CN118067203B - Fiber winding quality monitoring method, equipment and medium for high-pressure gas cylinder - Google Patents
Fiber winding quality monitoring method, equipment and medium for high-pressure gas cylinder Download PDFInfo
- Publication number
- CN118067203B CN118067203B CN202410479484.6A CN202410479484A CN118067203B CN 118067203 B CN118067203 B CN 118067203B CN 202410479484 A CN202410479484 A CN 202410479484A CN 118067203 B CN118067203 B CN 118067203B
- Authority
- CN
- China
- Prior art keywords
- winding
- parameter
- carbon fiber
- defect
- preset
- 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
- 238000004804 winding Methods 0.000 title claims abstract description 632
- 238000012544 monitoring process Methods 0.000 title claims abstract description 106
- 239000000835 fiber Substances 0.000 title claims abstract description 87
- 238000000034 method Methods 0.000 title claims abstract description 50
- 229920000049 Carbon (fiber) Polymers 0.000 claims abstract description 306
- 239000004917 carbon fiber Substances 0.000 claims abstract description 306
- VNWKTOKETHGBQD-UHFFFAOYSA-N methane Chemical compound C VNWKTOKETHGBQD-UHFFFAOYSA-N 0.000 claims abstract description 304
- 230000007547 defect Effects 0.000 claims abstract description 206
- 238000012937 correction Methods 0.000 claims description 85
- 238000009730 filament winding Methods 0.000 claims description 14
- 238000004458 analytical method Methods 0.000 claims description 11
- 239000007789 gas Substances 0.000 description 35
- 230000008569 process Effects 0.000 description 10
- 238000004519 manufacturing process Methods 0.000 description 9
- 230000008859 change Effects 0.000 description 8
- 238000012806 monitoring device Methods 0.000 description 7
- 238000012549 training Methods 0.000 description 5
- 230000009286 beneficial effect Effects 0.000 description 3
- 238000004364 calculation method Methods 0.000 description 3
- 230000000694 effects Effects 0.000 description 3
- 238000012986 modification Methods 0.000 description 2
- 230000004048 modification Effects 0.000 description 2
- UFHFLCQGNIYNRP-UHFFFAOYSA-N Hydrogen Chemical compound [H][H] UFHFLCQGNIYNRP-UHFFFAOYSA-N 0.000 description 1
- 229910000831 Steel Inorganic materials 0.000 description 1
- 230000007797 corrosion Effects 0.000 description 1
- 238000005260 corrosion Methods 0.000 description 1
- 238000011161 development Methods 0.000 description 1
- 238000010586 diagram Methods 0.000 description 1
- 239000000284 extract Substances 0.000 description 1
- 230000004438 eyesight Effects 0.000 description 1
- 229910052739 hydrogen Inorganic materials 0.000 description 1
- 239000001257 hydrogen Substances 0.000 description 1
- 230000006872 improvement Effects 0.000 description 1
- 238000011031 large-scale manufacturing process Methods 0.000 description 1
- 238000003062 neural network model Methods 0.000 description 1
- 238000012545 processing Methods 0.000 description 1
- 230000000750 progressive effect Effects 0.000 description 1
- 238000011160 research Methods 0.000 description 1
- 238000012216 screening Methods 0.000 description 1
- 239000010959 steel Substances 0.000 description 1
- 230000000007 visual effect Effects 0.000 description 1
Landscapes
- Moulding By Coating Moulds (AREA)
Abstract
The application provides a fiber winding quality monitoring method, equipment and medium for a high-pressure gas cylinder, and belongs to the technical field of high-pressure gas cylinder winding monitoring. The problem that the winding quality monitoring of the fiber of the existing high-pressure gas cylinder is difficult to accurately and efficiently carry out due to the fact that labor cost is excessively consumed in the winding quality monitoring of the fiber of the existing high-pressure gas cylinder is solved. The method can determine a carbon fiber winding parameter group, and performs parameter value matching with a winding defect comparison parameter list to determine whether a winding defect exists at the current moment; if the defect parameter exists, determining a defect parameter regular interval set based on the winding defect parameter value, the carbon fiber winding parameter sequence of the last preset period and the DBSCAN model. Then determining whether to correct the winding control parameter sequence of the digital winding machine by utilizing the carbon fiber winding parameter set and the defect parameter regular interval set; if yes, correcting the winding control parameter sequence, controlling the operation of the numerical control winding machine according to the corrected winding control parameter sequence, and continuously monitoring and early warning the carbon fiber winding parameter group.
Description
Technical Field
The application relates to the technical field of high-pressure gas cylinder winding monitoring, in particular to a fiber winding quality monitoring method, equipment and medium for a high-pressure gas cylinder.
Background
High pressure hydrogen cylinders such as type III and type IV cylinders are used for storing and transporting compressed gas and are formed by winding fibers such as carbon fibers outside the liner. Compared with the traditional steel gas cylinder, the fiber winding gas cylinder has the advantages of light weight, high strength, corrosion resistance, fatigue resistance and the like, and is widely applied to the fields of industry, medical treatment, scientific research and the like.
At present, when carbon fiber winding is carried out on a gas cylinder, the fiber winding is required to be monitored, the problems that fiber breakage, poor adhesion, uneven thickness and the like exist during winding are avoided, the quality and the service life of a finished gas cylinder are influenced, and the use safety performance of a high-pressure gas cylinder is ensured.
Currently, the gas cylinder winding process is monitored mainly by means of visual means, in a large-scale production environment, the monitoring result of the gas cylinder winding process depends on the quality of monitoring staff too much, the experience and eyesight of the staff directly influence the accuracy of the result, and more labor cost is required to be input for human cultivation. In addition, the actual gas cylinder production environment is complex, a plurality of factors such as unstable voltage and unstable equipment operation caused by equipment vibration and the like possibly exist, accurate winding monitoring by monitoring staff is more difficult, and accurate, timely and effective winding quality monitoring and optimizing treatment of the carbon fiber winding process are difficult.
Disclosure of Invention
The embodiment of the application provides a fiber winding quality monitoring method, equipment and medium for a high-pressure gas cylinder, which are used for solving the problems that the existing high-pressure gas cylinder fiber winding quality monitoring is excessively labor-consuming and difficult to accurately and efficiently monitor the winding quality.
In one aspect, an embodiment of the present application provides a filament winding quality monitoring method for a high pressure gas cylinder, the method comprising:
acquiring carbon fiber winding monitoring data from a winding monitoring equipment group in real time; the winding monitoring equipment group at least comprises a tension sensor, a vibration sensor, image acquisition equipment and a numerical control winding machine;
based on the carbon fiber winding monitoring data, determining a carbon fiber winding parameter set corresponding to the current moment, and performing parameter value matching on the carbon fiber winding parameter set and a preset winding defect comparison parameter list to determine whether a winding defect exists at the current moment;
under the condition that the winding defect exists at the current moment, determining a defect parameter regular interval set corresponding to the winding defect parameter value based on the winding defect parameter value corresponding to the winding defect, a carbon fiber winding parameter sequence corresponding to the last preset time period and a pre-trained DBSCAN model;
Determining whether to correct a winding control parameter sequence corresponding to the numerical control winding machine based on a matching result of the carbon fiber winding parameter set and the defect parameter regular interval set; wherein, the winding control parameters in the winding control parameter sequence and the carbon fiber winding parameter set have a dependency relationship;
If yes, correcting the winding control parameter sequence, and controlling the numerical control winding machine to operate according to the corrected winding control parameter sequence so as to continuously monitor and pre-warn the carbon fiber winding parameter set; wherein the correction of the winding control parameter sequence is performed based on a correction factor set sequence and the winding control parameter sequence; the correction factor set sequence is determined according to correction curves matched with the difference curves; the difference curve is generated by the carbon fiber winding parameter sequence corresponding to the previous preset time period and the carbon fiber winding parameter set.
