CN115983684B - Copper pipe production defect management method and system for refrigeration - Google Patents
Copper pipe production defect management method and system for refrigeration Download PDFInfo
- Publication number
- CN115983684B CN115983684B CN202211630123.4A CN202211630123A CN115983684B CN 115983684 B CN115983684 B CN 115983684B CN 202211630123 A CN202211630123 A CN 202211630123A CN 115983684 B CN115983684 B CN 115983684B
- Authority
- CN
- China
- Prior art keywords
- defect
- copper pipe
- frequency
- sensitive
- production
- 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
- 230000007547 defect Effects 0.000 title claims abstract description 249
- RYGMFSIKBFXOCR-UHFFFAOYSA-N Copper Chemical compound [Cu] RYGMFSIKBFXOCR-UHFFFAOYSA-N 0.000 title claims abstract description 166
- 229910052802 copper Inorganic materials 0.000 title claims abstract description 166
- 239000010949 copper Substances 0.000 title claims abstract description 166
- 238000004519 manufacturing process Methods 0.000 title claims abstract description 165
- 238000007726 management method Methods 0.000 title claims abstract description 53
- 238000005057 refrigeration Methods 0.000 title claims abstract description 48
- 238000000034 method Methods 0.000 claims abstract description 213
- 230000008569 process Effects 0.000 claims abstract description 190
- 238000001514 detection method Methods 0.000 claims abstract description 64
- 238000004458 analytical method Methods 0.000 claims abstract description 42
- 238000002360 preparation method Methods 0.000 claims abstract description 39
- 238000010206 sensitivity analysis Methods 0.000 claims abstract description 12
- 238000012216 screening Methods 0.000 claims description 27
- 238000009826 distribution Methods 0.000 claims description 12
- 238000010276 construction Methods 0.000 claims description 11
- 230000000694 effects Effects 0.000 claims description 6
- 238000007621 cluster analysis Methods 0.000 claims description 5
- 238000002955 isolation Methods 0.000 claims description 4
- 238000013075 data extraction Methods 0.000 claims description 3
- 238000012545 processing Methods 0.000 description 5
- 238000010586 diagram Methods 0.000 description 3
- 238000007689 inspection Methods 0.000 description 3
- 238000012544 monitoring process Methods 0.000 description 3
- 239000003507 refrigerant Substances 0.000 description 3
- 238000004364 calculation method Methods 0.000 description 2
- 238000001125 extrusion Methods 0.000 description 2
- 229910000831 Steel Inorganic materials 0.000 description 1
- 230000002950 deficient Effects 0.000 description 1
- 238000011156 evaluation Methods 0.000 description 1
- 238000002474 experimental method Methods 0.000 description 1
- 238000004880 explosion Methods 0.000 description 1
- 230000007246 mechanism Effects 0.000 description 1
- 238000012986 modification Methods 0.000 description 1
- 230000004048 modification Effects 0.000 description 1
- 230000002265 prevention Effects 0.000 description 1
- 238000012797 qualification Methods 0.000 description 1
- 230000004044 response Effects 0.000 description 1
- 230000035945 sensitivity Effects 0.000 description 1
- 239000010959 steel Substances 0.000 description 1
- 239000002699 waste material Substances 0.000 description 1
Landscapes
- General Factory Administration (AREA)
Abstract
The invention provides a copper pipe production defect management method and system for refrigeration, which relate to the technical field of refrigeration copper pipe production management and control, and are characterized in that product quality detection data are collected for defect analysis to generate a plurality of high-frequency defect characteristics, then a copper pipe production process is generated by a product preparation process, sensitivity analysis is performed on the high-frequency defect characteristics to generate sensitive process parameters, a copper pipe preparation early warning model is constructed for analyzing a real-time copper pipe production process, a sensitive process matching result is output for copper pipe production defect early warning management, the technical problems that the copper pipe production defect management and control method for refrigeration in the prior art is insufficient in intelligence, so that production defect management is complicated, meanwhile, the analysis on production defects is not strict enough, the efficiency is low and the accuracy is insufficient are solved, reverse recurrence is performed based on the product quality detection data, and further sensitive parameter detection analysis is performed by modeling, so that intelligent, accurate and efficient production management is realized.
Description
Technical Field
The invention relates to the technical field of refrigeration copper pipe production management and control, in particular to a refrigeration copper pipe production defect management method and system.
Background
The refrigeration copper pipe is used as a refrigerant pipe, can be applied to the air conditioner and is used as refrigeration equipment of the air conditioner, the refrigerant circulation flow between the inner machine and the outer machine is realized, the production quality requirement on the refrigeration copper pipe is higher, otherwise, the copper pipe is broken due to extrusion deformation of external force, or the pipe burst is caused due to uneven thickness, so as to influence the refrigeration effect of the air conditioner, when the refrigeration copper pipe is produced, the qualification rate of products cannot be guaranteed, so that more defective products cause resource waste, the defect tracing is mainly carried out, the corresponding response mechanism is formulated for management and control, and the management and control capacity is required to be improved.
In the prior art, the intelligent degree of the copper pipe production defect management and control method for refrigeration is insufficient, so that production defect management is complicated, meanwhile, analysis on production defects is not strict enough, and the efficiency is low and the accuracy is insufficient.
Disclosure of Invention
The application provides a copper pipe production defect management method and system for refrigeration, which are used for solving the technical problems that in the prior art, the intelligent degree of the copper pipe production defect management method for refrigeration is insufficient, so that production defect management is complicated, meanwhile, analysis on production defects is not strict enough, and efficiency is low and accuracy is insufficient.
