0042] Figure 3 is a flowchart of a specific method for obtaining order priority levels in one embodiment of the present invention;

[0043] Figure 4 It is a schematic diagram of the overall structure of a material storage control system for a welded pipe production line in one embodiment of the present invention;

[0044] Figure 5 It is a structural diagram of a control module in one embodiment of the present invention. DETAILED DESCRIPTION

[0045] In order to make the purpose, technical solution and advantages of the present invention more clearly understood, the present invention is further described in detail below in conjunction with its embodiments. It should be understood that the specific implementation methods described herein are only used to explain the present invention and do not limit the protection scope of the present invention.

[0046] It should be noted that when an element is referred to as being "fixed to" another element, it may be directly on the other element or there may be a central element. When an element is considered to be "connected to" another element, it may be directly connected to the other element or there may be a central element at the same time. The terms "vertical", "horizontal", "left", "right" and similar expressions used herein are for illustrative purposes only and do not represent the only implementation method.

[0047] Unless otherwise defined, all technical and scientific terms used herein have the same meaning as those commonly understood by those skilled in the art of the present invention. The terms used herein in the specification of the present invention are only for the purpose of describing specific embodiments and are not intended to limit the present invention. The term "and / or" used herein includes any and all combinations of one or more related listed items.

[0048] The “first” and “second” mentioned in the present invention do not represent specific quantities and orders, but are merely used to distinguish names.

[0049] Before describing the embodiments of the present invention, a brief introduction to the prior art is given first.

[0050] For welded pipe production lines, there are often multiple orders piled up. Different orders have different delivery times and different urgency levels. If the urgency of the orders is not reasonably and effectively sorted, and the orders are sorted and issued based on the personal experience of the managers, it is easy to cause confusion in the order and even the scheduling of welded pipe storage, which will reduce the overall production efficiency of the welded pipe production line.

[0051] In many cases, the different parameters and equipment status of the welding machine in the welding workshop will result in different compatibility with the storage material. The different compatibility between the welding machine and the storage material will not only lead to different equipment commissioning time, but also easily affect the welding quality of the welded pipe material.

[0052] In addition, in the welded pipe storage warehouse, due to the differences in suppliers, the impact of environmental changes, and the differences in warehousing quality standards, even the same type of welded pipe storage has different overall defect status and quality.

[0053] As for the existing storage management methods of welded pipe production lines, there is currently no method that can comprehensively consider process compatibility, welded pipe storage quality and order urgency to obtain order priority and thus realize intelligent storage scheduling.

[0054] In order to realize intelligent monitoring and scheduling of material storage of welded pipe production line and improve the overall intelligence level and production efficiency, an embodiment of the present invention provides a material storage control method for welded pipe production line, such as Figure 1 As shown, the following steps are included:

[0055] S1, obtaining multi-dimensional stock material characteristic information and welding parameter weights corresponding to the multi-dimensional stock material characteristic information, and obtaining process compatibility according to the multi-dimensional stock material characteristic information and the welding parameter weights. The multi-dimensional stock material characteristic information corresponds to the multi-dimensional welding parameter weights one by one.

[0056] For welded pipe stock, it is stored in the warehouse or workshop of the welded pipe production line by shelves or stacking. The stock characteristic information can be understood as reflecting the physical, chemical and process characteristics of the stock. The multi-dimensional stock characteristic information includes but is not limited to the material strength of the welded pipe stock (such as the tensile strength of Q235 steel), pipe diameter accuracy (such as ±0.5MM), surface treatment status (such as galvanized layer thickness) and wall thickness (such as 2.0 or 2.5MM). The welding parameter weight can be understood as the influence weight of different welding process parameters on the actual state of the welding equipment and the degree of matching of the stock. It can be dynamically adjusted according to the actual state of the welding equipment of the welded pipe production line, or set by the technicians according to experience. For the stock characteristic information and the welded pipe parameter weight, preferably, they are dimensionless and normalized to facilitate subsequent calculations.

[0057] Preferably, the method for dynamically adjusting the welding parameter weight according to the actual state of the welding equipment of the welded pipe production line includes: obtaining the basic parameters of the welding machine including rated power, electrode life, time of use, rated voltage, melting rate and welding speed, and assigning corresponding weight coefficients to different types of basic parameters, calculating the weighted average of the basic parameters and the weight coefficients, and characterizing the actual state of the welding machine with the obtained weighted average; constructing a mapping function of the actual state of the welding equipment and the welding parameter weight, and obtaining the multi-dimensional welding parameter weight according to the mapping function; wherein, in the mapping function, according to the actual state of the welding equipment, the welding parameter weight corresponding to the multi-dimensional material storage characteristic information is output. In this way, through the mapping function, the multi-dimensional welding parameter weight can be quickly obtained.

[0058] Specifically, for example, the welded pipe stock characteristics include material strength, pipe diameter accuracy, and surface treatment status. The welded pipe stock is 304 stainless steel. After dimensionless and normalized processing, the material strength , Pipe diameter accuracy And surface treatment status The current welding equipment has high requirements on material strength, followed by pipe diameter accuracy, and the weight of surface treatment status is relatively low. According to the actual status of the welding equipment and the mapping function, the corresponding welding weight parameters are obtained as follows: , , Therefore, the process compatibility can be obtained based on the multi-dimensional material storage characteristic information and welding parameter weights, which can be understood as the material strength. , Pipe diameter accuracy , Surface treatment status as well as , , Obtain process compatibility.

[0059] Preferably, process compatibility ;in, The number of dimensions representing the stock characteristic information, Indicates The storage characteristic values ​​of the dimensions, Indicates The welding parameter weights corresponding to the stock characteristic values ​​in each dimension.