In one implementation manner of the present application, determining a carbon fiber winding parameter set corresponding to a current time based on each carbon fiber winding monitoring data specifically includes:
According to the winding monitoring equipment for collecting the carbon fiber winding monitoring data, determining the data type corresponding to the carbon fiber winding monitoring data; wherein the winding monitoring equipment corresponds to the data types one by one; the data types at least comprise a fiber tension value type, a winding machine vibration frequency type, a carbon fiber winding image type and a winding control parameter type; the winding control parameter type corresponding parameters at least comprise: winding speed, fiber winding angle, fiber winding direction, fiber winding thickness and fiber winding layering;
Inputting the carbon fiber winding monitoring data corresponding to the carbon fiber winding image type into a winding drop point identification model trained in advance so as to determine drop point coordinates of a carbon fiber section with a preset length in a core mold three-dimensional coordinate system at the current moment; the drop point coordinates are center point coordinates of winding positions where the carbon fiber sections are located after the carbon fiber sections are wound on the core mold;
Determining a carbon fiber winding interval based on the distance between the falling point coordinates and adjacent fiber segment coordinates corresponding to the carbon fiber segments; wherein the coordinates of the adjacent fiber segments are the coordinates of the wound carbon fiber segments meeting preset conditions; the preset conditions include: the same fiber winding layering as the carbon fiber section, the same winding control parameter set as the carbon fiber section and the closest distance to the carbon fiber section; the carbon fiber winding interval is an interval distance value that the carbon fiber section intersects, is tangent to or is separated from the wound carbon fiber section;
And respectively taking the carbon fiber winding monitoring data and the carbon fiber winding interval corresponding to the fiber tension value type, the winding machine vibration frequency type and the winding control parameter type as carbon fiber winding parameter values of corresponding parameter dimensions, and sequentially adding the carbon fiber winding parameter values to the carbon fiber winding parameter sets.
In one implementation manner of the present application, the carbon fiber winding monitoring data corresponding to the carbon fiber winding image type is input into a winding drop point identification model trained in advance to determine drop point coordinates of a carbon fiber segment with a preset length in a core mold three-dimensional coordinate system at the current time, and the method specifically includes:
Determining a historical falling point dividing line in a carbon fiber winding image from the image acquisition equipment through the winding falling point identification model; the historical falling point dividing line is obtained based on a carbon fiber winding image at the previous moment;
According to a sliding window algorithm, taking the historical falling point dividing line as a starting point, moving a sliding window matched with the width of the carbon fiber along the winding direction of the carbon fiber by the preset length according to a preset step length, and determining the falling point dividing line corresponding to the current moment according to the sliding window after the movement is stopped;
and taking the carbon fiber between the historical drop point dividing line and the drop point dividing line as the carbon fiber section, and determining the drop point coordinate corresponding to the carbon fiber section according to the three-dimensional coordinate system of the mandrel.
In one implementation manner of the present application, before the parameter value matching is performed between the carbon fiber winding parameter set and a preset winding defect comparison parameter list to determine whether a winding defect exists at the current moment, the method further includes:
Acquiring a plurality of history defect-free winding records in a preset database;
according to each history defect-free winding record and a preset winding progress grade, calculating a mean parameter set of the carbon fiber winding parameter sets corresponding to the same preset winding progress grade; the preset winding progress classification is obtained by fiber winding layering classification; the average parameter group comprises an average value of carbon fiber winding parameter values of each parameter dimension;
and adding the average parameter group corresponding to each preset winding progress grade to the preset winding defect comparison parameter list according to the sequence of the preset winding progress grades.
In one implementation manner of the present application, determining the defect parameter regular interval set corresponding to the winding defect parameter value based on the winding defect parameter value corresponding to the winding defect, the carbon fiber winding parameter sequence corresponding to the last preset period and the pre-trained DBSCAN model specifically includes:
Performing parameter analysis on the winding defect parameter value and a carbon fiber winding parameter sequence corresponding to the last preset period, and inputting the winding defect parameter value into the DBSCAN model trained in advance when the parameter analysis result meets the preset defect regulation standard; wherein the preset defect regulation standard comprises a burst defect standard and a preset defect judgment standard;
Under the condition that the preset defect regulation standard is the burst defect standard, determining whether a corresponding burst defect has a history correction record through a first DBSCAN sub-model, acquiring a defect parameter regulation interval set corresponding to the burst defect based on the history correction record or specified operation from a user terminal, and correcting the corresponding preset defect standard;
And under the condition that the preset defect regulation standard is the preset defect standard, clustering the winding defect parameter values through a second DBSCAN sub-model, and determining the defect parameter regulation interval set corresponding to the preset defect standard.
In one implementation of the present application, the method further includes:
Under the condition that the corresponding sudden defect exists in the history correction records, corresponding history correction parameter sets in a plurality of history correction records are determined;
According to each history correction parameter set, determining a parameter maximum value and a parameter minimum value respectively corresponding to each same parameter dimension;
And generating a defect parameter regular interval corresponding to each parameter dimension according to the parameter maximum value and the parameter minimum value corresponding to each parameter dimension, so as to add each defect parameter regular interval to the defect parameter regular interval set.
In one implementation manner of the present application, based on a matching result of the carbon fiber winding parameter set and the defect parameter regular interval set, determining whether to correct a winding control parameter sequence corresponding to the numerical control winding machine specifically includes:
Comparing each carbon fiber winding parameter in the carbon fiber winding parameter group with a corresponding defect parameter regular interval, and determining whether the carbon fiber winding parameter is in the corresponding defect parameter regular interval;
And under the condition that any carbon fiber winding parameter is not in the corresponding defect parameter regular interval, determining the mismatch between the carbon fiber winding parameter group and the defect parameter regular interval set so as to correct the winding control parameter sequence corresponding to the numerical control winding machine.
In one implementation of the present application, the correcting the winding control parameter sequence specifically includes:
Calculating a difference sequence of the carbon fiber winding parameter set and the carbon fiber winding parameter sequence corresponding to the last preset period of time to generate a corresponding difference curve; wherein, the ordinate of the difference curve is the parameter variation amplitude and the abscissa is the time;
according to the difference curve, matching correction curves in a preset curve comparison list to determine corresponding correction factor set sequences according to the matched correction curves;
and multiplying the correction factor set sequence by each winding control parameter in the winding control parameter sequence in a corresponding dimension to obtain a corrected winding control parameter sequence by taking the product value as the corrected winding control parameter.
In another aspect, an embodiment of the present application further provides a filament winding quality monitoring apparatus for a high pressure gas cylinder, the apparatus including:
At least one processor; and a memory communicatively coupled to the at least one processor; wherein the memory stores instructions executable by the at least one processor to enable the at least one processor to:
acquiring carbon fiber winding monitoring data from a winding monitoring equipment group in real time; the winding monitoring equipment group at least comprises a tension sensor, a vibration sensor, image acquisition equipment and a numerical control winding machine;
based on the carbon fiber winding monitoring data, determining a carbon fiber winding parameter set corresponding to the current moment, and performing parameter value matching on the carbon fiber winding parameter set and a preset winding defect comparison parameter list to determine whether a winding defect exists at the current moment;
under the condition that the winding defect exists at the current moment, determining a defect parameter regular interval set corresponding to the winding defect parameter value based on the winding defect parameter value corresponding to the winding defect, a carbon fiber winding parameter sequence corresponding to the last preset time period and a pre-trained DBSCAN model;
Determining whether to correct a winding control parameter sequence corresponding to the numerical control winding machine based on a matching result of the carbon fiber winding parameter set and the defect parameter regular interval set; wherein, the winding control parameters in the winding control parameter sequence and the carbon fiber winding parameter set have a dependency relationship;
If yes, correcting the winding control parameter sequence, and controlling the numerical control winding machine to operate according to the corrected winding control parameter sequence so as to continuously monitor and pre-warn the carbon fiber winding parameter set; wherein the correction of the winding control parameter sequence is performed based on a correction factor set sequence and the winding control parameter sequence; the correction factor set sequence is determined according to correction curves matched with the difference curves; the difference curve is generated by the carbon fiber winding parameter sequence corresponding to the previous preset time period and the carbon fiber winding parameter set.
