CN118940043B - Stainless steel pipe processing control method and device based on defect detection - Google Patents
Stainless steel pipe processing control method and device based on defect detection Download PDFInfo
- Publication number
- CN118940043B CN118940043B CN202411096565.4A CN202411096565A CN118940043B CN 118940043 B CN118940043 B CN 118940043B CN 202411096565 A CN202411096565 A CN 202411096565A CN 118940043 B CN118940043 B CN 118940043B
- Authority
- CN
- China
- Prior art keywords
- processing
- defect
- rate
- nodes
- machining
- Prior art date
- Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
- Active
Links
Classifications
-
- G—PHYSICS
- G01—MEASURING; TESTING
- G01N—INVESTIGATING OR ANALYSING MATERIALS BY DETERMINING THEIR CHEMICAL OR PHYSICAL PROPERTIES
- G01N21/00—Investigating or analysing materials by the use of optical means, i.e. using sub-millimetre waves, infrared, visible or ultraviolet light
- G01N21/84—Systems specially adapted for particular applications
- G01N21/88—Investigating the presence of flaws or contamination
- G01N21/8851—Scan or image signal processing specially adapted therefor, e.g. for scan signal adjustment, for detecting different kinds of defects, for compensating for structures, markings, edges
-
- G—PHYSICS
- G06—COMPUTING OR CALCULATING; COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F18/00—Pattern recognition
- G06F18/20—Analysing
- G06F18/21—Design or setup of recognition systems or techniques; Extraction of features in feature space; Blind source separation
- G06F18/214—Generating training patterns; Bootstrap methods, e.g. bagging or boosting
-
- G—PHYSICS
- G06—COMPUTING OR CALCULATING; COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F18/00—Pattern recognition
- G06F18/20—Analysing
- G06F18/22—Matching criteria, e.g. proximity measures
-
- G—PHYSICS
- G06—COMPUTING OR CALCULATING; COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F18/00—Pattern recognition
- G06F18/20—Analysing
- G06F18/24—Classification techniques
- G06F18/241—Classification techniques relating to the classification model, e.g. parametric or non-parametric approaches
- G06F18/2415—Classification techniques relating to the classification model, e.g. parametric or non-parametric approaches based on parametric or probabilistic models, e.g. based on likelihood ratio or false acceptance rate versus a false rejection rate
-
- G—PHYSICS
- G06—COMPUTING OR CALCULATING; COUNTING
- G06N—COMPUTING ARRANGEMENTS BASED ON SPECIFIC COMPUTATIONAL MODELS
- G06N7/00—Computing arrangements based on specific mathematical models
- G06N7/01—Probabilistic graphical models, e.g. probabilistic networks
-
- G—PHYSICS
- G06—COMPUTING OR CALCULATING; COUNTING
- G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
- G06T7/00—Image analysis
- G06T7/0002—Inspection of images, e.g. flaw detection
- G06T7/0004—Industrial image inspection
-
- G—PHYSICS
- G01—MEASURING; TESTING
- G01N—INVESTIGATING OR ANALYSING MATERIALS BY DETERMINING THEIR CHEMICAL OR PHYSICAL PROPERTIES
- G01N21/00—Investigating or analysing materials by the use of optical means, i.e. using sub-millimetre waves, infrared, visible or ultraviolet light
- G01N21/84—Systems specially adapted for particular applications
- G01N21/88—Investigating the presence of flaws or contamination
- G01N21/8851—Scan or image signal processing specially adapted therefor, e.g. for scan signal adjustment, for detecting different kinds of defects, for compensating for structures, markings, edges
- G01N2021/8883—Scan or image signal processing specially adapted therefor, e.g. for scan signal adjustment, for detecting different kinds of defects, for compensating for structures, markings, edges involving the calculation of gauges, generating models
Landscapes
- Engineering & Computer Science (AREA)
- Physics & Mathematics (AREA)
- Theoretical Computer Science (AREA)
- Data Mining & Analysis (AREA)
- General Physics & Mathematics (AREA)
- Computer Vision & Pattern Recognition (AREA)
- General Engineering & Computer Science (AREA)
- Life Sciences & Earth Sciences (AREA)
- Artificial Intelligence (AREA)
- Evolutionary Computation (AREA)
- Evolutionary Biology (AREA)
- Bioinformatics & Computational Biology (AREA)
- Bioinformatics & Cheminformatics (AREA)
- Probability & Statistics with Applications (AREA)
- Mathematical Optimization (AREA)
- Signal Processing (AREA)
- Quality & Reliability (AREA)
- Mathematical Physics (AREA)
- Software Systems (AREA)
- Pure & Applied Mathematics (AREA)
- Algebra (AREA)
- Mathematical Analysis (AREA)
- Computational Mathematics (AREA)
- Computing Systems (AREA)
- Health & Medical Sciences (AREA)
- Chemical & Material Sciences (AREA)
- Analytical Chemistry (AREA)
- Biochemistry (AREA)
- General Health & Medical Sciences (AREA)
- Immunology (AREA)
- Pathology (AREA)
- General Factory Administration (AREA)
Abstract
本发明公开了基于缺陷检测的不锈钢管加工控制方法及装置，涉及智能控制技术领域。所述方法包括：连接不锈钢管加工控制系统，获取N个加工节点；建立N个数据检测区块，获取加工检测数据集，生成历史加工检测样本库；对历史加工检测样本库进行缺陷识别，获取N个加工节点对应的N个加工缺陷记忆库；建立N个概率密度函数；获取N个实时加工速率，对N个实时加工速率进行缺陷概率计算，输出N个缺陷概率；不锈钢管加工控制系统接收N个缺陷概率，生成预警信号，用于对异常加工节点进行加工速率控制。解决了现有技术中由于加工速率变化导致钢管出现缺陷的技术问题，通过智能控制异常加工节点的加工速率，达到了提高加工质量的技术效果。
The present invention discloses a stainless steel pipe processing control method and device based on defect detection, which relates to the field of intelligent control technology. The method includes: connecting a stainless steel pipe processing control system to obtain N processing nodes; establishing N data detection blocks, obtaining processing detection data sets, and generating a historical processing detection sample library; performing defect identification on the historical processing detection sample library, and obtaining N processing defect memory libraries corresponding to N processing nodes; establishing N probability density functions; obtaining N real-time processing rates, performing defect probability calculation on the N real-time processing rates, and outputting N defect probabilities; the stainless steel pipe processing control system receives N defect probabilities and generates an early warning signal for controlling the processing rate of abnormal processing nodes. The method solves the technical problem of defects in steel pipes due to changes in processing rate in the prior art, and achieves the technical effect of improving processing quality by intelligently controlling the processing rate of abnormal processing nodes.