In view of the above problems, the present application provides a method and a system for managing defects in production of copper tubes for refrigeration.
In a first aspect, the present application provides a method for managing defects in production of copper tubes for refrigeration, the method comprising:
traversing the model of the copper pipe for refrigeration, and collecting product quality detection data with preset time granularity;
performing defect analysis on the product quality detection data to generate a plurality of high-frequency defect characteristics;
collecting a product preparation process according to the plurality of high-frequency defect characteristics to generate a copper pipe production process;
performing sensitivity analysis on the copper pipe production process to generate sensitive process parameters;
constructing a copper pipe preparation early warning model according to the sensitive process parameters;
acquiring a real-time copper pipe production process, inputting the copper pipe preparation early warning model, and outputting a sensitive process matching result;
and carrying out early warning management on the production defects of the copper pipe according to the matching result of the sensitive process.
In a second aspect, the present application provides a copper tube production defect management system for refrigeration, the system comprising:
the data acquisition module is used for traversing the type of the copper pipe for refrigeration and acquiring product quality detection data with preset time granularity;
the feature generation module is used for carrying out defect analysis on the product quality detection data to generate a plurality of high-frequency defect features;
the process generation module is used for collecting a product preparation process according to the plurality of high-frequency defect characteristics and generating a copper pipe production process;
the parameter generation module is used for carrying out sensitivity analysis on the copper pipe production process and generating sensitive process parameters;
the model construction module is used for constructing a copper pipe preparation early warning model according to the sensitive process parameters;
the process matching module is used for acquiring a real-time copper pipe production process, inputting the copper pipe preparation early warning model and outputting a sensitive process matching result;
and the production management module is used for carrying out early warning management on the defects of copper pipe production according to the matching result of the sensitive process.
One or more technical solutions provided in the present application have at least the following technical effects or advantages:
according to the copper pipe production defect management method for refrigeration, the model of the copper pipe for refrigeration is traversed, product quality detection data with preset time granularity are collected, defect analysis is conducted on the product quality detection data, and a plurality of high-frequency defect characteristics are generated; collecting a product preparation process according to the plurality of high-frequency defect characteristics to generate a copper pipe production process; performing sensitivity analysis on the copper pipe production process to generate sensitive process parameters; constructing a copper pipe preparation early warning model according to the sensitive process parameters; the method comprises the steps of obtaining a real-time copper pipe production process, inputting a copper pipe preparation early warning model, outputting a sensitive process matching result, carrying out copper pipe production defect early warning management according to the sensitive process matching result, solving the technical problems of insufficient precision, low efficiency and insufficient precision of production defect analysis, carrying out reverse recursion based on product quality inspection data, further modeling to carry out sensitive parameter detection analysis, and realizing intelligent, accurate and efficient production management, wherein the intelligent defect management is complicated.
Drawings
Fig. 1 is a schematic flow chart of a method for managing defects in copper tube production for refrigeration;
FIG. 2 is a schematic diagram of a process for generating sensitive process parameters in a method for managing defects in copper tube production for refrigeration;
fig. 3 is a schematic diagram of a construction flow of a copper pipe preparation early warning model in a copper pipe production defect management method for refrigeration;
fig. 4 is a schematic structural diagram of a copper tube production defect management system for refrigeration.
Reference numerals illustrate: the system comprises a data acquisition module 11, a characteristic generation module 12, a process generation module 13, a parameter generation module 14, a model construction module 15, a process matching module 16 and a production management module 17.
Detailed Description
The application provides a copper pipe production defect management method and system for refrigeration for solve the technical problems that in the prior art, the copper pipe production defect management method for refrigeration is insufficient in intelligence, so that production defect management is complicated, meanwhile, analysis on production defects is not strict enough, and efficiency is low and accuracy is insufficient.
Example 1
As shown in fig. 1, the present application provides a method for managing defects in production of copper tubes for refrigeration, the method comprising:
step S100: traversing the model of the copper pipe for refrigeration, and collecting product quality detection data with preset time granularity;
specifically, the refrigeration copper pipe is used as the refrigeration equipment of the air conditioner, the refrigerant circulation flow between the inner machine and the outer machine is realized, the production quality requirement on the refrigeration copper pipe is higher, otherwise, the copper pipe is broken due to extrusion deformation by external force, or the pipe burst is caused due to uneven thickness, and the like, so that the refrigeration effect of the air conditioner is influenced. The method comprises the steps of obtaining various copper pipe models for refrigeration, wherein each model has specification difference, setting the preset time granularity, namely, the time period for data acquisition, and based on the preset time granularity, respectively carrying out product quality detection on the various copper pipe models for refrigeration based on quality inspection standards, such as product specification parameters, whether plastic deformation and leakage occur in an application process, and the like, obtaining product quality detection data, and providing theoretical basis for subsequent process analysis by taking the product quality detection data as source data of defect analysis.
Step S200: performing defect analysis on the product quality detection data to generate a plurality of high-frequency defect characteristics;
specifically, defect analysis is performed on the collected product quality detection data, the product quality detection data comprises a plurality of pieces of product defect record data, defect cluster analysis is performed on the product quality detection data to determine a plurality of corresponding defect types, a plurality of defect triggering frequencies and a plurality of defect triggering times, defect parameter screening is performed based on the defect types, the plurality of high-frequency defect characteristics are obtained to eliminate accidental defects, and complexity of subsequent defect management is reduced on the basis of maximizing a guaranteed defect management effect.