In yet another aspect, embodiments of the present application also provide a non-volatile computer storage medium storing computer-executable instructions configured to:
acquiring carbon fiber winding monitoring data from a winding monitoring equipment group in real time; the winding monitoring equipment group at least comprises a tension sensor, a vibration sensor, image acquisition equipment and a numerical control winding machine;
based on the carbon fiber winding monitoring data, determining a carbon fiber winding parameter set corresponding to the current moment, and performing parameter value matching on the carbon fiber winding parameter set and a preset winding defect comparison parameter list to determine whether a winding defect exists at the current moment;
under the condition that the winding defect exists at the current moment, determining a defect parameter regular interval set corresponding to the winding defect parameter value based on the winding defect parameter value corresponding to the winding defect, a carbon fiber winding parameter sequence corresponding to the last preset time period and a pre-trained DBSCAN model;
Determining whether to correct a winding control parameter sequence corresponding to the numerical control winding machine based on a matching result of the carbon fiber winding parameter set and the defect parameter regular interval set; wherein, the winding control parameters in the winding control parameter sequence and the carbon fiber winding parameter set have a dependency relationship;
If yes, correcting the winding control parameter sequence, and controlling the numerical control winding machine to operate according to the corrected winding control parameter sequence so as to continuously monitor and pre-warn the carbon fiber winding parameter set; wherein the correction of the winding control parameter sequence is performed based on a correction factor set sequence and the winding control parameter sequence; the correction factor set sequence is determined according to correction curves matched with the difference curves; the difference curve is generated by the carbon fiber winding parameter sequence corresponding to the previous preset time period and the carbon fiber winding parameter set.
According to the technical scheme, whether winding defects exist or not is judged by utilizing multidimensional carbon fiber winding monitoring data, and whether correction is needed or not is further judged by utilizing a continuously monitored carbon fiber winding parameter sequence. If correction is needed, the winding control parameters of the numerical control winding machine are automatically corrected, and a great amount of manpower is not needed to be input in the real-time monitoring process of fiber winding. Meanwhile, high-quality fiber winding monitoring can be more efficiently and conveniently realized, the efficiency of high-pressure gas cylinder manufacturing is improved, the yield of high-pressure gas cylinders is improved, the intelligent level of high-pressure gas cylinders during manufacturing is improved, and the high-pressure gas cylinder monitoring device is more suitable for large-scale high-pressure gas cylinder production and manufacturing scenes.
Drawings
The accompanying drawings, which are included to provide a further understanding of the application and are incorporated in and constitute a part of this specification, illustrate embodiments of the application and together with the description serve to explain the application and do not constitute a limitation on the application. In the drawings:
FIG. 1 is a schematic flow chart of a method for monitoring the quality of filament winding for a high pressure gas cylinder according to an embodiment of the present application;
FIG. 2 is a schematic diagram of filament winding in a filament winding quality monitoring method for a high pressure gas cylinder according to an embodiment of the present application;
Fig. 3 is a schematic structural view of a filament winding quality monitoring apparatus for a high pressure gas cylinder according to an embodiment of the present application.
Detailed Description
In order to make the objects, technical solutions and advantages of the present application more apparent, the technical solutions of the present application will be clearly and completely described below with reference to specific embodiments of the present application and corresponding drawings. 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.
The embodiment of the application provides a fiber winding quality monitoring method, equipment and medium for a high-pressure gas cylinder, which are used for solving the problems that the existing high-pressure gas cylinder fiber winding quality monitoring is too labor-consuming and difficult to accurately and efficiently monitor the winding quality.
Various embodiments of the present application are described in detail below with reference to the attached drawing figures.
The embodiment of the application provides a fiber winding quality monitoring method for a high-pressure gas cylinder, which can comprise the following steps of S101-S105 as shown in fig. 1:
S101, the server acquires carbon fiber winding monitoring data from a winding monitoring equipment set in real time.
The winding monitoring equipment group at least comprises a tension sensor, a vibration sensor, image acquisition equipment and a numerical control winding machine. The tension sensor is arranged at a wire guide port of the numerical control winding machine and is used for detecting the tension applied to the carbon fiber when the carbon fiber is led out from the wire guide port; the vibration sensor is arranged on the numerical control winding machine body and is used for detecting a vibration signal of the numerical control winding machine body, the vibration comes from external factors or self factors when the machine operates, and the vibration can influence the falling point of the fiber on the core mold so as to damage the winding rule of the fiber during winding; the image acquisition equipment can be arranged on the body of the numerical control winding machine and can also be arranged in a fixed area, and can acquire carbon fiber winding images of the falling points of the fibers on the core mold; and controlling the wire guiding port to guide out the carbon fiber by a control computer of the numerical control winding machine, winding the carbon fiber onto a core mold according to a preset winding rule until the carbon fiber wound by a preset fiber winding thickness is obtained on the core mold to obtain the high-pressure gas cylinder. The winding rule can form a space continuous curve track on the surface of the core mold by the carbon fiber, and the winding rule is such as a process of circumferential winding, spiral winding and the like.
In the embodiment of the application, each device in the winding monitoring device group can be connected with a preset server, and can interact with the server in information to transmit control signals and monitoring data. It should be noted that the server is merely an example as an execution body of the filament winding quality monitoring method for the high-pressure gas cylinder, and the execution body is not limited to the server, and the present application is not particularly limited thereto.
When carbon fiber winding is carried out, the server acquires carbon fiber winding monitoring data collected by each device in the winding monitoring device group in real time, so that quality monitoring of fiber winding is carried out.
S102, the server determines a carbon fiber winding parameter set corresponding to the current moment based on the carbon fiber winding monitoring data, and performs parameter value matching on the carbon fiber winding parameter set and a preset winding defect comparison parameter list to determine whether a winding defect exists at the current moment.
In the embodiment of the application, based on the monitoring data of each carbon fiber winding, the carbon fiber winding parameter group corresponding to the current moment is determined, which specifically comprises:
Firstly, the server determines the data type corresponding to each carbon fiber winding monitoring data according to the winding monitoring equipment which collects each carbon fiber winding monitoring data. Wherein, winding monitoring facilities and data type one-to-one. The data types at least comprise a fiber tension value type, a winding machine vibration frequency type, a carbon fiber winding image type and a winding control parameter type. The winding control parameter type corresponding parameters at least comprise: winding speed, fiber winding angle, fiber winding direction, fiber winding thickness, and fiber winding layering.
And then, the server inputs carbon fiber winding monitoring data corresponding to the type of the carbon fiber winding image into a pre-trained winding drop point identification model so as to determine the drop point coordinates of the carbon fiber section with the preset length in the three-dimensional coordinate system of the mandrel at the current moment. The drop point coordinates are center point coordinates of winding positions where the carbon fiber sections are located after the carbon fiber sections are wound on the core mold.
And then, the server determines the carbon fiber winding interval based on the distance between the coordinates of the falling point and the coordinates of the adjacent fiber segment corresponding to the carbon fiber segment. The coordinates of the adjacent fiber segments are the coordinates of the wound carbon fiber segments meeting preset conditions. The preset conditions comprise: the same fiber as the carbon fiber section is wound and layered, the same winding control parameter set as the carbon fiber section is arranged, and the distance between the carbon fiber section and the winding control parameter set is nearest. The carbon fiber winding interval is an interval distance value that the carbon fiber section intersects, is tangent to or is separated from the wound carbon fiber section.