Description
Technical Field
The invention relates to the technical field of intelligent control, in particular to a stainless steel tube processing control method and device based on defect detection.
Background
In the field of production and processing of stainless steel pipes, with the increase of market competition and the continuous improvement of product quality requirements of customers, the balance of processing efficiency and product quality is the focus of industry attention. In the conventional machining process, in order to pursue higher production efficiency, the machining rate is often increased, but such an increase in the machining rate is often accompanied by an increase in defects of the steel pipe, such as cracks, burrs, surface roughness, and the like. These defects not only affect the appearance quality of the steel pipe, but may also have serious effects on the service performance and life thereof. Therefore, how to effectively control and reduce the defects of the steel pipe while ensuring the processing efficiency becomes a technical problem to be solved urgently in the industry.
Disclosure of Invention
The embodiment of the application provides a stainless steel tube processing control method and device based on defect detection, which solve the technical problem that the steel tube is defective due to the change of processing speed in the prior art.
In view of the above problems, the embodiments of the present application provide a method and apparatus for controlling processing of stainless steel pipes based on defect detection.
In a first aspect of the embodiment of the present application, there is provided a stainless steel pipe processing control method based on defect detection, the method comprising:
The method comprises the steps of connecting a stainless steel pipe machining control system to obtain N machining nodes, establishing N data detection blocks according to the N machining nodes, obtaining corresponding machining detection data sets under each machining rate according to the N data detection blocks to generate a historical machining detection sample library, carrying out defect identification on the historical machining detection sample library to obtain N machining defect memory libraries corresponding to the N machining nodes, wherein each machining defect memory library comprises machining defect characteristics corresponding to each machining rate, establishing N probability density functions according to the N machining defect memory libraries, obtaining N real-time machining rates, carrying out defect probability calculation on the N real-time machining rates according to the N probability density functions, and outputting N defect probabilities, and the stainless steel pipe machining control system receives the N defect probabilities to generate early warning signals for controlling abnormal machining nodes.
In a second aspect of the embodiment of the present application, there is provided a stainless steel pipe machining control device based on defect detection, the device comprising:
The system comprises a stainless steel pipe processing control system, a node acquisition module, a sample library generation module, a defect identification module and a function establishment module, wherein the node acquisition module is used for connecting the stainless steel pipe processing control system to acquire N processing nodes, the sample library generation module is used for establishing N data detection blocks according to the N processing nodes, acquiring corresponding processing detection data sets under each processing rate according to the N data detection blocks to generate a historical processing detection sample library, the defect identification module is used for carrying out defect identification on the historical processing detection sample library to acquire N processing defect memory libraries corresponding to the N processing nodes, each processing defect memory library comprises processing defect characteristics corresponding to each processing rate, the function establishment module is used for establishing N probability density functions according to the N processing defect memory libraries, the calculation module is used for acquiring N real-time processing rates, carrying out defect probability calculation on the N real-time processing rates according to the N probability density functions and outputting N defect probabilities, and the control module is used for receiving the N defect probabilities by the stainless steel pipe processing control system to generate an abnormal processing rate early warning signal and carrying out abnormal processing rate control on the processing nodes.
One or more technical schemes provided by the application have at least the following technical effects or advantages:
Firstly, a stainless steel pipe processing control system is connected to obtain N processing nodes. And then, establishing N data detection blocks according to the N processing nodes, acquiring corresponding processing detection data sets under each processing rate according to the N data detection blocks, and generating a historical processing detection sample library. And then, carrying out defect identification on the historical processing detection sample library to obtain N processing defect memory libraries corresponding to the N processing nodes, wherein each processing defect memory library comprises processing defect characteristics corresponding to each processing rate. And then, according to the N processing defect memory banks, establishing N probability density functions. Further, N real-time processing rates are obtained, defect probability calculation is carried out on the N real-time processing rates according to the N probability density functions, and N defect probabilities are output. And finally, the stainless steel pipe machining control system receives the N defect probabilities and generates an early warning signal for controlling the machining rate of the abnormal machining node. The technical problem that defects appear in the steel pipe due to the change of the machining speed in the prior art is solved, and the technical effect of improving the machining quality is achieved by intelligently controlling the machining speed of abnormal machining nodes.