[0060] Refer to the multi-dimensional material storage characteristics information and welding parameter weights listed above and substitute them into the formula Available, process compatibility It should be noted that for different stock materials, based on their different characteristic information, their process compatibility with welding equipment is also different, that is, they correspond to different process compatibilities.

[0061] According to the calculated process compatibility, the stock with high process compatibility can be dispatched to the welding pipe production line for welding. If the process compatibility of a batch of stock is less than the preset compatibility threshold, the early warning mechanism is triggered, requiring the welded pipe order corresponding to the batch of stock to be re-inspected to ensure the quality of the welded pipe.

[0062] S2, obtaining multi-dimensional stock material defect information, obtaining a comprehensive defect coefficient based on the multi-dimensional stock material defect information, and obtaining a stock material quality index based on the comprehensive defect coefficient.

[0063] The multi-dimensional defect information includes but is not limited to cracks, dents, rust, white spots and burrs. For a certain type of stock material, the multi-dimensional stock material defect information can be obtained by randomly sampling a number of stock materials and detecting the average value of the defect characteristics of the sampled stock materials.

[0064] Preferably, the stock quality index ;

[0065] in, represents the comprehensive defect coefficient after normalization, Indicates the preset quality index adjustment coefficient, comprehensive defect coefficient , The number of dimensions representing the stock defect information, Indicates The storage defect value of the dimension, Indicates The weight coefficients corresponding to the storage defect values ​​in each dimension can be set by the technicians. Indicates Exponential sensitivity factors for stock defects in dimensions.

[0066] The index sensitivity factor is used to quantify the influence of a specific type of defect on the stock quality index. By setting the index sensitivity factor, the key influencing factors of the stock quality index can be effectively identified, and it can be determined which defects play a leading role in the final stock quality index.

[0067] The existing method of obtaining the material quality index based on multi-dimensional defect information often obtains the material quality index by calculating the weighted average of multiple defects of different types, without considering the degree of influence of different types of defects on the quality index, and without distinguishing the defect information that plays a leading role and a general role in the final material quality index, and it is difficult to quantify the sensitivity of certain types of defect information to the quality index. This embodiment sets the index sensitivity factor, and can effectively identify the key influencing factors of the stock material quality index according to the actual situation in the welded pipe production process, determine which defects play a leading role in the final stock material quality index, and then obtain a more accurate stock material quality index, so as to assist in the accurate acquisition of the stock material priority level, which is conducive to improving the overall production efficiency.

[0068] For exponential sensitivity factors , generally speaking, its default value is 1.0. Specifically, if Exponential sensitivity factor of stock defects in dimensions ,but , which is equal to 1.5 power. Thus, by setting the exponential sensitivity factor , which can amplify the impact of certain types of defects on the quality index.

[0069] Preferably, if Figure 2 As shown, for the exponential sensitivity factor , and its acquisition method includes the following steps:

[0070] S21, obtain the number of defect customer complaints for the same type of welded pipe orders within the preset period , Regional sensitivity coefficient of the area where the welded pipe production line is located and the industry average recall cycle for welded pipe orders Calculate the market feedback correction index based on the number of defect complaints, regional sensitivity coefficient and industry average recall cycle .

[0071] Among them, for different regions, the same type of defects correspond to different regional sensitivity coefficients. For example, in the northern region, the weld brittleness increases under low temperature environment, and the regional sensitivity coefficient corresponding to the defect crack can be set to 1.2, and in the south it is set to 1.0 or 0.9. In coastal areas, since the salt spray environment will accelerate the expansion of corrosion and rust defects, the regional sensitivity coefficient corresponding to the defect rust can be set to 1.5, and in the northern inland areas it is set to 1.0, 0.9 or 0.8. By setting the regional sensitivity coefficient, the corresponding regional sensitivity coefficient can be assigned according to the geographical environmental risk corresponding to the welded pipe production line, and the differences in the impact of climate and usage scenarios on defects can be quantified.

[0072] The number of defect customer complaints for similar welded pipe orders within a preset period can be obtained based on the defect customer complaint data of the past 12 months. By adopting a rolling update mechanism to obtain the number of defect customer complaints, the timeliness of the data can be ensured. The source of the defect customer complaint data can be the complaint work orders recorded by the enterprise customer service system, which are obtained by classifying and counting by defect type. It is preferably an industry complaint database shared by industry associations to improve the quality and effectiveness of the data. One of the functions of obtaining the number of defect customer complaints is to reflect the market's tolerance for specific defects to a certain extent, so that the index sensitivity factor is consistent with the actual market situation. Of course, it is also possible to establish an API data interface between the enterprise customer service system and the industry complaint database shared by the industry association to update the number of defect customer complaints and the industry's average recall cycle in real time.

[0073] The industry average recall cycle can be understood as the average time (days) from defect discovery to completion of product recall, which can be used to reflect the industry's quality response efficiency. Specifically, you can refer to the annual quality report released by the industry association or key enterprise cases. Since the longer the average recall cycle, the lower the market trust, the cube root processing is used to weaken the long tail effect, reduce the dimensional difference of the recall cycle, and balance the nonlinear impact of the long cycle time.

[0074] Modify the index calculation formula based on market feedback In the above example, add one to the number of customer complaints and take the logarithm. , its role is to avoid interference from extreme values ​​while retaining the incremental nature of the data.

[0075] By quantifying market feedback and regional sensitivity coefficients, we can provide a dynamic basis for the correction of the index sensitivity factor of defects, and improve the timeliness and accuracy of the stockpile quality index. In addition, by introducing regional sensitivity coefficients, we can also correct the deviation of market data to a certain extent.