And finally, the server sequentially adds carbon fiber winding monitoring data and carbon fiber winding intervals corresponding to the fiber tension value type, the winding machine vibration frequency type and the winding control parameter type to the carbon fiber winding parameter set respectively serving as carbon fiber winding parameter values of corresponding parameter dimensions.
In other words, each device in the winding monitoring device group can collect carbon fiber winding monitoring data with different data types, which can be understood as that the carbon fiber winding monitoring data collected by different winding monitoring devices can be marked with different data types, and the data types are in one-to-one correspondence with the winding monitoring devices. The server can mark the data type of the carbon fiber winding monitoring data through the carbon fiber winding monitoring data source equipment. Meanwhile, the server determines carbon fiber winding monitoring data of the carbon fiber winding image type by utilizing the data type, and further processes the data of the data type by utilizing a winding drop point identification model. The wrap-around drop point recognition model may be a pre-trained image recognition model, such as a neural network model. And obtaining the carbon fiber winding interval by identifying the distance between the coordinates of the falling point and the adjacent fiber ends. And adding the carbon fiber winding monitoring data corresponding to the fiber tension value type, the winding machine vibration frequency type and the winding control parameter type to the carbon fiber winding parameter set.
The above-mentioned drop point coordinates are understood to be center point coordinates at a winding position after the carbon fiber is wound on the mandrel, that is, center point coordinates of an area where a carbon fiber segment of a predetermined length to be wound on the mandrel is wound on the mandrel. The preset length may be set by a user during actual use, or may be set to different winding areas of the mandrel, which is not particularly limited in the present application. The wrapped carbon fiber segment corresponding to the coordinates of the adjacent fiber segments can be understood as being wrapped around the carbon fiber segment according to the wrapping rule, the same fiber is wrapped around and layered in the previous wrapping period, and the wrapped carbon fiber segment is wrapped by using the same wrapping control parameter set and is the wrapped carbon fiber segment closest to the carbon fiber segment. In addition, when the carbon fiber segment does not have the wound carbon fiber segment meeting the preset condition, the server can search the historical winding record in the database, match the historical carbon fiber segment with the same landing point coordinate as the carbon fiber segment, and take the historical adjacent carbon fiber segment corresponding to the historical carbon fiber segment as the wound carbon fiber segment, thereby obtaining the adjacent fiber segment coordinate. The relationship between carbon fiber segments and wrapped carbon fiber segments is shown in fig. 2, the broken lines 201 are carbon fiber segments, the wrapped carbon fiber segments are 202, and the two are in an intersecting, tangential or separated relationship, the relationship is a separated relationship in the figure, and the carbon fiber wrapping interval is greater than 0.
In the embodiment of the present application, the inputting the carbon fiber winding monitoring data corresponding to the carbon fiber winding image type into the winding drop point identification model trained in advance to determine the drop point coordinates of the carbon fiber segment with the preset length in the three-dimensional coordinate system of the mandrel at the current moment specifically includes:
The server determines a historical drop point dividing line in the carbon fiber winding image from the image acquisition device through the winding drop point identification model. The historical falling point dividing line is obtained based on a carbon fiber winding image at the previous moment. And then, the server moves the sliding window matched with the width of the carbon fiber along the winding direction of the carbon fiber by a preset length according to a preset step length by taking the historical falling point dividing line as a starting point according to a sliding window algorithm, so as to determine the falling point dividing line corresponding to the current moment according to the sliding window after the movement is stopped. And then, the server takes the carbon fiber between the historical falling point dividing line and the falling point dividing line as a carbon fiber section to determine the falling point coordinate corresponding to the carbon fiber section according to the three-dimensional coordinate system of the core mold.
The winding drop point identification model is obtained through training a plurality of fiber drop point image samples, the fiber drop point image samples are marked with an acquisition time, a historical drop point dividing line label, a drop point dividing line label and a carbon fiber section label, and the server carries out the step of identifying the historical drop point dividing line after receiving a carbon fiber winding image acquired by the image acquisition equipment, namely, receiving carbon fiber winding monitoring data corresponding to the type of the carbon fiber winding image.
The server obtains a drop point dividing line from a carbon fiber winding image at the previous moment of the current moment, and takes the drop point dividing line at the previous moment as a historical drop point dividing line of the carbon fiber winding image at the current moment. When the carbon fiber winding image does not exist before the current moment, the user is used for designating or taking the starting point of the carbon fiber in the carbon fiber winding image as a historical falling point dividing line, and the selection is specifically carried out according to actual use. And then moving a sliding window on the carbon fiber which is to be wound on the core mold in the carbon fiber winding image by a preset length according to the carbon fiber winding direction through a sliding window algorithm, wherein when the sliding window does not move, the moving distance judging line of the sliding window is positioned on the historical falling point dividing line.
The preset step length of each movement is designated by a user, the preset step length is smaller than or equal to the preset length, the preset length can be divided by the preset step length, and after the preset length of the sliding window is moved, the position of the judgment line is used as a falling point dividing line according to the movement distance of the sliding window. The historical drop point dividing line is shown as 203 in fig. 2, and the drop point dividing line is shown as 204. The three-dimensional coordinate system of the mandrel can be pre-established by a user and can be established by taking a certain position in the numerically-controlled winding machine body as a coordinate origin. The drop point coordinates of the carbon fiber segment may be the center point coordinates of the carbon fiber segment, such as 205 in fig. 2.
In an embodiment of the present application, before the matching between the carbon fiber winding parameter set and the preset winding defect comparison parameter list to determine whether a winding defect exists at the current time, the method further includes:
The server acquires a plurality of history defect-free winding records in a preset database. And calculating the average parameter set of the carbon fiber winding parameter set corresponding to the same preset winding progress grade according to each history defect-free winding record and the preset winding progress grade. The preset winding progress grades are obtained according to fiber winding layering division. The mean parameter set includes a mean of the carbon fiber winding parameter values for each parameter dimension. And then adding the average parameter group corresponding to each preset winding progress grade to the preset winding defect comparison parameter list according to the sequence of the preset winding progress grades.
That is, the server may acquire a plurality of history defect-free winding records in a preset database connected thereto, and the history defect-free winding records may be acquired when winding carbon fibers in the past, and include carbon fiber winding parameter sets at various moments. The historical defect-free winding record is also marked with a preset winding progress rating, which is obtained by dividing the fiber winding hierarchy, for example, the fiber winding thickness needs to reach N cm, the corresponding fiber winding hierarchy to be wound is N layers, the preset winding progress rating can be N layers, or the N layers of fiber winding hierarchy can be further divided into winding progress ratings, for example, m ratings, i.e., N/m comprises N/m layers of fiber winding hierarchy for each winding progress rating.
The server matches the average parameter set corresponding to the history defect-free winding records in the same preset winding progress classification through the history defect-free winding records and the preset winding progress classification. The method comprises the steps of screening historical defect-free winding records classified in the same preset winding progress, calculating the average value of carbon fiber winding parameter values of each parameter dimension by using each carbon fiber winding parameter set corresponding to the screened historical defect-free winding records, and adding the average value of the carbon fiber winding parameters to the average value parameter set according to the parameter dimensions to obtain the average value parameter set. Meanwhile, the server can add the average parameter group corresponding to each winding progress grade to the preset winding defect comparison parameter list according to the sequence of the preset winding progress grades.
S103, under the condition that the winding defect exists at the current moment, the server determines a defect parameter regular interval set corresponding to the winding defect parameter value based on the winding defect parameter value corresponding to the winding defect, a carbon fiber winding parameter sequence corresponding to the last preset time period and a pre-trained DBSCAN model.