Drawings
In order to more clearly illustrate the technical solutions of the embodiments of the present invention, the following description will briefly explain the drawings needed in the description of the embodiments, which are merely examples of the present invention, and other drawings can be obtained according to these drawings without inventive effort for a person skilled in the art.
FIG. 1 is a schematic flow chart of a stainless steel tube processing control method based on defect detection according to an embodiment of the present application;
fig. 2 is a schematic structural diagram of a stainless steel tube processing control device based on defect detection according to an embodiment of the present application.
Reference numerals illustrate a node acquisition module 11, a sample library generation module 12, a defect identification module 13, a function establishment module 14, a calculation module 15 and a control module 16.
Detailed Description
The embodiment of the application solves the technical problem that the steel pipe is defective due to the change of the processing speed in the prior art by providing the stainless steel pipe processing control method and the stainless steel pipe processing control device based on defect detection.
The technical solutions in the embodiments of the present application will be clearly and completely described below with reference to the accompanying drawings in the embodiments of the present application. It will be apparent that the described embodiments are only some, but not all, embodiments of the application. All other embodiments, which can be made by those skilled in the art based on the embodiments of the application without making any inventive effort, are intended to be within the scope of the application.
It should be noted that the terms "comprises" and "comprising" are intended to cover a non-exclusive inclusion, such that a process, method, system, article, or server that comprises a list of steps or elements is not necessarily limited to those steps or elements expressly listed or inherent to such process, method, article, or apparatus.
Example 1
As shown in fig. 1, an embodiment of the present application provides a method for controlling processing of a stainless steel pipe based on defect detection, wherein the method includes:
And connecting a stainless steel pipe processing control system to obtain N processing nodes.
The stainless steel pipe processing control system is used for controlling the processing of the stainless steel pipe, N processing nodes are obtained through connecting the stainless steel pipe processing control system, the processing nodes represent different stages or links in the processing process, the links comprise cutting, forming, welding, polishing, heat treatment and the like of materials, each node corresponds to specific processing operation and technological parameters, and each node directly influences the quality and performance of a final product.
And establishing N data detection blocks according to the N processing nodes, acquiring corresponding processing detection data sets under each processing rate according to the N data detection blocks, and generating a historical processing detection sample library.
And establishing N data detection blocks based on the N processing nodes, wherein the data detection blocks are special data collection and analysis areas which are specially set for each processing node and are used for monitoring and recording the running states and processing parameters of each node in real time. Each data detection block is provided with a corresponding sensor and detection device for capturing key data in the processing process, such as processing speed, temperature, pressure, vibration and the like. Along with the continuous processing, corresponding processing detection data sets under each processing rate are obtained from the N data detection blocks, and the processing detection data sets record the running condition and the processing effect of each processing node under different processing rates. Along with the continuous accumulation of the processing detection data set, a historical processing detection sample library is further generated, and data support is provided for defect identification.
And carrying out defect identification on the historical processing detection sample library to obtain N processing defect memory libraries corresponding to the N processing nodes, wherein each processing defect memory library comprises processing defect characteristics corresponding to each processing rate.
After a historical processing detection sample library containing rich processing data is established, defects possibly occurring at different processing rates of different processing nodes are identified by performing defect identification on the data. And obtaining processing defect characteristics (such as cracks, burrs, deformation and the like) of the N processing nodes from the historical processing detection sample library to form corresponding processing defect memory libraries, wherein each processing defect memory library aims at a specific processing node and contains various defect characteristics possibly occurring at different processing rates of the node.
Further, the method further includes obtaining N machining defect memory banks corresponding to the N machining nodes:
The method comprises the steps of analyzing the processing sequence of N processing nodes, identifying the N processing defect memory banks according to the sequence, establishing a similarity analysis network, calling the similarity analysis network to conduct similarity identification on a first processing defect memory bank and a second processing defect memory bank, outputting primary similar defect characteristics, conducting similarity identification on the similarity analysis network according to the primary similar defect characteristics and a third processing defect memory bank, outputting secondary similar defect characteristics, and the like until N-1 similar defect characteristics are output, and taking the secondary similar defect characteristics as characteristic elements for constructing the processing defect memory banks according to the N-1 similar defect characteristics.
N processing nodes correspond to N processing defect memory banks, the processing sequence directly influences the working relation and potential defect transmission mechanism among each node, and the N processing defect memory banks can be identified according to the processing sequence by analyzing the processing sequence of the N processing nodes, so that the processing defect memory banks are ensured to accurately correspond to the corresponding processing nodes. Training a deep learning model based on historical processing defect data to identify potential links and similarities among different processing defect memory banks, wherein the model is set as a similarity analysis network after training and is used for capturing similar defect characteristics of different processing nodes. And (3) invoking a similarity analysis network to perform similarity recognition on the first machining defect memory bank and the second machining defect memory bank, and outputting primary similar defect characteristics, wherein the primary similar defect characteristics represent common or similar defect modes in the two memory banks. And (3) taking the primary similar defect characteristic as input, calling a similarity analysis network to perform similarity recognition with a third processing defect memory library, outputting a secondary similar defect characteristic, and the like until the N-1 times of similar defect characteristics are output, and taking the N-1 times of similar defect characteristics as characteristic elements for constructing the processing defect memory library. The iterative process is actually to construct a hierarchical structure of defect features, and similar defect features of each layer are obtained based on similarity identification of a previous layer and a next processing defect memory bank.