[0076] S22, obtaining the total number of searches for keywords corresponding to the defect through multiple search engines , the speed at which defect-related videos spread (times / hour) and the ratio of public opinion heat, based on the total search volume, propagation speed and the ratio of public opinion heat Calculate keyword spread index In this formula It comprehensively considers factors such as the search popularity of defect-related keywords, video dissemination rate, and public opinion comparison. It can be used to quantify online public opinion and dissemination dynamics, measure the dissemination effect of defect-related keywords, and reflect the attention and spread risk of defects in the public and the market.

[0077] Among them, the total number of searches It can be understood as the total search volume or weighted average of the search volume for a keyword in a preset time period (30 days or 45 days, etc.) on multiple search engines (such as Toutiao / Douyin / Baidu / 360, etc.), which reflects the popularity of users' active search behavior. The higher the value, the higher the attention paid to the keyword. A sudden increase may indicate that the relevant welded pipe quality accident has caused the spread of public opinion. Specifically, the standardized search index can be obtained through the search engine, and the weighted average of the search indexes of multiple search engines is used as the total search volume.

[0078] Propagation speed It is possible to understand the rate at which videos related to negative information (such as welded pipe defects) spread within 24 hours. One of its functions is to quantify the speed at which negative public opinion spreads. The faster the spread rate, the greater the potential public opinion risk. It can be calculated by counting the growth rate of forwarding, comments, likes and playback volume of relevant videos within 24 hours. The square root in is used to reduce the impact of extreme transmission events and filter network noise. For example, a video of a certain type of welded pipe defect was transmitted 100,000 times in 24 hours. The actual impact is calculated as Participate in calculations.

[0079] Public opinion heat ratio Used to reflect the relative attention of welded pipe defect events. It indicates the heat of corporate public opinion, which can be understood as the current heat value of corporate public opinion within a specific time window. It can be calculated by integrating media reports, social discussions and other data. It is mainly used to measure the intensity of public opinion attention on the company. The higher the value, the more active the public opinion. For corporate public opinion heat, the methods include: 1. Data monitoring, using public opinion monitoring systems (such as the WeChat Business Public Opinion Monitoring System, etc.) to monitor the big data of the entire network and filter out all mentions related to the company, including social blogs, news reports, forum posts, etc.; 2. Heat calculation, based on the filtered data, by constructing an indicator system to calculate the attention index, heat index, etc., to measure the heat of corporate public opinion.

[0080] It represents the industry benchmark heat, which can be understood as the average public opinion heat benchmark value of enterprises in the same industry in the same time period. It is mainly used to provide industry comparison standards to determine whether the public opinion heat of enterprises deviates from the normal range. The industry benchmark heat can be taken as a reference by taking the historical average of the same industry or the public opinion level of the leading enterprises, and can be updated synchronously every month to avoid the invalidation of the keyword communication index due to policy or market changes.

[0081] In this formula middle, It is used to integrate the user's active search behavior and passive propagation effect, and capture the critical point from potential concern to large-scale diffusion of welded pipe defects. By setting the square root, the exponential deviation caused by "viral transmission" can be avoided, which is more in line with the actual transmission attenuation law. It integrates the two factors of user active search and passive transmission, which not only reflects the real needs of users, but also quantifies the rate of information diffusion.

[0082] It is mainly used to make nonlinear adjustments through exponential functions to amplify the difference between the heat of corporate public opinion and the basic heat of the industry. By introducing the industry benchmark heat, it can prevent the extreme impact of the overheating or overcooling of individual corporate public opinion on the index. For the keyword communication index, a corresponding communication index threshold can be set. If it exceeds the communication index threshold for several consecutive days (such as 5 days or 7 days), the manual verification process is started to eliminate interference factors such as brushing. Of course, for the formula ,when When >2, the manual review mechanism can be triggered to eliminate interference from false information.

[0083] By obtaining the keyword propagation index , which can identify the focus of market attention and thus better adjust the index sensitivity factors of different types of defects.

[0084] S23, calculate the index sensitivity factor based on the market feedback correction index and keyword communication index ;in, They respectively represent the weight coefficients of the market feedback correction index and the keyword dissemination index.

[0085] By calculating the index sensitivity factor through the market feedback correction index and the keyword dissemination index, which integrates the market feedback mechanism and the Internet dissemination characteristics, it can more accurately amplify the impact of certain types of defects on the quality index, thereby assisting in the accurate acquisition of stockpile priority levels to improve overall production efficiency.

[0086] S3, obtain the latest delivery time of the welded pipe order, and obtain the urgency of the welded pipe order delivery according to the current time and the latest delivery time of the welded pipe order.

[0087] Here, the urgency of the delivery date of the welded pipe order can be obtained by calculating the difference between the current time and the latest delivery time of the welded pipe order (which can be understood as the delivery date of the welded pipe order), and then combining it with the customer level weighting.

[0088] S4, obtains the order priority according to the process compatibility, the stock material quality index and the urgency of the welded pipe order delivery date, and schedules the stock material according to the order priority.

[0089] Preferably, order priority ;in, Indicates the urgency of the delivery date of the welded pipe order. They respectively represent the weight coefficients of process compatibility, stock material quality index and welded pipe order delivery urgency.

[0090] Based on the calculated order priority, the stockpiles can be scheduled and matched according to the order priority, and the stockpiles of orders with high priority levels are transported to the welded pipe production line for welding first.