In actual use, it is not necessary to perform a correction process if a winding defect exists, since the winding defect may not need to be resolved immediately. For example, when the fiber is shifted in a predetermined direction during winding, the shift amount does not reach the degree of correction of the control parameter, and only when the shift amount is continuously shifted to a certain degree according to the shift direction, the correction is required. Accordingly, the present application solves the above-described problems by executing the following technical means. The last preset period may be a period that is specified in advance by the user according to the winding time, or may be a period from the time when the server determines that the defect type is the same as the current winding defect and the defect finding time that does not reach the correction degree to the time before the current time.
In the embodiment of the present application, determining a defect parameter regular interval set corresponding to a winding defect parameter value based on the winding defect parameter value corresponding to the winding defect, a carbon fiber winding parameter sequence corresponding to a last preset period, and a pre-trained DBSCAN model specifically includes:
firstly, the server performs parameter analysis on the winding defect parameter value and a carbon fiber winding parameter sequence corresponding to the last preset period, and inputs the winding defect parameter value into a pre-trained DBSCAN model when the parameter analysis result accords with the preset defect regulation standard. The preset defect regulation standard comprises a burst defect standard and a preset defect judgment standard. Under the condition that the preset defect regulation standard is the burst defect standard, determining whether a corresponding burst defect has a history correction record through a first DBSCAN sub-model, acquiring a defect parameter regulation interval set corresponding to the burst defect based on the history correction record or specified operation from a user terminal, and correcting the corresponding preset defect standard. And under the condition that the preset defect regulation standard is the preset defect standard, clustering the winding defect parameter values through a second DBSCAN sub-model, and determining a defect parameter regulation interval set corresponding to the preset defect standard.
The Density-based clustering algorithm (Density-Based Spatial Clustering of Applications with Noise, DBSCAN) model comprises two sub-models which are respectively used for clustering under the sudden defect standard and the pre-judging defect standard. The first DBSCAN sub-model performs clustering training through training samples corresponding to a plurality of burst defects, so that the burst defects and corresponding correction records thereof can be clustered, and the burst defects can be identified and whether the burst defects have corresponding history correction records or not can be clustered after the training is finished. The second DBSCAN sub-model is obtained by training a plurality of winding defect parameter value samples and corresponding defect parameter regular interval set samples, and sample data can be from a database appointed by a user.
The method comprises the steps of matching a carbon fiber winding parameter set with a preset winding defect comparison parameter list, and if the parameter value of a certain parameter dimension in the carbon fiber winding parameter set is not in an error interval of a corresponding winding defect comparison parameter or the difference value between the parameter of the certain dimension in the carbon fiber winding parameter set and the corresponding winding defect comparison parameter is larger than a preset threshold value, indicating that the parameter of the parameter dimension is the winding defect parameter value. Then, the server performs parameter analysis on the winding defect parameter value and the carbon fiber winding parameter sequence, wherein the parameter analysis process may be to generate a parameter change curve of the winding defect parameter value and the carbon fiber winding parameter sequence of the previous preset period, and determine the change rate and the increase amplitude of the corresponding parameter change curve at each moment, and the change rate is obtained according to the slope of the curve at the moment.
If the absolute value of the change rate is larger than a first preset threshold value, the parameter analysis result is proved to accord with the burst defect standard; if the absolute value of the change rate is smaller than or equal to a first preset threshold value and the increase amplitude is smaller than a second preset threshold value, the parameter analysis result does not accord with the preset defect regulation standard; if the absolute value of the change rate is smaller than or equal to a first preset threshold value and the increase amplitude is larger than or equal to a second preset threshold value, the parameter analysis result meets the pre-judging defect standard. The fact that the pre-judging defect standard is met means that the winding curve parameter is greatly increased, for example, the accumulated carbon fiber winding interval is larger than or equal to a second preset threshold value, and the fact that the pre-judging defect standard is met is judged; the step of meeting the sudden defect standard refers to the sudden change of parameters of the winding curve, for example, the sudden difference of the carbon fiber winding interval development trend corresponding to the carbon fiber winding parameter sequence and the carbon fiber winding interval is judged to occur, and the event meeting the sudden defect standard is judged to occur.
In addition, when it is determined that the history correction record does not exist in the burst defect, the server sends prompt information to the user terminal, so that the user specifies the defect parameter regular interval set through the user terminal. The user terminal may be understood as a user mobile phone, a computer, etc., which is not particularly limited in the present application.
And under the condition that the corresponding history correction record exists in the burst defect, determining each corresponding history correction parameter set in a plurality of history correction records. And determining the parameter maximum value and the parameter minimum value respectively corresponding to each same parameter dimension according to each history correction parameter set. And generating defect parameter regular intervals corresponding to the parameter dimensions according to the parameter maximum values and the parameter minimum values corresponding to the parameter dimensions, so as to add the defect parameter regular intervals to the defect parameter regular interval set.
That is, the server obtains a corresponding history correction parameter set through the history correction record, where the history correction parameter set is obtained after correcting the carbon fiber winding parameter set of the history burst defect. The server extracts the maximum value and the minimum value of the parameters with the same parameter dimension in the historical correction parameter set to generate a corresponding defect parameter regular interval, and then generates a set of defect parameter regular intervals with different parameter dimensions according to the parameter dimension.
S104, the server determines whether to correct the winding control parameter sequence corresponding to the digital winding machine based on the matching result of the carbon fiber winding parameter set and the defect parameter regular interval set.
Wherein, the winding control parameters in the winding control parameter sequence have a dependency relationship with the carbon fiber winding parameter group.
In the embodiment of the present application, the determining whether to correct the winding control parameter sequence corresponding to the numerical control winding machine based on the matching result of the carbon fiber winding parameter set and the defect parameter regular interval set specifically includes:
And the server compares each carbon fiber winding parameter value in the carbon fiber winding parameter set with the corresponding defect parameter regular interval to determine whether the carbon fiber winding parameter value is in the corresponding defect parameter regular interval. And under the condition that any carbon fiber winding parameter value is not in the corresponding defect parameter regular interval, determining the mismatch between the carbon fiber winding parameter set and the defect parameter regular interval set so as to correct the winding control parameter sequence corresponding to the digital winding machine.
In other words, the server may compare the carbon fiber winding parameter value of each parameter dimension with the corresponding defect parameter regular interval to determine whether the carbon fiber winding parameter value is within the defect parameter regular interval of the corresponding parameter dimension. For example, the carbon fiber winding parameter value "carbon fiber winding interval A" is the defect parameter regular interval of the corresponding parameter dimension is [ a, b ], and whether A is within the interval [ a, b ] is judged. If the carbon fiber winding parameter values of all the parameter dimensions are in the defect parameter regular interval, the matching of the carbon fiber winding parameter set and the defect parameter regular interval set is described; if the carbon fiber winding parameter value of any parameter dimension is not in the corresponding defect parameter regular interval, the mismatching of the carbon fiber winding parameter set and the defect parameter regular interval set is indicated, and then the correction processing of the winding control parameter sequence is carried out.
The dependency relationship between the winding control parameters and the carbon fiber winding parameter group in the winding control parameter sequence can be established through a knowledge graph, such as the dependency relationship between the winding speed, the fiber winding angle, the fiber winding direction, the fiber winding thickness and the fiber winding layering, and the fiber tension value, the vibration frequency of the winding machine and the carbon fiber winding interval, and can be determined through the following formulas:
Wherein, the parameters a, b, c, d, e in the formula, 、/>、/>In actual use, a knowledge graph is established according to a plurality of winding control parameters and carbon fiber winding parameter groups, and the dependence relation coefficient obtained by analyzing the relation among the parameters is analyzed. /(I)Representing winding speed,/>Is the winding angle of the fiber,/>Is the filament winding direction,/>For filament winding thickness,/>For the fiber winding and layering of the fiber,Is the fiber tension value,/>For the vibration frequency of winding machine,/>Is a carbon fiber winding interval.