And establishing N probability density functions according to the N processing defect memory banks.
And extracting defect characteristic data corresponding to different processing rates from N processing defect memory libraries, selecting a proper probability density function model, such as Gaussian distribution, poisson distribution, exponential distribution and the like, and establishing N probability density functions.
Further, according to the N machining defect memory banks, establishing N probability density functions, the method includes:
analyzing the defect change relation of the N machining defect memory banks under each machining speed, identifying probability distribution types according to the defect change relation, obtaining N initialization probability distributions, and training the N initialization probability distributions by the N machining defect memory banks to generate the N probability density functions.
Preferably, the defect characteristic data at different processing rates are extracted from N processing defect memory banks, including the number, type and severity of defects, statistical analysis and visualization means (such as scatter diagram, line diagram or box diagram) are utilized to show the variation trend of defects at each processing rate, namely the defect variation relation, according to the distribution situation of the defect data at different processing rates, the probability distribution type most matched with the defect data is identified by comparing common probability distributions (such as normal distribution, poisson distribution, index distribution and the like) so as to identify the probability distribution type most matched with the defect data, if the data shows that the defect rate presents the characteristic of normal distribution along with the change of the rate, normal distribution is used, if the probability of occurrence of defects is exponentially reduced along with the increase of the processing rate, index distribution can be used, if the distribution of the defect rate data is in a bias state, especially when the long tail phenomenon is obvious, the normal distribution can be more suitable, a corresponding probability distribution is initialized for each processing defect memory bank according to the determined probability distribution type, the corresponding initialization probability distribution is respectively carried out on the data in each processing defect memory bank, and each memory function can be obtained through the optimization of each iteration of the probability density through a specific processing function.
And obtaining N real-time processing rates, performing defect probability calculation on the N real-time processing rates by using the N probability density functions, and outputting N defect probabilities.
Obtaining N real-time processing rates, and carrying out defect probability calculation on the N real-time processing rates by utilizing the N probability density functions, namely finding out a corresponding probability density function for each real-time processing rate, substituting the real-time processing rate as an input value into the corresponding probability density function, and calculating the probability that the defect probability exceeds a preset threshold value under the given real-time processing rate to obtain N defect probabilities.
Further, the method includes performing defect probability calculation on the N real-time processing rates with the N probability density functions, and outputting N defect probabilities, the method including:
setting a preset defect probability threshold, converting the preset defect probability threshold to output preset machining rates, inputting the N real-time machining rates into the N probability density functions, and calculating an integral which falls into a position larger than the preset machining rate to obtain output N defect probabilities.
Preferably, a preset defect probability threshold is set for judging whether the machining rate possibly causes an excessively high defect rate, a machining rate range of the preset defect probability threshold is found through probability density functions, namely, for each probability density function, a corresponding cumulative distribution function is found, the preset defect probability threshold is used for finding a corresponding machining rate value, namely, a preset machining rate, on the cumulative distribution function, N real-time machining rates are respectively input into the corresponding N probability density functions, and the integral value of the probability density function when the real-time machining rate is larger than the preset machining rate is calculated, so that the defect probability that N machining nodes exceed the preset machining rate under the respective real-time machining rate is obtained.
Further, if the probability density function of the processing node is subjected to normal distribution, the integral expression of the defect probability is calculated according to the real-time processing rate as follows:
Wherein, the method comprises the steps of, Represents the probability of defect occurrence at the real-time processing rate v,Representing a preset threshold value of probability of defect,A Cumulative Distribution Function (CDF) representing a normal distribution,Represents a preset machining rate corresponding based on a preset defect probability threshold,Represents the mean value of the probability of defects corresponding to the processing rate in the normal distribution,The degree of variation of the defect probability corresponding to the processing rate in the normal distribution is shown.
If the probability density function of the processing node is subjected to normal distribution, calculating an integral expression of defect probability according to the real-time processing rate asWherein, the method comprises the steps of,Represents the probability of defect occurrence at the real-time processing rate v,Representing a preset threshold value of probability of defect,A Cumulative Distribution Function (CDF) representing a normal distribution,Represents a preset machining rate corresponding based on a preset defect probability threshold,Represents the mean value of the probability of defects corresponding to the processing rate in the normal distribution, namely the center point of the normal distribution,The degree of variation (standard deviation) of the defect probability corresponding to the processing rate in the normal distribution is shown. Calculating probability of defect occurrence at real-time processing rate using normally distributed CDF。
And the stainless steel pipe machining control system receives the N defect probabilities and generates an early warning signal for controlling the machining rate of the abnormal machining node.
After the stainless steel tube processing control system receives N defect probabilities, a threshold value of the defect probability can be set based on historical data and quality control requirements, and when the predicted defect probability exceeds the threshold value, early warning is triggered, an early warning signal is generated, and the system is controlled to reduce the processing rate.