Further, the step S200 of performing defect analysis on the product quality detection data to generate a plurality of high-frequency defect features further includes:
step S210: extracting a plurality of pieces of product defect record data from the product quality detection data;
step S220: performing defect cluster analysis according to the plurality of product defect record data to generate a plurality of defect types, a plurality of defect trigger frequencies and a plurality of defect trigger times;
step S230: and screening the defect types according to the defect triggering frequencies and the defect triggering times to generate the high-frequency defect characteristics.
Further, the step S230 of the present application further includes screening the plurality of defect types according to the plurality of defect triggering frequencies and the plurality of defect triggering times to generate the plurality of high-frequency defect features:
step S231: constructing a first coordinate axis based on the defect trigger frequency and constructing a second coordinate axis based on the defect trigger time;
step S232: constructing a high-frequency defect screening coordinate system according to the first coordinate axis and the second coordinate axis;
step S233: setting a high-frequency triggering area in the high-frequency defect screening coordinate system;
step S234: inputting the defect triggering frequencies and the defect triggering times into the high-frequency defect screening coordinate system to generate defect type distribution information;
step S235: and adding the plurality of defect types belonging to the high-frequency trigger area into the plurality of high-frequency defect characteristics according to the defect type distribution information.
Specifically, quality detection is performed by traversing the copper pipe model for refrigeration, and product quality detection data is obtained, wherein the product quality detection data comprises a plurality of product defect record data corresponding to the copper pipe model for refrigeration, so that the actual fitting degree of a subsequent analysis result is ensured. And carrying out data extraction and defect clustering on the defect parameters, and determining the defect types, the defect triggering frequencies and the defect triggering time, wherein the more the clustering clusters are, the more accurate the corresponding clustering results are, the defect parameters are associated and correspond, for example, the copper pipe thickness is uneven to cause pipe explosion, the corresponding defect triggering frequencies and the defect triggering time are extracted based on the corresponding recorded data, and the products of different types have differences for the related parameters of the same defect type. And screening the defect types by taking the defect triggering frequencies and the defect triggering times as screening evaluation standards to determine the high-frequency defect characteristics, wherein the high-frequency defect characteristics are required to be subjected to heavy management monitoring in the process of production.
Specifically, the defect triggering frequency corresponds to the defect triggering time, corresponding coordinate axes are respectively determined based on the defect triggering frequency and the defect triggering time, and the first coordinate axis and the second coordinate axis are constructed, wherein the higher the frequency in the first coordinate axis is, the higher the occurrence probability is; and the second coordinate axis takes the interval duration of the trigger time and the current time as a coordinate unit, and the shorter the duration is, the stronger the high-frequency attribute is. And determining a coordinate origin based on the first coordinate axis and the second coordinate axis, and constructing a two-dimensional coordinate system serving as the high-frequency defect screening coordinate system. The high-frequency defect screening coordinate system is used as an auxiliary screening tool, the coordinate area is divided into areas, the high-frequency triggering area is determined, and the coordinate axial threshold values, namely the critical value for area definition, can be respectively determined, the threshold value positioning is carried out in the high-frequency defect screening coordinate system, and the range meeting the coordinate axial threshold value is set as the high-frequency triggering area.
And based on the plurality of defect type distribution coordinates and the plurality of defect triggering times, performing position positioning in the high-frequency defect screening coordinate system, determining corresponding coordinate points, and determining the defect type distribution information based on the arrangement condition of the coordinate points. And taking the set high-frequency trigger area as a coordinate point delineating range, acquiring a plurality of defect types in the high-frequency trigger area in the defect type distribution information, adding the defect types into the plurality of high-frequency defect characteristics, and carrying out defect positioning division by constructing a coordinate system, so that the division efficiency and accuracy can be effectively improved.
Step S300: collecting a product preparation process according to the plurality of high-frequency defect characteristics to generate a copper pipe production process;
further, the step S300 of generating a copper pipe production process according to the preparation process of the plurality of high-frequency defect feature collection products further includes:
step S310: constructing a process classification constraint array according to the plurality of high-frequency defect characteristics, wherein the process classification constraint array comprises a plurality of process classification constraint conditions, and any one of the plurality of process classification constraint conditions comprises one or more high-frequency defect characteristics;
step S320: traversing the process classification constraint conditions, and collecting the copper pipe production process based on an industrial internet.
Specifically, the product quality detection data is analyzed and screened to obtain the plurality of high-frequency defect characteristics, and the plurality of high-frequency defect characteristics are used as screening conditions to perform production process matching. And determining a plurality of process classification constraints based on the plurality of high-frequency defect features, wherein any one of the plurality of process classification constraints comprises one or more high-frequency defect features, which may be caused by a plurality of process parameters, and generating the process classification constraint array. And further performing process matching based on the industrial Internet, wherein the process classification constraint array is used for performing process classification limiting, ensuring the suitability of the matched production process and high-frequency defect characteristics, determining the copper pipe production process according to the process classification constraint array, and performing parameter analysis and early warning based on the obtained copper pipe production process.