S105, the server corrects the winding control parameter sequence under the condition that the corresponding winding control parameter sequence of the digital winding machine is determined to be corrected, and controls the digital winding machine to operate according to the corrected winding control parameter sequence so as to continuously monitor and pre-warn the carbon fiber winding parameter group.
The correction of the winding control parameter sequence is performed based on the correction factor set sequence and the winding control parameter sequence; the correction factor set sequence is determined according to correction curves matched with the difference curves; the difference curve is generated by a carbon fiber winding parameter sequence corresponding to the previous preset time period and the carbon fiber winding parameter set.
In the embodiment of the application, if the server determines that the correction of the winding control parameter sequence is not required, the subsequent correction operation is not executed.
When the server corrects the winding control parameter sequence, the method can comprise the following steps:
Firstly, the server calculates a difference sequence of the carbon fiber winding parameter set and a carbon fiber winding parameter sequence corresponding to a last preset period to generate a corresponding difference curve. The ordinate of the difference curve is the parameter variation amplitude, and the abscissa is the time. And then, matching correction curves in a preset curve comparison list according to the difference curves, so as to determine corresponding correction factor set sequences according to the matched correction curves. And then, carrying out corresponding dimension multiplication on the correction factor set sequence and each winding control parameter in the winding control parameter sequence, so as to obtain a corrected winding control parameter sequence by taking the product value as the corrected winding control parameter.
That is, the server may generate a difference curve by using a parameter difference between the carbon fiber winding parameter set and the carbon fiber winding parameter sequence of the previous preset period, where the parameter difference may be understood as a subtraction calculation of the carbon fiber winding parameters of the corresponding parameter dimensions according to adjacent moments. The difference value calculation is carried out on the carbon fiber winding parameter set at the current moment and the carbon fiber winding parameter set at the last moment of the current moment in the last preset period according to time, corresponding parameter dimensions are subtracted when the difference value is calculated, and then the difference value calculation is carried out on the last moment and the last moment until all the moments of the last preset period are traversed. And calculating the difference value of different parameter dimensions at the same time as a difference curve value at the same time through the calculated difference value sequence of the parameter dimensions and a pre-designated defect generation coefficient, wherein the defect generation coefficient of each parameter dimension such as winding speed, fiber winding angle, fiber winding direction, fiber winding thickness, fiber winding layering, fiber tension value, winding machine vibration frequency, carbon fiber winding interval and the like is m1, m2, m3, m4, m5, m6, m7 and m8 at each time. The defect generation coefficient may be set by a user according to time or other factors during actual use, which is not particularly limited in the present application.
After the difference curve values at all the moments are obtained, a difference curve can be established in a plane rectangular coordinate system. The server uses the difference curve to match with the correction curve in the preset curve comparison list, the matching can be to calculate the cosine similarity between every two curves, and if the cosine similarity is greater than a second preset threshold, the matching of the two curves is indicated, so that the matched correction curve is obtained. At the same time, the correction curve has a correction factor set sequence associated with it, which can be understood as a correction factor set at the current moment and in a time thereafter, the correction factor set containing correction factors for the winding control parameters of the respective parameter dimension.
The server records winding control parameter sequences under different winding time when the fiber is wound, multiplies the correction factor set sequence by each winding control parameter in the winding control parameter sequence to be corrected according to the parameter dimension, namely multiplies the winding control parameters with different parameter dimensions by the correction factors to obtain a product value, thereby obtaining corrected winding control parameters, and generating a new winding control parameter sequence according to the time sequence.
And (3) controlling the operation of the numerical control winding machine by utilizing the corrected winding control parameter sequence, continuously monitoring the real-time carbon fiber winding parameter set according to the steps S101-S105, and giving an early warning when winding defects exist, so as to prompt a user.
According to the technical scheme, whether winding defects exist or not is judged by utilizing multidimensional carbon fiber winding monitoring data, and whether correction is needed or not is further judged by utilizing a continuously monitored carbon fiber winding parameter sequence. If correction is needed, the winding control parameters of the numerical control winding machine are automatically corrected, and a great amount of manpower is not needed to be input in the real-time monitoring process of fiber winding. Meanwhile, high-quality fiber winding monitoring can be more efficiently and conveniently realized, the efficiency of high-pressure gas cylinder manufacturing is improved, the yield of high-pressure gas cylinders is improved, the intelligent level of high-pressure gas cylinders during manufacturing is improved, and the high-pressure gas cylinder monitoring device is more suitable for large-scale high-pressure gas cylinder production and manufacturing scenes.
Fig. 3 provides a schematic structural view of a filament wound quality monitoring apparatus for a high pressure gas cylinder, as shown in fig. 3, the apparatus comprising:
At least one processor; and a memory communicatively coupled to the at least one processor. Wherein the memory stores instructions executable by the at least one processor, the instructions being executable by the at least one processor to enable the at least one processor to:
And acquiring carbon fiber winding monitoring data from the winding monitoring equipment group in real time. The winding monitoring equipment group at least comprises a tension sensor, a vibration sensor, image acquisition equipment and a numerical control winding machine. And determining a carbon fiber winding parameter set corresponding to the current moment based on the carbon fiber winding monitoring data, and performing parameter value matching on the carbon fiber winding parameter set and a preset winding defect comparison parameter list to determine whether a winding defect exists at the current moment. Under the condition that the winding defect exists at the current moment, determining a defect parameter regular interval set corresponding to the winding defect parameter value based on the winding defect parameter value corresponding to the winding defect, a carbon fiber winding parameter sequence corresponding to the last preset period and a pre-trained DBSCAN model. And determining whether to correct the winding control parameter sequence corresponding to the numerical control winding machine based on the matching result of the carbon fiber winding parameter set and the defect parameter regular interval set. Wherein, the winding control parameters in the winding control parameter sequence have a dependency relationship with the carbon fiber winding parameter group. If yes, correcting the winding control parameter sequence, and controlling the operation of the numerical control winding machine according to the corrected winding control parameter sequence so as to continuously monitor and pre-warn the carbon fiber winding parameter set. The correction of the winding control parameter sequence is performed based on the correction factor set sequence and the winding control parameter sequence; the correction factor set sequence is determined according to correction curves matched with the difference curves; the difference curve is generated by a carbon fiber winding parameter sequence corresponding to the previous preset time period and the carbon fiber winding parameter set.
The embodiment of the application also provides a nonvolatile computer storage medium, which stores computer executable instructions, wherein the computer executable instructions are configured to:
And acquiring carbon fiber winding monitoring data from the winding monitoring equipment group in real time. The winding monitoring equipment group at least comprises a tension sensor, a vibration sensor, image acquisition equipment and a numerical control winding machine. And determining a carbon fiber winding parameter set corresponding to the current moment based on the carbon fiber winding monitoring data, and performing parameter value matching on the carbon fiber winding parameter set and a preset winding defect comparison parameter list to determine whether a winding defect exists at the current moment. Under the condition that the winding defect exists at the current moment, determining a defect parameter regular interval set corresponding to the winding defect parameter value based on the winding defect parameter value corresponding to the winding defect, a carbon fiber winding parameter sequence corresponding to the last preset period and a pre-trained DBSCAN model. And determining whether to correct the winding control parameter sequence corresponding to the numerical control winding machine based on the matching result of the carbon fiber winding parameter set and the defect parameter regular interval set. Wherein, the winding control parameters in the winding control parameter sequence have a dependency relationship with the carbon fiber winding parameter group. If yes, correcting the winding control parameter sequence, and controlling the operation of the numerical control winding machine according to the corrected winding control parameter sequence so as to continuously monitor and pre-warn the carbon fiber winding parameter set. The correction of the winding control parameter sequence is performed based on the correction factor set sequence and the winding control parameter sequence; the correction factor set sequence is determined according to correction curves matched with the difference curves; the difference curve is generated by a carbon fiber winding parameter sequence corresponding to the previous preset time period and the carbon fiber winding parameter set.