Further, the stainless steel pipe machining control system receives the N defect probabilities, and after identifying an abnormal machining node greater than a preset threshold, the method further includes:
The method comprises the steps of obtaining associated processing nodes of the abnormal processing nodes based on the N processing nodes, wherein the associated processing nodes are nodes with processing rate association with the abnormal processing nodes, determining adjustment processing rates of the abnormal processing nodes according to the preset threshold, and synchronously adjusting the processing rates of the associated processing nodes according to the adjustment processing rates.
When the stainless steel pipe machining control system receives the defect probability of N machining nodes, the system immediately analyzes the defect probability of each node, a preset threshold value of the defect probability is set based on factors such as historical data, quality requirements and production standards, the system compares the defect probability of each machining node with the preset threshold value, and when the defect probability of a certain node exceeds the preset threshold value, the node is identified as an abnormal machining node. Once an abnormal processing node is identified, it is necessary to further determine other nodes that have a processing rate correlation with the abnormal node, i.e., associated processing nodes, which are nodes adjacent to the abnormal node on the production line. And determining a proper machining rate according to the preset threshold value and the actual condition of the abnormal machining node, and ensuring that the adjusted rate can meet the production requirement and the product quality. After the adjustment processing rate of the abnormal processing node is determined, the associated processing node with the processing rate relevance with the node is synchronously adjusted, and the aim of synchronous adjustment is to ensure that the processing rate of the whole production line can be kept consistent, so that the operation efficiency and quality of the whole production line are prevented from being influenced due to the rate change of a certain node.
Further, the processing rate synchronization adjustment is performed on the associated processing node according to the adjusted processing rate, and the method comprises the following steps:
The production efficiency influence indexes corresponding to the associated processing nodes are obtained, an objective function is built to minimize the production efficiency influence indexes of the associated processing nodes, and a synchronous adjustment strategy is generated, wherein the synchronous adjustment expression of each associated processing node is as follows: ; indicating the adjusted processing rate of the associated node j, Showing the processing rate of the associated node j after adjustment,An influence coefficient representing the adjustment of the processing rate of the abnormal node i and the processing rate of the associated node j, for quantifying the change in the rate of the associated node j correspondingly adjusted when the rate of the abnormal node i is changed,Indicating the adjusted machining rate of the abnormal node i,And adjusting the associated processing node according to the synchronous adjustment strategy.
Preferably, the production efficiency influence indexes of the relevant processing nodes are obtained from historical data, including but not limited to production cycle, equipment utilization rate, energy consumption, defective rate and the like, an objective function is established aiming at minimizing the sum of the production efficiency influence indexes of the relevant processing nodes so as to find a set of processing speed adjustment values and minimize the production efficiency influence of the whole production line, a synchronous adjustment strategy is generated according to the objective function and the production efficiency influence indexes of the relevant processing nodes, and the synchronous adjustment expression of each relevant processing node is as follows;Indicating the adjusted processing rate of the associated node j,Showing the processing rate of the associated node j after adjustment,An influence coefficient representing the adjustment of the processing rate of the abnormal node i and the processing rate of the associated node j, for quantifying the change in the rate of the associated node j correspondingly adjusted when the rate of the abnormal node i is changed, wherein,In the positive, the processing rate of i increases, the processing rate of j also increases,Negative, i increases the processing rate, j decreases the processing rate,Indicating the adjusted machining rate of the abnormal node i,Generating a synchronous adjustment strategy based on the synchronous adjustment expression, and adjusting the associated processing node according to the synchronous adjustment strategy.
In summary, the embodiment of the application has at least the following technical effects:
Firstly, a stainless steel pipe processing control system is connected to obtain N processing nodes. And then, establishing N data detection blocks according to the N processing nodes, acquiring corresponding processing detection data sets under each processing rate according to the N data detection blocks, and generating a historical processing detection sample library. And then, carrying out defect identification on the historical processing detection sample library to obtain N processing defect memory libraries corresponding to the N processing nodes, wherein each processing defect memory library comprises processing defect characteristics corresponding to each processing rate. And then, according to the N processing defect memory banks, establishing N probability density functions. Further, N real-time processing rates are obtained, defect probability calculation is carried out on the N real-time processing rates according to the N probability density functions, and N defect probabilities are output. And finally, the stainless steel pipe machining control system receives the N defect probabilities and generates an early warning signal for controlling the machining rate of the abnormal machining node. The technical problem that defects appear in the steel pipe due to the change of the machining speed in the prior art is solved, and the technical effect of improving the machining quality is achieved by intelligently controlling the machining speed of abnormal machining nodes.
Example two
Based on the same inventive concept as the stainless steel pipe machining control method based on defect detection in the foregoing embodiments, as shown in fig. 2, the present application provides a stainless steel pipe machining control device based on defect detection, and the device and method embodiments in the embodiments of the present application are based on the same inventive concept. Wherein, the device includes:
The system comprises a node acquisition module 11, a sample library generation module 12, a defect identification module 13, a function creation module 14 and a control module 16, wherein the node acquisition module 11 is used for connecting a stainless steel pipe machining control system to acquire N machining nodes, the sample library generation module 12 is used for creating N data detection blocks according to the N machining nodes, acquiring corresponding machining detection data sets at all machining rates according to the N data detection blocks to generate a historical machining detection sample library, the defect identification module 13 is used for carrying out defect identification on the historical machining detection sample library to acquire N machining defect memory libraries corresponding to the N machining nodes, each machining defect memory library comprises machining defect characteristics corresponding to all machining rates, the function creation module 14 is used for creating N probability density functions according to the N machining defect memory libraries, the calculation module 15 is used for acquiring N real-time machining rates, carrying out defect probability calculation on the N real-time machining rates according to the N probability density functions and outputting N defect probabilities, and the control module 16 is used for controlling the stainless steel pipe machining error probability to receive the N machining defect probability signals and carrying out early warning on the machining nodes.