Step S400: performing sensitivity analysis on the copper pipe production process to generate sensitive process parameters;
specifically, the copper pipe production process is determined by performing process matching, a plurality of process nodes are determined by performing process disassembly, a single variable or a plurality of related variables, namely concurrent variables caused by the single variable, are determined based on each node, the influence degree of process parameters corresponding to the nodes on the finished product state is determined by performing variable analysis, further influence degree threshold judgment is performed, when the threshold is met, the corresponding process parameters are judged as sensitive process parameters, variable adjustment analysis is performed on the copper pipe production process for a plurality of times, all the process parameters meeting the influence degree threshold are added into the sensitive process parameters by performing influence degree judgment, the sensitive process parameters are production parameters to be subjected to important monitoring management, and the acquisition of the sensitive process parameters is the basis for subsequent copper pipe preparation pre-warning and tamping.
Further, as shown in fig. 2, the sensitivity analysis is performed on the copper pipe production process to generate sensitive process parameters, and step S400 of the present application further includes:
step S410: process node disassembly is carried out on the copper pipe production process, and a copper pipe production process node network is generated;
step S420: matching a production process parameter group sequence with copper pipe finished product state information according to the copper pipe production process node network;
step S430: performing production comparison analysis according to the copper pipe production process node network, the production process parameter group sequence and the copper pipe finished product state information to generate a plurality of process parameter comparison influence values;
step S440: and adding the process parameters, of which the control influence degree of the process parameters meets the control influence degree threshold, into the sensitive process parameters.
Further, the step S430 of the present application further includes:
step S431: setting a first variable group according to the copper pipe production process node network; setting a second variable group according to the production process parameter group sequence;
step S432: randomly setting a first type variable according to the first variable group, wherein the first type variable comprises a process node or a plurality of first type variables;
step S433: inputting the first type variable into the second variable group, and randomly setting a second type variable, wherein the second type variable comprises one technological parameter or a plurality of technological parameters;
step S434: according to the second type variable, carrying out production comparison analysis on the copper pipe finished product state information to obtain defect quantity influence and defect grade influence;
step S435: and fusing the defect quantity influence degree and the defect grade influence degree according to preset weight distribution to generate first process parameter comparison influence degree, wherein the first process parameter comparison influence degree belongs to the plurality of process parameter comparison influence degrees.
Specifically, the copper pipe production process is acquired by collecting a product preparation process, the required equipment corresponding to each process step is determined by carrying out process disassembly, each production equipment is used as a processing node, the processing nodes are connected based on the process steps, and the copper pipe production process node network is generated. And carrying out production information matching on each node in the copper pipe production process node network, determining the characteristic value of the production parameter corresponding to corresponding production equipment, corresponding to the process nodes one by one, and using the characteristic value as the production process parameter group sequence, and simultaneously determining the corresponding processing state of the copper pipe after processing is finished and using the corresponding processing state as the copper pipe finished product state information, wherein any one of the characteristic value sequences of the production parameters corresponds to one finished product state, and the quantity and the defect grade of defects can be represented.
Specifically, a plurality of process nodes are determined based on the copper pipe production process node network and used as the first variable group, and one process node or a plurality of process nodes are randomly set as the first type variable based on the first variable group and used for carrying out production defect analysis of single process nodes or multiple process node combinations; and determining node process parameters to generate the second variable group based on the production process parameter group sequence, inputting the first type variable into the second variable group, determining one process parameter or a plurality of process parameters corresponding to the first type variable, and setting the first type variable as the second type variable. And carrying out production control analysis on the finished product state of the copper pipe based on the second type of variables, and limiting a single variable or a single group of variables by carrying out control experiments so as to analyze the influence degree of different process parameters on the finished product state. And determining the influence degree of the second type of variables on the defect quantity and the defect grade of the finished state of the copper pipe based on the finished state deviation, wherein the influence degree is proportional to the finished state deviation, and the influence grade can be set to express the influence degree.
Further obtaining a preset weight, namely the set influence degree weight of the defect quantity and the defect grade, carrying out weighted calculation on the influence degree of the defect quantity and the influence degree of the defect grade based on the preset weight, taking a calculation result as the first process parameter comparison influence degree, adjusting variables for a plurality of times, carrying out comparison influence analysis based on the analysis steps, generating corresponding process parameter comparison influence degree, and further adding the corresponding process parameter comparison influence degree into the process parameter comparison influence degrees to determine the state influence of different process parameters on a copper pipe finished product.
Setting the influence threshold, namely, carrying out influence threshold of related parameter sensitivity analysis, judging the influence degree of the process parameters based on the influence threshold, adding the process parameters which are larger than or equal to the influence threshold into the sensitive process parameters, namely, carrying out high-precision screening of the sensitive process parameters to carry out targeted monitoring management of steel pipe production, and avoiding invalid work.
Step S500: constructing a copper pipe preparation early warning model according to the sensitive process parameters;
further, as shown in fig. 3, according to the sensitive process parameters, the method constructs a copper pipe preparation early warning model, and step S500 of the present application further includes:
step S510: dividing the sensitive process parameters into a first type sensitive process parameter and a second type sensitive process parameter and up to an N type sensitive process parameter, wherein the type of any one type of sensitive process parameter is defined as 1, and the characteristic value data quantity of any one sensitive process parameter in any one type is defined as 2;
step S520: constructing a first isolated detection tree according to the first type sensitive process parameters;
step S530: constructing an N-th isolated detection tree according to the N-th type sensitive process parameters;
step S540: and setting the first isolation detection tree to the Nth isolation detection tree as parallel nodes, and generating the copper pipe preparation early warning model.