The embodiments of the present application are described in a progressive manner, and the same and similar parts of the embodiments are all referred to each other, and each embodiment is mainly described in the differences from the other embodiments. In particular, for the apparatus and medium embodiments, since they are substantially similar to the method embodiments, the description is relatively simple, and reference is made to the description of the method embodiments for relevant points.
The device, the medium and the method provided by the embodiment of the application are in one-to-one correspondence, so that the device and the medium also have similar beneficial technical effects as the corresponding method, and the beneficial technical effects of the device and the medium are not repeated here because the beneficial technical effects of the method are described in detail above.
It should also be noted that the terms "comprises," "comprising," or any other variation thereof, are intended to cover a non-exclusive inclusion, such that a process, method, article, or apparatus that comprises a list of elements does not include only those elements but may include other elements not expressly listed or inherent to such process, method, article, or apparatus. Without further limitation, an element defined by the phrase "comprising one … …" does not exclude the presence of other like elements in a process, method, article or apparatus that comprises the element.
The foregoing is merely exemplary of the present application and is not intended to limit the present application. Various modifications and variations of the present application will be apparent to those skilled in the art. Any modification, equivalent replacement, improvement, etc. which come within the spirit and principles of the application are to be included in the scope of the claims of the present application.
Claims (10)
1. A filament winding quality monitoring method for a high pressure gas cylinder, the method comprising:
acquiring carbon fiber winding monitoring data from a winding monitoring equipment group in real time; the winding monitoring equipment group at least comprises a tension sensor, a vibration sensor, image acquisition equipment and a numerical control winding machine;
based on the carbon fiber winding monitoring data, determining a carbon fiber winding parameter set corresponding to the current moment, and performing parameter value matching on the carbon fiber winding parameter set and a preset winding defect comparison parameter list to determine whether a winding defect exists at the current moment;
under the condition that the winding defect exists at the current moment, determining a defect parameter regular interval set corresponding to the winding defect parameter value based on the winding defect parameter value corresponding to the winding defect, a carbon fiber winding parameter sequence corresponding to the last preset time period and a pre-trained DBSCAN model;
Determining whether to correct a winding control parameter sequence corresponding to the numerical control winding machine based on a matching result of the carbon fiber winding parameter set and the defect parameter regular interval set; wherein, the winding control parameters in the winding control parameter sequence and the carbon fiber winding parameter set have a dependency relationship;
If yes, correcting the winding control parameter sequence, and controlling the numerical control winding machine to operate according to the corrected winding control parameter sequence so as to continuously monitor and pre-warn the carbon fiber winding parameter set; wherein the correction of the winding control parameter sequence is performed based on a correction factor set sequence and the winding control parameter sequence; the correction factor set sequence is determined according to correction curves matched with the difference curves; the difference curve is generated by the carbon fiber winding parameter sequence corresponding to the previous preset time period and the carbon fiber winding parameter set.
2. The method for monitoring the fiber winding quality of a high-pressure gas cylinder according to claim 1, wherein the determining of the set of carbon fiber winding parameters corresponding to the current time based on each of the carbon fiber winding monitoring data comprises:
According to the winding monitoring equipment for collecting the carbon fiber winding monitoring data, determining the data type corresponding to the carbon fiber winding monitoring data; wherein the winding monitoring equipment corresponds to the data types one by one; the data types at least comprise a fiber tension value type, a winding machine vibration frequency type, a carbon fiber winding image type and a winding control parameter type; the winding control parameter type corresponding parameters at least comprise: winding speed, fiber winding angle, fiber winding direction, fiber winding thickness and fiber winding layering;
Inputting the carbon fiber winding monitoring data corresponding to the carbon fiber winding image type into a winding drop point identification model trained in advance so as to determine drop point coordinates of a carbon fiber section with a preset length in a core mold three-dimensional coordinate system at the current moment; the drop point coordinates are center point coordinates of winding positions where the carbon fiber sections are located after the carbon fiber sections are wound on the core mold;
Determining a carbon fiber winding interval based on the distance between the falling point coordinates and adjacent fiber segment coordinates corresponding to the carbon fiber segments; wherein the coordinates of the adjacent fiber segments are the coordinates of the wound carbon fiber segments meeting preset conditions; the preset conditions include: the carbon fiber section is positioned in the same fiber winding layering, is the same winding control parameter set as the carbon fiber section and is nearest to the carbon fiber section; the carbon fiber winding interval is an interval distance value that the carbon fiber section intersects, is tangent to or is separated from the wound carbon fiber section;
And respectively taking the carbon fiber winding monitoring data and the carbon fiber winding interval corresponding to the fiber tension value type, the winding machine vibration frequency type and the winding control parameter type as carbon fiber winding parameter values of corresponding parameter dimensions, and sequentially adding the carbon fiber winding parameter values to the carbon fiber winding parameter sets.
3. The fiber winding quality monitoring method for a high-pressure gas cylinder according to claim 2, wherein the carbon fiber winding monitoring data corresponding to the carbon fiber winding image type is input into a winding drop point identification model trained in advance to determine the drop point coordinates of a carbon fiber segment with a preset length in a core mold three-dimensional coordinate system at the current time, and specifically comprises the following steps:
Determining a historical falling point dividing line in a carbon fiber winding image from the image acquisition equipment through the winding falling point identification model; the historical falling point dividing line is obtained based on a carbon fiber winding image at the previous moment;
According to a sliding window algorithm, taking the historical falling point dividing line as a starting point, moving a sliding window matched with the width of the carbon fiber along the winding direction of the carbon fiber by the preset length according to a preset step length, and determining the falling point dividing line corresponding to the current moment according to the sliding window after the movement is stopped;
and taking the carbon fiber between the historical drop point dividing line and the drop point dividing line as the carbon fiber section, and determining the drop point coordinate corresponding to the carbon fiber section according to the three-dimensional coordinate system of the mandrel.
4. The fiber winding quality monitoring method for a high pressure gas cylinder according to claim 1, wherein the carbon fiber winding parameter set is matched with a preset winding defect comparison parameter list in parameter values to determine whether or not a winding defect exists at the current time, the method further comprising:
Acquiring a plurality of history defect-free winding records in a preset database;
according to each history defect-free winding record and a preset winding progress grade, calculating a mean parameter set of the carbon fiber winding parameter sets corresponding to the same preset winding progress grade; the preset winding progress classification is obtained by fiber winding layering classification; the average parameter group comprises an average value of carbon fiber winding parameter values of each parameter dimension;
and adding the average parameter group corresponding to each preset winding progress grade to the preset winding defect comparison parameter list according to the sequence of the preset winding progress grades.
5. The fiber winding quality monitoring method for a high-pressure gas cylinder according to claim 1, wherein determining the defect parameter regular interval set corresponding to the winding defect parameter value based on the winding defect parameter value corresponding to the winding defect, a carbon fiber winding parameter sequence corresponding to a last preset period and a pre-trained DBSCAN model specifically comprises:
Performing parameter analysis on the winding defect parameter value and a carbon fiber winding parameter sequence corresponding to the last preset period, and inputting the winding defect parameter value into the DBSCAN model trained in advance when the parameter analysis result meets the preset defect regulation standard; wherein the preset defect regulation standard comprises a burst defect standard and a preset defect judgment standard;
Under the condition that the preset defect regulation standard is the burst defect standard, determining whether a corresponding burst defect has a history correction record through a first DBSCAN sub-model, acquiring a defect parameter regulation interval set corresponding to the burst defect based on the history correction record or specified operation from a user terminal, and correcting the corresponding preset defect standard;
And under the condition that the preset defect regulation standard is the preset defect standard, clustering the winding defect parameter values through a second DBSCAN sub-model, and determining the defect parameter regulation interval set corresponding to the preset defect standard.