[0091] In summary, the material storage control method for the welded pipe production line obtains the process compatibility, the material storage quality index and the urgency of the welded pipe order delivery date, and obtains the priority of the order acceptance according to the process compatibility, the material storage quality index and the urgency of the welded pipe order delivery date, and then schedules the material storage, comprehensively considering the multi-dimensional factors that may affect the overall production efficiency of the welded pipe production line, and can improve the intelligence level and overall production efficiency of the welded pipe production line, and solves the problem that the existing technology does not consider the compatibility between materials and processes and the urgency of orders, which is prone to chaotic scheduling of welded pipe material storage, thereby leading to low overall production efficiency of the welded pipe production line.

[0092] As a preferred technical solution, Figure 3 As shown, in step S4, obtaining the order priority level includes:

[0093] S41, a prediction model for predicting stock quality decay is constructed based on LSTM neural network.

[0094] LSTM (Long Short-Term Memory) is a time recurrent neural network designed to solve the long-term dependency problem of general RNN (recurrent neural network). All RNNs have a chain form of repeated neural network modules, which are mainly used to process and predict important events with very long intervals and delays in time series. LSTM can effectively process and remember long-term dependency problems through its unique design structure.

[0095] The test model can be expressed as ;in, express The mass decay of stockpiles at a certain moment can be understood as the prediction target of the prediction model. It is usually measured in percentage, actual mass unit (such as tons) or stockpiles mass index, reflecting the mass loss of stockpiles due to environmental factors (such as oxidation, humidity, temperature) or physical stacking state (such as compaction degree).

[0096] is the weight matrix of environmental parameters, which is used to dynamically adjust the influence of different environmental variables on quality attenuation. The weight matrix is ​​automatically learned through the training of the LSTM network to capture the nonlinear relationship between environmental parameters and quality attenuation at different time steps. For example, temperature and humidity may correspond to higher weights because they directly affect the oxidation rate of materials, leading to an increase in rust defects, while the vibration frequency of the storage warehouse may indirectly affect quality attenuation through physical compaction, leading to an increase in indentation defects.

[0097] express The environmental parameter input vector at each moment includes key variables that affect the storage quality index, such as environmental monitoring data such as temperature, humidity, air pressure, vibration parameters of the storage warehouse (amplitude, frequency), stacking pressure, and storage time.

[0098] It is a bias term used to adjust the model output benchmark value and enhance the model's adaptability to different storage scenarios. Represents an activation function (such as Sigmoid, ReLU, or Tanh), which is used to introduce nonlinear relationships. In LSTM, the activation function is usually combined with a gating mechanism, for example: the Sigmoid function controls the information retention ratio (such as the forget gate); the Tanh function adjusts the output range.

[0099] Indicates the length of the historical time window, that is, the prediction model uses the past The environmental parameters of the time step predict the current quality decay. For example: The prediction model uses the environmental data from the past 7 days arrive Input LSTM network to predict the mass decay of storage materials The length of the historical window needs to be adjusted according to the data periodicity and attenuation mechanism.

[0100] Specifically, the relationship between the prediction model and the LSTM structure is as follows:

[0101] Input Layer: As the input sequence, it is processed by the gating mechanism (input gate, forget gate, output gate) of the LSTM unit to retain long-term dependencies.

[0102] Hidden layer: weight matrix Through back-propagation optimization, the importance of environment parameters at different time steps is dynamically adjusted.

[0103] Output layer: The LSTM hidden state is mapped to The predicted value of Join at this stage.

[0104] Preferably, the loss function of the prediction model for predicting the quality decay of stored materials based on the LSTM neural network is expressed as: .

[0105] in, Represents the dynamic loss term of the main task, which can also be understood as an improved weighted MAE-MSE hybrid loss term.

[0106] in, represents the temperature, humidity and pressure coupling coefficient and , Indicates the ambient temperature, Indicates the ambient humidity. Indicates the stacking pressure, Represents the Sigmoid function, which is used to realize the nonlinear fusion of environmental parameters. , , They represent the ambient temperature weight coefficient, ambient humidity weight coefficient and stacking pressure weight coefficient respectively, all of which can be set by technicians. The role of the temperature, humidity and pressure coupling coefficient is to quantify the comprehensive impact of environmental factors on mass attenuation. For example, in a high temperature and high humidity environment, Tends to 1, which can significantly amplify the loss value and force the prediction model to pay more attention to the prediction accuracy in harsh environments.

[0107] represents the time decay factor, Indicates the storage time of welded pipes. It indicates the corrosion rate parameter of welded pipe storage material, which can be obtained through experimental calibration, such as fitting in laboratory accelerated aging experiment, and its value range can be set to 0.01-0.1 / day. The role of is to reflect the nonlinear cumulative effect of storage time on quality attenuation; the initial stage ( When the value is small), the time decay factor Large, the prediction model focuses on short-term fluctuations; when the welded pipe stock is stored for a long time, Attenuation, the loss weight is reduced to avoid overfitting short-term noise.

[0108] It represents the number of samples, which can also be understood as the number of defect types of welded pipe stock in the training data set used to train the prediction model. It is used to convert the total loss into the average loss to avoid numerical instability caused by differences in sample size. It represents the error shape adjustment factor, which is an adjustable parameter and can be set by technicians. Its value range is 0.3-0.7, which is used to balance the weights of MAE (mean absolute error) and MSE (mean square error). When it approaches 1, MAE is dominant, reducing the sensitivity of abnormal values ​​(such as occasional sensor noise, etc.) and improving the robustness of the prediction model; When it approaches 0, MSE is dominant, and the optimization of the prediction accuracy of conventional working conditions is strengthened to enhance convergence.