Specifically, the sensitive process parameters are obtained through carrying out sensitive analysis on copper pipe production process, the sensitive process parameters are divided, preferably, process nodes can be used as dividing standards of the sensitive process parameters, the first type sensitive process parameters and the second type sensitive process parameters are obtained until the Nth type sensitive process parameters, the sensitive process parameters are defined as 1 based on the first type sensitive process parameters, the characteristic value data quantity of any one sensitive process parameter is at least 2, such as a temperature control parameter and a pressure control parameter, two groups of identical characteristic values are set, the first isolated detection tree is constructed, when normal process parameters enter, the first isolated detection tree cannot be divided into any sensitive process parameters, the first isolated detection tree is judged to be finally differentiated into new nodes, the data quantity of the differentiated nodes is 1, and the process parameters are regarded as normal; if the node with the data quantity of 1 cannot be differentiated, namely the input process parameter and the characteristic value of the sensitive process parameter belong to the same class, and the current data quantity is 3, the process parameter is considered to be a parameter which is easier to cause production defects, and early warning management is needed.
Similarly, a second isolated detection tree is constructed based on the second type sensitive process parameters until the construction of the N-th isolated detection tree is completed, and the first isolated detection tree to the N-th isolated detection tree are set as parallel nodes, namely peer nodes, so that the copper pipe preparation early warning model is formed. A plurality of isolated detection trees are embedded in the copper pipe preparation early warning model so as to carry out targeted detection of input data, and the data detection efficiency and accuracy can be improved to a certain extent.
Step S600: acquiring a real-time copper pipe production process, inputting the copper pipe preparation early warning model, and outputting a sensitive process matching result;
step S700: and carrying out early warning management on the production defects of the copper pipe according to the matching result of the sensitive process.
Specifically, a real-time copper pipe production process, namely a current process for product production, is obtained, process parameters of all process nodes are determined and input into the copper pipe preparation early warning model, sensitivity detection of corresponding process parameters is carried out based on a matching result by matching input data with an isolated detection tree so as to carry out efficient and accurate detection of the process parameters, sensitive process parameters in the input data are determined and are used as a model output of the sensitive process matching result, the real-time copper pipe production process is further identified based on the sensitive process matching result so as to carry out early warning and warning of the process production process, and high-quality production of the copper pipe is realized by carrying out defect prevention of copper pipe production.
The copper pipe production defect management method for refrigeration provided by the embodiment of the application has the following technical effects:
1. the method comprises the steps of collecting product quality detection data for defect analysis, generating a plurality of high-frequency defect characteristics, further collecting a product preparation process to generate a copper pipe production process, carrying out sensitivity analysis on the copper pipe production process to generate sensitive process parameters, constructing a copper pipe preparation early warning model for analyzing a real-time copper pipe production process, outputting a sensitive process matching result, carrying out copper pipe production defect early warning management, solving the technical problems of insufficient intelligence, complicated production defect management, insufficient precision, low efficiency and insufficient accuracy of analysis of production defects in the prior art, carrying out reverse recursion based on product quality detection data, further modeling to carry out sensitive parameter detection analysis, and realizing intelligent, accurate and efficient production management.
2. Based on product quality inspection data, a high-frequency defect screening coordinate system is constructed to divide defect characteristics, the demand suitability of screening defects is guaranteed, a single variable or a single group of variables are further limited, sensitive process parameters are determined by analyzing the influence degree of different process parameters on the state of a finished product, the accuracy of the parameters is guaranteed, a copper pipe preparation early warning model is constructed to conduct real-time production process analysis, and the objectivity and actual production compliance of analysis results are guaranteed, so that production defect management is conducted on the basis.
Example two
Based on the same inventive concept as the copper tube production defect management method for refrigeration in the foregoing embodiment, as shown in fig. 4, the present application provides a copper tube production defect management system for refrigeration, the system comprising:
the data acquisition module 11 is used for traversing the type of the copper pipe for refrigeration and acquiring product quality detection data with preset time granularity;
a feature generation module 12, wherein the feature generation module 12 is configured to perform defect analysis on the product quality detection data to generate a plurality of high-frequency defect features;
the process generation module 13 is used for collecting a product preparation process according to the plurality of high-frequency defect characteristics and generating a copper pipe production process;
the parameter generation module 14 is used for performing sensitivity analysis on the copper pipe production process to generate sensitive process parameters;
the model construction module 15 is used for constructing a copper pipe preparation early warning model according to the sensitive process parameters;
the process matching module 16 is used for acquiring a real-time copper pipe production process, inputting the copper pipe preparation early warning model and outputting a sensitive process matching result;
and the production management module 17 is used for carrying out early warning management on the production defects of the copper pipe according to the matching result of the sensitive process by the production management module 17.
Further, the system further comprises:
the data extraction module is used for extracting a plurality of pieces of product defect record data from the product quality detection data;
the defect parameter generation module is used for carrying out defect cluster analysis according to the plurality of product defect record data to generate a plurality of defect types, a plurality of defect trigger frequencies and a plurality of defect trigger times;
and the defect feature generation module is used for screening the defect types according to the defect trigger frequencies and the defect trigger times to generate the high-frequency defect features.