Further, the control module 16 is configured to perform the following method:
The method comprises the steps of obtaining associated processing nodes of the abnormal processing nodes based on the N processing nodes, wherein the associated processing nodes are nodes with processing rate association with the abnormal processing nodes, determining adjustment processing rates of the abnormal processing nodes according to the preset threshold, and synchronously adjusting the processing rates of the associated processing nodes according to the adjustment processing rates.
Further, the control module 16 is configured to perform the following method:
The production efficiency influence indexes corresponding to the associated processing nodes are obtained, an objective function is built to minimize the production efficiency influence indexes of the associated processing nodes, and a synchronous adjustment strategy is generated, wherein the synchronous adjustment expression of each associated processing node is as follows: ; indicating the adjusted processing rate of the associated node j, Showing the processing rate of the associated node j after adjustment,An influence coefficient representing the adjustment of the processing rate of the abnormal node i and the processing rate of the associated node j, for quantifying the change in the rate of the associated node j correspondingly adjusted when the rate of the abnormal node i is changed,Indicating the adjusted machining rate of the abnormal node i,And adjusting the associated processing node according to the synchronous adjustment strategy.
Further, the function creation module 14 is configured to perform the following method:
analyzing the defect change relation of the N machining defect memory banks under each machining speed, identifying probability distribution types according to the defect change relation, obtaining N initialization probability distributions, and training the N initialization probability distributions by the N machining defect memory banks to generate the N probability density functions.
Further, the computing module 15 is configured to perform the following method:
setting a preset defect probability threshold, converting the preset defect probability threshold to output preset machining rates, inputting the N real-time machining rates into the N probability density functions, and calculating an integral which falls into a position larger than the preset machining rate to obtain output N defect probabilities.
Further, the computing module 15 is configured to perform the following method:
If the probability density function of the processing node is subjected to normal distribution, calculating an integral expression of defect probability according to the real-time processing rate as follows: Wherein, the method comprises the steps of, Represents the probability of defect occurrence at the real-time processing rate v,Representing a preset threshold value of probability of defect,A Cumulative Distribution Function (CDF) representing a normal distribution,Represents a preset machining rate corresponding based on a preset defect probability threshold,Represents the mean value of the probability of defects corresponding to the processing rate in the normal distribution,The degree of variation of the defect probability corresponding to the processing rate in the normal distribution is shown.
Further, the defect identifying module 13 is configured to perform the following method:
The method comprises the steps of analyzing the processing sequence of N processing nodes, identifying the N processing defect memory banks according to the sequence, establishing a similarity analysis network, calling the similarity analysis network to conduct similarity identification on a first processing defect memory bank and a second processing defect memory bank, outputting primary similar defect characteristics, conducting similarity identification on the similarity analysis network according to the primary similar defect characteristics and a third processing defect memory bank, outputting secondary similar defect characteristics, and the like until N-1 similar defect characteristics are output, and taking the secondary similar defect characteristics as characteristic elements for constructing the processing defect memory banks according to the N-1 similar defect characteristics.
It should be noted that the sequence of the embodiments of the present application is only for description, and does not represent the advantages and disadvantages of the embodiments. And the foregoing description has been directed to specific embodiments of this specification. The processes depicted in the accompanying drawings do not necessarily require the particular order shown, or sequential order, to achieve desirable results. In some embodiments, multitasking and parallel processing are also possible or may be advantageous.
The foregoing description of the preferred embodiments of the application is not intended to limit the application to the precise form disclosed, and any such modifications, equivalents, and alternatives falling within the spirit and scope of the application are intended to be included within the scope of the application.
The specification and figures are merely exemplary illustrations of the present application and are considered to cover any and all modifications, variations, combinations, or equivalents that fall within the scope of the application. It will be apparent to those skilled in the art that various modifications and variations can be made to the present application without departing from the scope of the application. Thus, the present application is intended to include such modifications and alterations insofar as they come within the scope of the application or the equivalents thereof.