[0109] Specifically, It can be selected as 0.6 to suppress the excessive penalty of MSE on outliers; when the quality of training data is high, It can be selected as 0.3 to improve the convergence efficiency by using the smooth gradient characteristics of MSE. It can be adjusted according to actual conditions.

[0110] Indicates The true value of the samples, Indicates The predicted value of a sample. Represents the MAE term, which directly measures the absolute deviation between the sample prediction value and the true value. It is generally applicable to scenarios where outliers need to be handled robustly. Represents the MSE term, which is used to amplify the impact of larger errors and force the prediction model to prioritize correcting predictions that deviate significantly from the true value.

[0111] In the main task dynamic loss term In, through and The prediction model for predicting the quality decay of storage materials based on the LSTM neural network can automatically adapt to the prediction requirements of different environmental conditions and welding pipe storage nodes. For example, in a high temperature and high humidity environment and long-term storage, the loss function pays more attention to the decay mode dominated by the environment rather than short-term random fluctuations. In this way, by setting the loss function of the prediction model with an improved weighted MAE-MSE mixed loss term, it is beneficial to dynamically adjust the loss weight according to the actual situation of the storage of welding pipe warehouses, thereby improving the accuracy of prediction.

[0112] Furthermore, the metal corrosion dynamics equation can be introduced to constrain the loss function of the prediction model to improve physical interpretability.

[0113] Specifically, . .in, , , They are all characteristic parameters of welded pipe storage materials, representing the corrosion rate constant, humidity influence index and temperature influence index of welded pipe storage materials respectively. Accelerated aging experiments (such as salt spray test chamber) can be used to simulate different temperature and humidity environmental conditions, measure the curve of corrosion amount changing with time, and use nonlinear least squares method to fit the parameters. For example, for carbon steel material Q235, the calibration results are obtained under 25-60 degrees Celsius and 60%-95% humidity conditions. , , .

[0114] are all power functions, Indicates time, The amount of material corrosion predicted by the model can be expressed in terms of mass loss percentage, thickness change, etc. It is used to indicate the degree of corrosion of the material under specific environmental conditions and its accuracy can be verified by actual test data. Represents the norm.

[0115] The corrosion rate constant of welded pipe stock can be understood as a rate constant related to the material type and surface treatment process, and is used to quantify the corrosion tendency of the material per unit time. The humidity influence index characterizes the nonlinear effect of humidity on the corrosion rate. When it is less than 1, it means that the corrosion rate slows down with increasing humidity. The temperature influence index can be understood as the contribution of temperature to the corrosion rate, which is usually related to the thermal expansion coefficient of the material and can be fitted through temperature control experiments. The value range of can be set to 0.5-1.2. The value range can be set to .

[0116] By introducing the metal corrosion dynamics equation as a constraint and as a regular term in the loss function, the physical interpretability and generalization ability of the prediction model for corrosion prediction can be improved.

[0117] S42, selecting a range of orders that need to readjust their priorities according to a preset strategy.

[0118] The preset strategy here can be set by technical personnel or management personnel. For example, after obtaining the order priority based on process compatibility, stock material quality index and welded pipe order delivery urgency, several later or consecutive order priorities are used as the order range that needs to readjust the priority. Specifically, if the order priority is obtained based on process compatibility, stock material quality index and welded pipe order delivery urgency, it is expressed as , according to the preset strategy, the order range that needs to be readjusted can be understood as , Or Select the order range that needs to be re-prioritized; among them, Indicates the number of welded pipe orders or order priorities in the selected range .

[0119] S43, continuously adjusting the priority of the orders in the selected range, and obtaining the quality attenuation of all the stored materials for each order in the selected range through the prediction model during each adjustment.

[0120] S44, calculating the total mass attenuation of all orders within the selected range.

[0121] The total mass attenuation of all orders within the selected range may be understood as the total mass attenuation of the stock of all orders within the selected range that need to readjust their priorities.

[0122] Each welded pipe order generally includes multiple welded pipes that need to be welded. When calculating the welding time of each welded pipe order, the historical welding data of welded pipe orders corresponding to the same type of stock can be obtained, and the historical welding time of multiple welded pipes can be obtained from the historical welding data. The average of the historical welding time of multiple welded pipes is used as the standard welding time, and the storage time of each welded pipe stock is calculated according to the standard welding time.

[0123] Specifically, if at the current moment, the order priority The corresponding storage time of welded pipe is , order priority The standard welding time of the corresponding welded pipe is , order priority The corresponding welded pipes have Root, then Root welded pipe storage time , ; If the order priority The standard welding time of the corresponding welded pipe is , order priority The corresponding welded pipes have Root, then Root welded pipe storage time , By analogy, the estimated storage time of each welded pipe stock of each order within the selected range can be obtained, and based on the estimated storage time, the mass attenuation of each welded pipe stock of each order can be obtained through the prediction model, and then the total mass attenuation of all orders within the selected range can be calculated.

[0124] For the orders within the selected range that need to readjust their priority levels, the mass attenuation weight of all stored materials is calculated for all orders before adjustment and after each adjustment.

[0125] S45, taking the minimum total quality attenuation and meeting the delivery date of the welded pipe order as the final priority adjustment condition, the priority of the orders in the selected range is adjusted, and the priority of all orders is obtained according to the adjusted priority of the orders in the selected range.

[0126] Specifically, if the order priority is expressed as And the order can be divided into segments, through Indicates that the order range that needs to be re-prioritized is ,right After adjustment, it is expressed as , then the priority of all orders obtained according to the adjusted priority of the selected range of orders is changed from the original ,become .