Further, the system further comprises:
the coordinate axis construction module is used for constructing a first coordinate axis based on the defect trigger frequency and constructing a second coordinate axis based on the defect trigger time;
the coordinate system construction module is used for constructing a high-frequency defect screening coordinate system according to the first coordinate axis and the second coordinate axis;
the region setting module is used for setting a high-frequency trigger region in the high-frequency defect screening coordinate system;
inputting the defect triggering frequencies and the defect triggering times into the high-frequency defect screening coordinate system to generate defect type distribution information;
and the defect type adding module is used for adding the plurality of defect types belonging to the high-frequency trigger area into the plurality of high-frequency defect characteristics according to the defect type distribution information.
Further, the system further comprises:
the array construction module is used for constructing a process classification constraint array according to the plurality of high-frequency defect characteristics, wherein the process classification constraint array comprises a plurality of process classification constraints, and any one of the plurality of process classification constraints comprises one or more high-frequency defect characteristics;
the production process acquisition module is used for traversing the process classification constraint conditions and acquiring the copper pipe production process based on an industrial internet.
Further, the system further comprises:
the node network generation module is used for carrying out process node disassembly on the copper pipe production process to generate a copper pipe production process node network;
the information matching module is used for matching the production process parameter group sequence and the copper pipe finished product state information according to the copper pipe production process node network;
the influence degree generation module is used for carrying out production comparison analysis according to the copper pipe production process node network, the production process parameter group sequence and the copper pipe finished product state information to generate a plurality of process parameter comparison influence degrees;
and the process parameter adding module is used for adding the process parameters, of which the process parameter comparison influence degree meets the comparison influence degree threshold, into the sensitive process parameters.
Further, the system further comprises:
the variable group setting module is used for setting a first variable group according to the copper pipe production process node network; setting a second variable group according to the production process parameter group sequence;
the first type variable setting module is used for randomly setting a first type variable according to the first variable group, wherein the first type variable comprises a process node or a plurality of first type variables;
the second type variable setting module is used for inputting the first type variable into the second variable group and randomly setting a second type variable, wherein the second type variable comprises one technological parameter or a plurality of technological parameters;
the defect influence degree acquisition module is used for carrying out production comparison analysis on the copper pipe finished product state information according to the second type variable to acquire defect quantity influence degree and defect grade influence degree;
the comparison influence degree generation module is used for fusing the defect quantity influence degree and the defect grade influence degree according to preset weight distribution to generate first process parameter comparison influence degree, and the first process parameter comparison influence degree belongs to the plurality of process parameter comparison influence degrees.
Further, the system further comprises:
the parameter dividing module is used for dividing the sensitive process parameters into a first type sensitive process parameter and a second type sensitive process parameter until an N type sensitive process parameter, wherein the type of any one type of sensitive process parameter is defined as 1, and the characteristic value data quantity of any one sensitive process parameter in any one type is defined as 2;
the first isolated detection tree construction module is used for constructing a first isolated detection tree according to the first type sensitive process parameters;
the N-th isolated detection tree construction module is used for constructing an N-th isolated detection tree according to the N-th type sensitive process parameters;
the model generation module is used for setting the first isolated detection tree to the N isolated detection tree as parallel nodes and generating the copper pipe preparation early warning model.
In the foregoing description of a method for managing defects in production of copper tubes for refrigeration, those skilled in the art will clearly understand that a method and a system for managing defects in production of copper tubes for refrigeration in this embodiment are relatively simple in description, and the relevant points refer to the description of the method section for the device disclosed in the embodiment, since the device corresponds to the method disclosed in the embodiment.
The previous description of the disclosed embodiments is provided to enable any person skilled in the art to make or use the present application. Various modifications to these embodiments will be readily apparent to those skilled in the art, and the generic principles defined herein may be applied to other embodiments without departing from the spirit or scope of the application. Thus, the present application is not intended to be limited to the embodiments shown herein but is to be accorded the widest scope consistent with the principles and novel features disclosed herein.
Claims (6)
1. The production defect management method of the copper pipe for refrigeration is characterized by comprising the following steps of:
traversing the model of the copper pipe for refrigeration, and collecting product quality detection data with preset time granularity;
performing defect analysis on the product quality detection data to generate a plurality of high-frequency defect characteristics;
collecting a product preparation process according to the plurality of high-frequency defect characteristics to generate a copper pipe production process;
performing sensitivity analysis on the copper pipe production process to generate sensitive process parameters;
constructing a copper pipe preparation early warning model according to the sensitive process parameters;
acquiring a real-time copper pipe production process, inputting the copper pipe preparation early warning model, and outputting a sensitive process matching result;
performing copper pipe production defect early warning management according to the sensitive process matching result;
performing defect analysis on the product quality detection data to generate a plurality of high-frequency defect characteristics, including:
extracting a plurality of pieces of product defect record data from the product quality detection data;
performing defect cluster analysis according to the plurality of product defect record data to generate a plurality of defect types, a plurality of defect trigger frequencies and a plurality of defect trigger times;
screening the defect types according to the defect triggering frequencies and the defect triggering times to generate the high-frequency defect characteristics;
the sensitivity analysis is carried out on the copper pipe production process to generate sensitive process parameters, which comprises the following steps:
process node disassembly is carried out on the copper pipe production process, and a copper pipe production process node network is generated;
matching a production process parameter group sequence with copper pipe finished product state information according to the copper pipe production process node network;
performing production comparison analysis according to the copper pipe production process node network, the production process parameter group sequence and the copper pipe finished product state information to generate a plurality of process parameter comparison influence values;
and adding the process parameters, of which the control influence degree of the process parameters meets the control influence degree threshold, into the sensitive process parameters.