Claims (4)
1. The stainless steel tube processing control method based on defect detection is characterized by comprising the following steps:
connecting a stainless steel tube processing control system to obtain N processing nodes;
Establishing N data detection blocks according to the N processing nodes, acquiring corresponding processing detection data sets under each processing rate according to the N data detection blocks, and generating a historical processing detection sample library;
Performing defect identification on the historical processing detection sample library to obtain N processing defect memory libraries corresponding to the N processing nodes, wherein each processing defect memory library comprises processing defect characteristics corresponding to each processing rate;
establishing N probability density functions according to the N processing defect memory banks;
Acquiring N real-time processing rates, performing defect probability calculation on the N real-time processing rates by using the N probability density functions, and outputting N defect probabilities;
the stainless steel pipe machining control system receives the N defect probabilities and generates an early warning signal for controlling the machining rate of the abnormal machining node;
according to the N processing defect memory banks, establishing N probability density functions, wherein the method comprises the following steps:
analyzing the defect change relation of the N machining defect memory banks under each machining rate;
Identifying probability distribution types according to the defect change relation, and acquiring N initialization probability distributions;
training the N initialization probability distributions by using the N processing defect memory banks respectively to generate N probability density functions;
The N real-time processing rates are subjected to defect probability calculation according to the N probability density functions, N defect probabilities are output, and the method comprises the following steps:
Setting a preset defect probability threshold;
Converting the preset defect probability threshold value and outputting a preset processing rate;
Inputting the N real-time processing rates into the N probability density functions, and calculating an integral which falls into a processing rate greater than the preset processing rate to obtain N defect probabilities;
if the probability density function of the processing node obeys normal distribution, the integral expression of the defect probability is calculated according to the real-time processing rate as follows:
;
Wherein, Representing the processing rate in real timeThe probability of occurrence of the defect is lower,Representing a preset threshold value of probability of defect,A Cumulative Distribution Function (CDF) representing a normal distribution,Represents a preset machining rate corresponding based on a preset defect probability threshold,Represents the mean value of the probability of defects corresponding to the processing rate in the normal distribution,The defect probability variation degree corresponding to the processing rate in normal distribution is represented;
Wherein N processing defect memory banks corresponding to the N processing nodes are obtained, the method further comprises the steps of:
analyzing the processing sequence of the N processing nodes, and identifying the N processing defect memory banks according to the sequence;
establishing a similarity analysis network, calling the similarity analysis network to perform similarity identification on the first machining defect memory bank and the second machining defect memory bank, and outputting primary similar defect characteristics;
The similarity analysis network carries out similarity identification according to the primary similar defect characteristic and a third processing defect memory library, outputs secondary similar defect characteristics, and so on until N-1 times of similar defect characteristics are output;
And according to the N-1 times similar defect characteristics, the characteristic elements are used for constructing a processing defect memory bank.
2. The method of claim 1, wherein the stainless steel pipe machining control system receives the N defect probabilities, and after identifying an abnormal machining node greater than a preset threshold, the method further comprises:
acquiring associated processing nodes of the abnormal processing nodes based on the N processing nodes, wherein the associated processing nodes are nodes with processing rate association with the abnormal processing nodes;
According to the preset threshold value, determining the adjustment processing rate of the abnormal processing node;
And synchronously adjusting the processing rate of the associated processing nodes according to the adjusted processing rate.
3. The method for controlling machining of a stainless steel pipe based on defect detection according to claim 2, wherein machining rate synchronization adjustment is performed on the associated machining node according to the adjusted machining rate, the method comprising:
acquiring production efficiency influence indexes corresponding to the associated processing nodes;
And establishing an objective function to generate a synchronous adjustment strategy by minimizing the production efficiency influence index sum of the associated processing nodes, wherein the synchronous adjustment expression of each associated processing node is as follows:
;
Representing associated nodes In the case of an adjusted processing rate,Showing associated nodesIn the case of an adjusted processing rate,Representing abnormal nodesProcessing rate adjustment and association node of (a)Influence coefficient of processing rate for quantifying when abnormal nodeCorresponding adjustment of associated nodes when the rate of (a) changesIs used for the rate of change of (a),Representing abnormal nodesIn the case of an adjusted processing rate,Representing abnormal nodesThe processing rate before adjustment;
And adjusting the associated processing nodes according to the synchronous adjustment strategy.
4. A stainless steel pipe machining control device based on defect detection, characterized by being used for implementing the stainless steel pipe machining control method based on defect detection according to any one of claims 1 to 3, the device comprising:
The node acquisition module is used for connecting a stainless steel tube processing control system and acquiring N processing nodes;
The sample library generation module is used for establishing N data detection blocks according to the N processing nodes, acquiring corresponding processing detection data sets at each processing rate according to the N data detection blocks and generating a historical processing detection sample library;
The defect identification module is used for carrying out defect identification on the historical processing detection sample library and obtaining N processing defect memory libraries corresponding to the N processing nodes, wherein each processing defect memory library comprises processing defect characteristics corresponding to each processing rate;