[0127] Through the above method, a prediction model for predicting the quality attenuation of stock materials is constructed based on the LSTM neural network, and the quality attenuation of all stock materials for each order in the selected range is obtained through the prediction model during each adjustment, and the total quality attenuation of all orders in the selected range is calculated. Finally, the priority of the orders in the selected range is adjusted based on the minimum total quality attenuation and meeting the delivery date of the welded pipe order as the final priority adjustment condition. This can not only obtain a more accurate total quality attenuation, but also help to further improve the intelligence level of the welded pipe production line and the overall production efficiency based on the minimum total quality attenuation and meeting the delivery date of the welded pipe order as the final priority adjustment condition.

[0128] An embodiment of the present invention further provides a material storage control system for a welded pipe production line, which is used to implement the material storage control method for a welded pipe production line, such as Figure 4 As shown, it includes a process compatibility acquisition module, a storage material quality index acquisition module, a delivery urgency acquisition module and a management and control module.

[0129] The process compatibility acquisition module is used to obtain multi-dimensional stock material characteristic information and welding parameter weights corresponding to the multi-dimensional stock material characteristic information, and obtain process compatibility based on the multi-dimensional stock material characteristic information and the welding parameter weights.

[0130] Preferably, the process compatibility acquisition module is based on the formula Get process compatibility. Among them, The number of dimensions representing the stock characteristic information, Indicates The storage characteristic values ​​of the dimensions, Indicates The welding parameter weights corresponding to the stock characteristic values ​​in each dimension.

[0131] For process compatibility, quantitative rules can be set. For example, when CP=1.0, it indicates full compatibility, the welded pipe stock can be directly transported to the welded pipe production line, and the welding equipment can be directly welded without debugging; when CP is greater than 0.8 and less than 1.0, it indicates high compatibility, and the welding equipment can be directly welded after slight debugging (such as adjusting only one of the process parameters); when CP is greater than 0.5 and less than 0.8, it indicates moderate compatibility, and the welding equipment can be welded after certain process parameter debugging (adjusting two or more process parameters, such as welding speed, power, shielding gas flow, etc.); if CP is less than 0.5, it indicates low compatibility or incompatibility, and multiple process parameters need to be adjusted and core components such as welding guns need to be replaced before the stock can be welded.

[0132] The stock material quality index acquisition module is used to obtain multi-dimensional stock material defect information, obtain a comprehensive defect coefficient based on the multi-dimensional stock material defect information, and obtain a stock material quality index based on the comprehensive defect coefficient.

[0133] Preferably, the stock quality index acquisition module is based on the formula Get the stock quality index. Among them, represents the comprehensive defect coefficient after normalization, Indicates the preset quality index adjustment coefficient. Generally speaking, the quality index adjustment coefficient can be Set to 1.0, comprehensive defect coefficient , The number of dimensions representing the stock defect information, Indicates The storage defect value of the dimension, Indicates The weight coefficient corresponding to the storage defect value of each dimension, Indicates Exponential sensitivity factors for stock defects in dimensions.

[0134] The index sensitivity factor is used to quantify the influence of a specific type of defect on the stock quality index. By setting the index sensitivity factor, the key influencing factors of the stock quality index can be effectively identified, and it can be determined which defects play a leading role in the final stock quality index.

[0135] The existing method of obtaining the material quality index based on multi-dimensional defect information often obtains the material quality index by calculating the weighted average of multiple defects of different types, without considering the degree of influence of different types of defects on the quality index, and without distinguishing the defect information that plays a leading role and a general role in the final material quality index, and it is difficult to quantify the sensitivity of certain types of defect information to the quality index. This embodiment sets the index sensitivity factor, and can effectively identify the key influencing factors of the stock material quality index according to the actual situation in the welded pipe production process, determine which defects play a leading role in the final stock material quality index, and then obtain a more accurate stock material quality index, so as to assist in the accurate acquisition of the stock material priority level, which is conducive to improving the overall production efficiency.

[0136] For exponential sensitivity factors , generally speaking, its default value is 1.0. Specifically, if Exponential sensitivity factor of stock defects in dimensions ,but , which is equal to 1.5 power. Thus, by setting the exponential sensitivity factor , which can amplify the impact of certain types of defects on the quality index.

[0137] The delivery urgency acquisition module is used to obtain the latest delivery time of the welded pipe order, and obtain the delivery urgency of the welded pipe order based on the current time and the latest delivery time of the welded pipe order.

[0138] The urgency of the delivery date of the welded pipe order can be obtained by calculating the difference between the current time and the latest delivery time of the welded pipe order (which can be understood as the delivery date of the welded pipe order), and then combining it with the customer level weighting. The customer level can be set by the enterprise management personnel based on experience.

[0139] The management and control module is used to obtain the order priority according to the process compatibility, stock material quality index and the urgency of the welded pipe order delivery date, and to schedule the stock materials according to the order priority, and to transport the welded pipe stock materials with high order priority to the welded pipe production line for welding.

[0140] Preferably, the control module is based on the formula Get the order priority level. Indicates the urgency of the delivery date of the welded pipe order. They respectively represent the weight coefficients of process compatibility, stock material quality index and welded pipe order delivery urgency.

[0141] Specifically, the weight coefficients of process compatibility, stock material quality index and welded pipe order delivery urgency. The typical value range can be set to 0.4-0.6, and the weight coefficient of the urgency of the delivery date of the welded pipe order can be increased when the capacity of the welded pipe production line is tight. The typical value range can be set at 0.2-0.35, and its weight value can be appropriately increased for high value-added welded pipe orders. The typical value range of can be set to 0.1-0.4, which can improve the weight coefficient of process compatibility in the scenario of multi-variety and small-batch welded pipe orders. Of course, according to the actual situation of the welded pipe manufacturer, the typical value range of can be set to 0.1-0.4, which can improve the weight coefficient of process compatibility in the scenario of multi-variety and small-batch welded pipe orders. Of course, according to the actual situation of the welded pipe manufacturer, Make appropriate and reasonable settings.