2. The method of claim 1, wherein the screening the plurality of defect types based on the plurality of defect trigger frequencies and the plurality of defect trigger times to generate the plurality of high frequency defect features comprises:
constructing a first coordinate axis based on the defect trigger frequency and constructing a second coordinate axis based on the defect trigger time;
constructing a high-frequency defect screening coordinate system according to the first coordinate axis and the second coordinate axis;
setting a high-frequency triggering area in the high-frequency defect screening coordinate system;
inputting the defect triggering frequencies and the defect triggering times into the high-frequency defect screening coordinate system to generate defect type distribution information;
and adding the plurality of defect types belonging to the high-frequency trigger area into the plurality of high-frequency defect characteristics according to the defect type distribution information.
3. The method of claim 1, wherein the collecting the product preparation process based on the plurality of high frequency defect characteristics generates a copper tube production process comprising:
constructing a process classification constraint array according to the plurality of high-frequency defect characteristics, wherein the process classification constraint array comprises a plurality of process classification constraint conditions, and any one of the plurality of process classification constraint conditions comprises one or more high-frequency defect characteristics;
traversing the process classification constraint conditions, and collecting the copper pipe production process based on an industrial internet.
4. The method of claim 1, wherein said performing a production control analysis based on said network of copper tube production process nodes, said sequence of production process parameter sets, and said copper tube finish state information to generate a plurality of process parameter control effects comprises:
setting a first variable group according to the copper pipe production process node network; setting a second variable group according to the production process parameter group sequence;
randomly setting a first type variable according to the first variable group, wherein the first type variable comprises one process node or a plurality of process nodes;
inputting the first type variable into the second variable group, determining one technological parameter or a plurality of technological parameters corresponding to the first type variable, and setting the first type variable as a second type variable, wherein the second type variable comprises one technological parameter or a plurality of technological parameters;
according to the second type variable, carrying out production comparison analysis on the copper pipe finished product state information to obtain defect quantity influence and defect grade influence;
and fusing the defect quantity influence degree and the defect grade influence degree according to preset weight distribution to generate first process parameter comparison influence degree, wherein the first process parameter comparison influence degree belongs to the plurality of process parameter comparison influence degrees.
5. The method of claim 1, wherein constructing a copper tube preparation pre-warning model based on the sensitive process parameters comprises:
dividing the sensitive process parameters into a first type sensitive process parameter and a second type sensitive process parameter and up to an N type sensitive process parameter, wherein the type of any one type of sensitive process parameter is defined as 1, and the characteristic value data quantity of any one sensitive process parameter in any one type is defined as 2;
constructing a first isolated detection tree according to the first type sensitive process parameters;
constructing an N-th isolated detection tree according to the N-th type sensitive process parameters;
and setting the first isolation detection tree to the Nth isolation detection tree as parallel nodes, and generating the copper pipe preparation early warning model.
6. A copper tube production defect management system for refrigeration, the system comprising:
the data acquisition module is used for traversing the type of the copper pipe for refrigeration and acquiring product quality detection data with preset time granularity;
the feature generation module is used for carrying out defect analysis on the product quality detection data to generate a plurality of high-frequency defect features;
the process generation module is used for collecting a product preparation process according to the plurality of high-frequency defect characteristics and generating a copper pipe production process;
the parameter generation module is used for carrying out sensitivity analysis on the copper pipe production process and generating sensitive process parameters;
the model construction module is used for constructing a copper pipe preparation early warning model according to the sensitive process parameters;
the process matching module is used for acquiring a real-time copper pipe production process, inputting the copper pipe preparation early warning model and outputting a sensitive process matching result;
the production management module is used for carrying out early warning management on copper pipe production defects according to the sensitive process matching result;
the data extraction module is used for extracting a plurality of pieces of product defect record data from the product quality detection data;
the defect parameter generation module is used for carrying out defect cluster analysis according to the plurality of product defect record data to generate a plurality of defect types, a plurality of defect trigger frequencies and a plurality of defect trigger times;
the defect feature generation module is used for screening the defect types according to the defect trigger frequencies and the defect trigger times to generate the high-frequency defect features;
the node network generation module is used for carrying out process node disassembly on the copper pipe production process to generate a copper pipe production process node network;
the information matching module is used for matching the production process parameter group sequence and the copper pipe finished product state information according to the copper pipe production process node network;