The function building module is used for building N probability density functions according to the N processing defect memory banks;
the calculation module is used for acquiring N real-time processing rates, carrying out defect probability calculation on the N real-time processing rates by using the N probability density functions, and outputting N defect probabilities;
The control module is used for receiving the N defect probabilities by the stainless steel pipe machining control system, generating an early warning signal and controlling the machining rate of the abnormal machining node.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202411096565.4A CN118940043B (en) | 2024-08-12 | 2024-08-12 | Stainless steel pipe processing control method and device based on defect detection |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202411096565.4A CN118940043B (en) | 2024-08-12 | 2024-08-12 | Stainless steel pipe processing control method and device based on defect detection |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN118940043A CN118940043A (en) | 2024-11-12 |
| CN118940043B true CN118940043B (en) | 2025-04-01 |
Family
ID=93349576
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN202411096565.4A Active CN118940043B (en) | 2024-08-12 | 2024-08-12 | Stainless steel pipe processing control method and device based on defect detection |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN118940043B (en) |
Citations (2)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN115240193A (en) * | 2022-07-29 | 2022-10-25 | 山东品正金属制品有限公司 | Surface treatment method and system for motor spindle of electric vehicle |
| CN115983684A (en) * | 2022-12-19 | 2023-04-18 | 常熟中佳新材料有限公司 | Method and system for managing production defects of copper pipe for refrigeration |
Family Cites Families (7)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| US7997139B2 (en) * | 2007-12-03 | 2011-08-16 | Fbs, Inc. | Guided wave pipeline inspection system and method with enhanced natural focusing techniques |
| JP2018116545A (en) * | 2017-01-19 | 2018-07-26 | オムロン株式会社 | Prediction model creating device, production facility monitoring system, and production facility monitoring method |
| CN114972325B (en) * | 2022-07-11 | 2022-11-04 | 爱普车辆股份有限公司 | Automobile hub defect detection method based on image processing |
| CN115690105B (en) * | 2022-12-30 | 2023-03-21 | 无锡康贝电子设备有限公司 | Milling cutter scratch detection method based on computer vision |
| CN117173151B (en) * | 2023-09-25 | 2024-03-08 | 江苏精益智控科技有限公司 | Visual identification device, method and system for defects of outer surface of long steel |
| CN117130265B (en) * | 2023-10-27 | 2024-01-02 | 江苏辰星药业股份有限公司 | Self-adaptive control method and system for health product transportation environment |
| CN118228411A (en) * | 2024-03-27 | 2024-06-21 | 复旦大学 | Method, equipment and product for determining reliability of defect tolerance of complex aviation assembly surfaces |
-
2024
- 2024-08-12 CN CN202411096565.4A patent/CN118940043B/en active Active
Patent Citations (2)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN115240193A (en) * | 2022-07-29 | 2022-10-25 | 山东品正金属制品有限公司 | Surface treatment method and system for motor spindle of electric vehicle |
| CN115983684A (en) * | 2022-12-19 | 2023-04-18 | 常熟中佳新材料有限公司 | Method and system for managing production defects of copper pipe for refrigeration |
Also Published As
| Publication number | Publication date |
|---|---|
| CN118940043A (en) | 2024-11-12 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN112508243B (en) | Training method and device for multi-fault prediction network model of power information system | |
| CN107273924B (en) | Power plant fault diagnosis method based on multi-data fusion based on fuzzy cluster analysis | |
| KR102366787B1 (en) | Real-time sliding window based anomaly detection system for multivariate data generated by manufacturing equipment | |
| CN116582309A (en) | A network intrusion detection method based on GAN-CNN-BiLSTM | |
| CN109469919A (en) | A method for monitoring ash blocking of air preheater in power station based on weighted clustering | |
| CN120009395B (en) | Extrusion casting piece molding quality detection device and system | |
| CN118279665A (en) | Casting defect detection system based on YOLO network | |
| CN116561692B (en) | A dynamic update method for real-time measurement data detection | |
| CN119293490A (en) | Anomaly detection and processing method and system based on deep learning | |
| CN118520389A (en) | Intelligent monitoring method and system for stock yard | |
| CN118940043B (en) | Stainless steel pipe processing control method and device based on defect detection | |
| CN119903454A (en) | A fault diagnosis method, device, equipment and medium for equipment lubrication system | |
| CN117732886B (en) | Hot rolling quality pre-control method based on cascading intelligent diagnosis | |
| CN118585903A (en) | A valve leakage monitoring method and system | |
| CN118626281A (en) | A fast proofreading method and system for error-sensitive word detection | |
| CN116704258A (en) | A Classification Method of Turning Tool Wear Based on Improved MobileNet_V2 Network Model | |
| CN115294382A (en) | A method for multi-damage intelligent identification of structures with limited training samples | |
| CN115905944A (en) | Intelligent labeling method for realizing industrial inspection data based on semantic metadata | |
| CN114943304A (en) | Bayesian-based 3C electronic product assembly error transfer method | |
| CN114579519A (en) | File system abnormality detection method and device, storage medium and electronic device | |
| CN111242215A (en) | Multi-class abnormity detection method based on Token | |
| CN119293491B (en) | An abnormal data monitoring method and system based on artificial intelligence | |
| CN116309521B (en) | A magnetic flux leakage detection method for oil and gas pipelines based on dual attention network | |
| CN118981407B (en) | AI-based facility digital monitoring analysis pre-training system | |
| CN115688050B (en) | A method for detecting time series change points of multi-sensor information based on fusion confidence strategy |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| PB01 | Publication | ||
| PB01 | Publication | ||
| SE01 | Entry into force of request for substantive examination | ||
| SE01 | Entry into force of request for substantive examination | ||
| GR01 | Patent grant | ||
| GR01 | Patent grant | ||
| TR01 | Transfer of patent right | ||
| TR01 | Transfer of patent right |
Effective date of registration: 20250717 Address after: No. 33 Jinding Road, Nantong High tech Industrial Development Zone, Nantong City, Jiangsu Province, 226000 Patentee after: JIANGSU YINYANG STAINLESS STEEL PIPE INDUSTRY Co.,Ltd. Country or region after: China Address before: 226000 Jiangsu Province Nantong City Gaoxin Development Zone Jinding Road 33# Patentee before: Jiangsu Yinjia stainless steel pipe industry Co.,Ltd. Country or region before: China |