[0142] The acquisition of order priority integrates process compatibility, stock material quality index and the urgency of welded pipe order delivery, and comprehensively considers the multi-dimensional factors that may affect the overall production efficiency of the welded pipe production line. By setting the weight coefficients of process compatibility, stock material quality index and the urgency of welded pipe order delivery, the priority of stock material welding can be automatically obtained according to the actual production situation of the welded pipe manufacturer, which can improve its intelligence and production efficiency, avoid the chaos of welded pipe stock material scheduling due to failure to consider factors such as compatibility between materials and processes and order urgency, and thus lead to the problem of low overall production efficiency of the welded pipe production line.

[0143] To sum up, the material storage control system of the welded pipe production line obtains the process compatibility, material storage quality index and the urgency of the welded pipe order delivery date, and obtains the priority of the order acceptance according to the process compatibility, material storage quality index and the urgency of the welded pipe order delivery date, and then schedules the material storage, comprehensively considering the multi-dimensional factors that may affect the overall production efficiency of the welded pipe production line, which can improve the intelligence level and overall production efficiency of the welded pipe production line, and solves the problem that the existing technology does not consider the compatibility between materials and processes and the urgency of orders, which is prone to chaotic scheduling of welded pipe material storage, thereby leading to low overall production efficiency of the welded pipe production line.

[0144] As a preferred technical solution, the management and control module includes a prediction model building unit, an order range selection unit, an adjustment unit, a total attenuation calculation unit and a priority adjustment unit.

[0145] The prediction model building unit is used to build a prediction model for predicting the quality decay of stored materials based on the LSTM neural network. The core of LSTM is that it contains three "gate" structures: forget gate, input gate and output gate. These gates control the retention and discarding of information through the sigmoid activation function, thereby achieving a balance between long-term memory and short-term memory of information.

[0146] Forget gate: determines which information needs to be forgotten. It outputs a value between 0 and 1 through the sigmoid function, indicating the proportion of retention.

[0147] Input gate: decides what new information needs to be added to the cell state. It also creates a "candidate cell state" containing new candidate values.

[0148] Output gate: determines which information needs to be output. It uses the sigmoid function to determine which parts of the information need to be output.

[0149] For the prediction model, the loss function can be constructed , in order to improve its prediction accuracy. Furthermore, the metal corrosion dynamics equation is introduced as a constraint and used as a regular term in the loss function, namely , , improving the physical interpretability and generalization ability of the prediction model for corrosion prediction.

[0150] Furthermore, the loss function , , represents the environmental mutation penalty term, which is used to punish periods of drastic changes in temperature and humidity (such as or ) imposes additional penalties.

[0151] in, Represents a set of mutation time points, with a user-defined threshold (Temperature Change Rate Threshold) and (humidity change rate threshold) is filtered out. For example, when time, time point Included ; represents the gradient operator, which can be regarded as a differential operation of the time series, used to capture mutation signals and strengthen the response of the prediction model to extreme environments. Represents the penalty coefficient.

[0152] Represents the prediction model at time The predicted value of Represents the prediction model at time The true value of Represents the L2 norm (mean square error), which is used to measure the degree of deviation between the predicted value and the true value. represents the prediction error penalty, The gradient operator of the predicted value reflects the rate of change of the predicted result in the time dimension (e.g. ), Represents the L1 norm, which is used to constrain the change range of the predicted value during the mutation period to avoid drastic fluctuations in the model output.

[0153] The loss function Introduce environmental mutation penalty term, through mutation time point set Screening the mutation period and imposing penalties in a targeted manner can force the prediction model to pay more attention to prediction stability when the temperature and humidity change drastically; among them, the L2 norm error term ensures the prediction accuracy, and the L1 gradient term can suppress the abrupt change of the prediction value during the mutation period.

[0154] By further optimizing the loss function of the prediction model based on the LSTM neural network for predicting the mass attenuation of stored materials, the accuracy of the model in predicting the mass attenuation of welded pipe stored materials can be further improved.

[0155] The order range selection unit is used to select the order range that needs to readjust the priority level according to the preset strategy; the adjustment unit is used to continuously adjust the priority level of the orders in the selected range, and obtain the quality attenuation of all the stored materials for each order in the selected range through the prediction model during each adjustment.

[0156] The total attenuation calculation unit is used to calculate the total mass attenuation of all orders within the selected range; the priority adjustment unit is used to adjust the priority of the orders in the selected range with the minimum total mass attenuation and meeting the delivery period of the welded pipe order as the final priority adjustment condition, and obtain the priority of all orders according to the adjusted priority of the orders in the selected range.

[0157] Through the above method, a prediction model for predicting the quality attenuation of stock materials is constructed based on the LSTM neural network, and the quality attenuation of all stock materials for each order in the selected range is obtained through the prediction model during each adjustment, and the total quality attenuation of all orders in the selected range is calculated. Finally, the priority of the orders in the selected range is adjusted based on the minimum total quality attenuation and meeting the delivery date of the welded pipe order as the final priority adjustment condition. This can not only obtain a more accurate total quality attenuation, but also help to further improve the intelligence level of the welded pipe production line and the overall production efficiency based on the minimum total quality attenuation and meeting the delivery date of the welded pipe order as the final priority adjustment condition.

[0158] The technical features of the above-described embodiments may be arbitrarily combined. To make the description concise, not all possible combinations of the technical features in the above-described embodiments are described. However, as long as there is no contradiction in the combination of these technical features, they should be considered to be within the scope of this specification.