the influence degree generation module is used for carrying out production comparison analysis according to the copper pipe production process node network, the production process parameter group sequence and the copper pipe finished product state information to generate a plurality of process parameter comparison influence degrees;
and the process parameter adding module is used for adding the process parameters, of which the process parameter comparison influence degree meets the comparison influence degree threshold, into the sensitive process parameters.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202211630123.4A CN115983684B (en) | 2022-12-19 | 2022-12-19 | Copper pipe production defect management method and system for refrigeration |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202211630123.4A CN115983684B (en) | 2022-12-19 | 2022-12-19 | Copper pipe production defect management method and system for refrigeration |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN115983684A CN115983684A (en) | 2023-04-18 |
| CN115983684B true CN115983684B (en) | 2024-01-26 |
Family
ID=85964125
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN202211630123.4A Active CN115983684B (en) | 2022-12-19 | 2022-12-19 | Copper pipe production defect management method and system for refrigeration |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN115983684B (en) |
Citations (7)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| KR20190032908A (en) * | 2017-09-20 | 2019-03-28 | 주식회사 에이치엔에스휴먼시스템 | Method for managing steel quality and system |
| KR102100499B1 (en) * | 2019-10-11 | 2020-04-13 | 강현광 | Quality management system using big data |
| CN111597729A (en) * | 2020-05-27 | 2020-08-28 | 北京天泽智云科技有限公司 | Method and system for optimizing technological parameters of processing equipment |
| CN112557406A (en) * | 2021-02-19 | 2021-03-26 | 浙江大胜达包装股份有限公司 | Intelligent inspection method and system for paper product production quality |
| CN113176281A (en) * | 2021-04-25 | 2021-07-27 | 程志伟 | System and method for detecting surface defects based on artificial intelligence drawing line |
| CN113723781A (en) * | 2021-08-19 | 2021-11-30 | 武汉慧远智控科技有限公司 | Product quality defect judgment system and method based on SPC analysis |
| CN115047162A (en) * | 2022-06-24 | 2022-09-13 | 张家港沙龙精密管业有限公司 | Defect detection method and system for steel pipe heat treatment |
-
2022
- 2022-12-19 CN CN202211630123.4A patent/CN115983684B/en active Active
Patent Citations (7)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| KR20190032908A (en) * | 2017-09-20 | 2019-03-28 | 주식회사 에이치엔에스휴먼시스템 | Method for managing steel quality and system |
| KR102100499B1 (en) * | 2019-10-11 | 2020-04-13 | 강현광 | Quality management system using big data |
| CN111597729A (en) * | 2020-05-27 | 2020-08-28 | 北京天泽智云科技有限公司 | Method and system for optimizing technological parameters of processing equipment |
| CN112557406A (en) * | 2021-02-19 | 2021-03-26 | 浙江大胜达包装股份有限公司 | Intelligent inspection method and system for paper product production quality |
| CN113176281A (en) * | 2021-04-25 | 2021-07-27 | 程志伟 | System and method for detecting surface defects based on artificial intelligence drawing line |
| CN113723781A (en) * | 2021-08-19 | 2021-11-30 | 武汉慧远智控科技有限公司 | Product quality defect judgment system and method based on SPC analysis |
| CN115047162A (en) * | 2022-06-24 | 2022-09-13 | 张家港沙龙精密管业有限公司 | Defect detection method and system for steel pipe heat treatment |
Non-Patent Citations (4)
| Title |
|---|
| Learning With Imbalanced Data in Smart Manufacturing: A Comparative Analysis;Yasmin Fathy等;IEEE Access;第9卷;2734-2757 * |
| 低温容器钢16MnDR内部点状缺陷分析与改进;庞洪轩等;天津冶金(第03期);57-59+67 * |
| 热轧工艺过程和质量管控平台的研发和应用;邵健等;轧钢;第35卷(第02期);6-11 * |
| 铜管三辊行星轧制过程轧辊轴力学行为分析;董营等;热加工工艺;第45卷(第05期);111-115 * |
Also Published As
| Publication number | Publication date |
|---|---|
| CN115983684A (en) | 2023-04-18 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| JP6817426B2 (en) | Yield prediction system and method for machine learning-based semiconductor manufacturing | |
| CN107273924B (en) | Multi-data fusion power plant fault diagnosis method based on fuzzy clustering analysis | |
| CN109634233A (en) | Industrial big data intellectual analysis decision-making technique, readable storage medium storing program for executing and terminal | |
| CN108491931B (en) | Method for improving nondestructive testing precision based on machine learning | |
| CN115601000B (en) | BIM5D model engineering construction cost method and system under complex environment | |
| CN116224930B (en) | Processing control method and system for numerically controlled grinder product | |
| CN109614451A (en) | Industrial big data intellectual analysis decision making device | |
| CN111130890A (en) | Network flow dynamic prediction system | |
| Ungermann et al. | Data analytics for manufacturing systems–a data-driven approach for process optimization | |
| CN116468160A (en) | Aluminum alloy die casting quality prediction method based on production big data | |
| CN116822353A (en) | Digital twin model rapid construction method in manufacturing process | |
| CN109345060A (en) | A kind of product quality characteristics Error Tracing & analysis method based on multi-source perception | |
| CN115983684B (en) | Copper pipe production defect management method and system for refrigeration | |
| CN108563201B (en) | DMAIC-driven small-batch part machining process quality improvement method | |
| CN111126477A (en) | Learning and reasoning method of hybrid Bayesian network | |
| Eger et al. | Knowledge capturing platform in multi-stage production systems for zero-defect manufacturing | |
| CN115106615B (en) | Welding deviation real-time detection method and system based on intelligent working condition identification | |
| CN116227896A (en) | Silicon carbide production process management method and system | |
| CN112884348B (en) | Method for diagnosing production deviation source of space detonator based on dynamic Bayesian network | |
| RU2622857C1 (en) | System of transformation, analysis and evaluation of information objects of the object | |
| CN115221796A (en) | Cloud manufacturing mode-oriented process flow simulation method and system | |
| CN115115213A (en) | Method and device for retrospective analysis of quality characteristics of steel products | |
| CN113312587A (en) | Sensor acquisition data missing value processing method based on ARIMA prediction and regression prediction | |
| Khanzadeh et al. | Profiling and optimizing the geometric accuracy of additively manufactured components via self-organizing map | |
| Su et al. | Characterisation and evaluation of identicality for digital twins for the manufacturing domain |
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 |