[0159] The above-mentioned embodiments only express several implementation methods of the present invention, and the descriptions thereof are relatively specific and detailed, but they cannot be understood as limiting the scope of the invention patent. It should be pointed out that, for ordinary technicians in this field, several variations and improvements can be made without departing from the concept of the present invention, and these all belong to the protection scope of the present invention. Therefore, the protection scope of the patent of the present invention shall be subject to the attached claims.

Claims

1. A material storage control method for a welded pipe production line, characterized in that ,include: Acquire multi-dimensional stock material characteristic information and welding parameter weights corresponding to the multi-dimensional stock material characteristic information, and acquire process compatibility according to the multi-dimensional stock material characteristic information and the welding parameter weights; Obtain multi-dimensional stock material defect information, obtain a comprehensive defect coefficient based on the multi-dimensional stock material defect information, and obtain a stock material quality index based on the comprehensive defect coefficient; Get the latest delivery time of the welded pipe order, and get the urgency of the welded pipe order delivery based on the current time and the latest delivery time of the welded pipe order; Obtain order priority based on process compatibility, stock material quality index, and welded pipe order delivery urgency, and schedule stock materials based on order priority; The welded pipe production line material storage control method also includes: A prediction model for predicting the quality decay of stockpiles is constructed based on LSTM neural network; Select the range of orders that need to be re-prioritized according to the preset strategy; Continuously adjust the priority of the orders in the selected range, and obtain the quality attenuation of all the stock for each order in the selected range through the prediction model during each adjustment; Calculate the total amount of mass decay for all orders within the selected range; Taking the minimum total quality attenuation and meeting the delivery date of the welded pipe order as the final priority adjustment condition, the priority of the selected range of orders is adjusted, and the priority of all orders is obtained according to the adjusted priority of the selected range of orders; Process compatibility Where n represents the number of dimensions of the storage characteristic information, s i represents the storage characteristic value of the i-th dimension, w i represents the welding parameter weight corresponding to the stock characteristic value of the i-th dimension; Stock quality index Q = ln(λ×(1-F')); Among them, F' represents the comprehensive defect coefficient after normalization, λ represents the preset quality index adjustment coefficient, and the comprehensive defect coefficient m represents the number of dimensions of the stock defect information, f i represents the storage defect value of the i-th dimension, α i represents the weight coefficient corresponding to the stock defect value of the i-th dimension, and ki represents the exponential sensitivity factor of the stock defect of the i-th dimension.

2. A method for controlling material storage in a welded pipe production line according to claim 1, characterized in that ,Order priority Among them, ED represents the urgency of the delivery date of the welded pipe order, k1, k2, and k3 represent the weight coefficients of process compatibility, stock material quality index, and the urgency of the delivery date of the welded pipe order, respectively.

3. A material storage control system for a welded pipe production line, used to implement the material storage control method for a welded pipe production line as described in any one of claims 1-2, characterized in that ,include: A process compatibility acquisition module is used to acquire multi-dimensional stock material characteristic information and welding parameter weights corresponding to the multi-dimensional stock material characteristic information, and acquire process compatibility according to the multi-dimensional stock material characteristic information and the welding parameter weights; A stock material quality index acquisition module is used to acquire multi-dimensional stock material defect information, acquire a comprehensive defect coefficient based on the multi-dimensional stock material defect information, and acquire a stock material quality index based on the comprehensive defect coefficient; The delivery urgency acquisition module is used to obtain the latest delivery time of the welded pipe order, and obtain the delivery urgency of the welded pipe order based on the current time and the latest delivery time of the welded pipe order; The control module is used to obtain the order priority level according to the process compatibility, stock material quality index and the urgency of the welded pipe order delivery date, and to schedule the stock material according to the order priority level; The control modules include: A prediction model building unit, used to build a prediction model for predicting the quality decay of stock based on an LSTM neural network; An order range selection unit is used to select the order range that needs to readjust the priority level according to the preset strategy; An adjustment unit is used to continuously adjust the priority of the orders within the selected range, and obtain the quality attenuation of all the stored materials for each order within the selected range through a prediction model during each adjustment; A total attenuation calculation unit is used to calculate the total mass attenuation of all orders within a selected range; The priority adjustment unit is used to adjust the priority of the selected range of orders with the minimum total quality attenuation and meeting the delivery date of the welded pipe order as the final priority adjustment condition, and obtain the priority of all orders according to the adjusted priority of the selected range of orders; The process compatibility acquisition module is based on the formula Obtain process compatibility; Where n represents the number of dimensions of the storage characteristic information, s i represents the storage characteristic value of the i-th dimension, w i represents the welding parameter weight corresponding to the stock characteristic value of the i-th dimension; The stock quality index acquisition module acquires the stock quality index according to the formula Q=ln(λ×(1-F')); Among them, F' represents the comprehensive defect coefficient after normalization, λ represents the preset quality index adjustment coefficient, and the comprehensive defect coefficient m represents the number of dimensions of the stock defect information, f i represents the storage defect value of the i-th dimension, α i represents the weight coefficient corresponding to the stock defect value of the i-th dimension, and ki represents the exponential sensitivity factor of the stock defect of the i-th dimension.

4. A material storage control system for a welded pipe production line as claimed in claim 3, characterized in that: The control module is based on the formula Get order priority level; Among them, ED represents the urgency of the delivery date of the welded pipe order, k1, k2, and k3 represent the weight coefficients of process compatibility, stock material quality index, and the urgency of the delivery date of the welded pipe order, respectively.

Images