6. A filament winding quality monitoring method for a high pressure gas cylinder according to claim 5, further comprising:
Under the condition that the corresponding sudden defect exists in the history correction records, corresponding history correction parameter sets in a plurality of history correction records are determined;
According to each history correction parameter set, determining a parameter maximum value and a parameter minimum value respectively corresponding to each same parameter dimension;
And generating a defect parameter regular interval corresponding to each parameter dimension according to the parameter maximum value and the parameter minimum value corresponding to each parameter dimension, so as to add each defect parameter regular interval to the defect parameter regular interval set.
7. The method for monitoring the winding quality of the fiber for the high-pressure gas cylinder according to claim 1, wherein determining whether to correct the corresponding winding control parameter sequence of the numerical control winding machine based on the matching result of the carbon fiber winding parameter group and the defect parameter regular interval set specifically comprises:
Comparing each carbon fiber winding parameter value in the carbon fiber winding parameter set with a corresponding defect parameter regular interval, and determining whether the carbon fiber winding parameter value is in the corresponding defect parameter regular interval;
And under the condition that any carbon fiber winding parameter value is not in the corresponding defect parameter regular interval, determining the mismatch between the carbon fiber winding parameter set and the defect parameter regular interval set so as to correct the winding control parameter sequence corresponding to the numerical control winding machine.
8. The method for monitoring the quality of filament winding for a high-pressure gas cylinder according to claim 1, characterized in that the sequence of winding control parameters is modified, in particular comprising:
Calculating a difference sequence of the carbon fiber winding parameter set and the carbon fiber winding parameter sequence corresponding to the last preset period of time to generate a corresponding difference curve; wherein, the ordinate of the difference curve is the parameter variation amplitude and the abscissa is the time;
according to the difference curve, matching correction curves in a preset curve comparison list to determine corresponding correction factor set sequences according to the matched correction curves;
and multiplying the correction factor set sequence by each winding control parameter in the winding control parameter sequence in a corresponding dimension to obtain a corrected winding control parameter sequence by taking the product value as the corrected winding control parameter.
9. A filament wound quality monitoring apparatus for a high pressure gas cylinder, the apparatus comprising:
at least one processor; and
A memory communicatively coupled to the at least one processor; wherein,
The memory stores instructions executable by the at least one processor to enable the at least one processor to perform a filament winding quality monitoring method for a high pressure gas cylinder as set forth in any one of the preceding claims 1-8.
10. A non-transitory computer storage medium storing computer executable instructions, wherein the computer executable instructions are capable of performing a filament winding quality monitoring method for a high pressure gas cylinder as claimed in any one of claims 1 to 8.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202410479484.6A CN118067203B (en) | 2024-04-22 | 2024-04-22 | Fiber winding quality monitoring method, equipment and medium for high-pressure gas cylinder |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202410479484.6A CN118067203B (en) | 2024-04-22 | 2024-04-22 | Fiber winding quality monitoring method, equipment and medium for high-pressure gas cylinder |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN118067203A CN118067203A (en) | 2024-05-24 |
| CN118067203B true CN118067203B (en) | 2024-06-25 |
Family
ID=91107694
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN202410479484.6A Active CN118067203B (en) | 2024-04-22 | 2024-04-22 | Fiber winding quality monitoring method, equipment and medium for high-pressure gas cylinder |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN118067203B (en) |
Citations (1)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN104441598A (en) * | 2014-11-03 | 2015-03-25 | 镇江奥立特机械制造有限公司 | Filament winding control system |
Family Cites Families (6)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| JP7119970B2 (en) * | 2018-12-12 | 2022-08-17 | トヨタ自動車株式会社 | POSITION DETECTION METHOD AND POSITION DETECTION DEVICE |
| CN111239166A (en) * | 2020-03-25 | 2020-06-05 | 嘉兴市特种设备检验检测院 | Microwave detection method for defects of carbon fiber wound composite gas cylinder |
| WO2022067756A1 (en) * | 2020-09-30 | 2022-04-07 | 深圳烯湾科技有限公司 | Control method for fiber winding and curing, and light-curing fiber winding device |
| CN113203795A (en) * | 2021-04-30 | 2021-08-03 | 天津市特种设备监督检验技术研究院(天津市特种设备事故应急调查处理中心) | Gas cylinder damage detection method, device, equipment and storage medium |
| US20230212040A1 (en) * | 2021-12-31 | 2023-07-06 | Wuhan University | Carbon fiber filter and regeneration method therefor, and water treatment device |
| CN114813789A (en) * | 2022-05-26 | 2022-07-29 | 嘉兴市特种设备检验检测院 | Carbon fiber winding IV-type hydrogen storage cylinder defect detection method |
-
2024
- 2024-04-22 CN CN202410479484.6A patent/CN118067203B/en active Active
Patent Citations (1)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN104441598A (en) * | 2014-11-03 | 2015-03-25 | 镇江奥立特机械制造有限公司 | Filament winding control system |
Non-Patent Citations (1)
| Title |
|---|
| 一种大容积钢质内胆碳纤维全缠绕高压气瓶的研制;刘泽林等;《中国化工装备》;20230118;全文 * |
Also Published As
| Publication number | Publication date |
|---|---|
| CN118067203A (en) | 2024-05-24 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN111639921B (en) | Intelligent equipment fault judgment and auxiliary disposal method based on expert system | |
| CN112508053A (en) | Intelligent diagnosis method, device, equipment and medium based on integrated learning framework | |
| CN116341771A (en) | Intelligent optimization method and system for low-temperature-resistant cable production process | |
| CN117220441A (en) | Information monitoring intelligent early warning method based on panoramic monitoring | |
| CN109634233A (en) | Industrial big data intellectual analysis decision-making technique, readable storage medium storing program for executing and terminal | |
| CN112348101B (en) | Steel rolling fuel consumption early warning method and system based on abnormal data analysis | |
| CN111950577A (en) | Point inspection method and device | |
| CN113256151A (en) | Hydrogen quality detection method, system and computer storage medium using the same | |
| CN117473350B (en) | Welded pipe welding production process monitoring method based on data processing | |
| CN118067203B (en) | Fiber winding quality monitoring method, equipment and medium for high-pressure gas cylinder | |
| CN114741844A (en) | Static equipment corrosion monitoring method, device, equipment and storage medium | |
| CN117076955A (en) | Fault detection method and system for high-voltage frequency converter | |
| CN117115169A (en) | Intelligent recognition method for abnormal deformation of surface of die-casting die of automobile part | |
| CN112884212A (en) | Cigarette single gram weight deviation analysis and prediction method | |
| CN110738272B (en) | Method for labeling visual mechanical continuous alarm samples of power transmission line channel | |
| CN111636123B (en) | Intelligent spinning production line fault detection system | |
| CN117829318A (en) | Digital twin system based on large model algorithm | |
| CN116629726B (en) | Environment monitoring management method and system for precise copper pipe packaging workshop | |
| CN112329868A (en) | CLARA clustering-based manufacturing and processing equipment group energy efficiency state evaluation method | |
| CN116277821A (en) | Monitoring method and equipment for abnormality in injection molding process of injection molding machine | |
| CN116308037A (en) | Ship spare part early warning method and system | |
| CN114462699A (en) | Optical fiber production qualification index prediction method based on random forest | |
| CN111444658A (en) | Groove rail geometric parameter trend prediction method and system | |
| CN118211190B (en) | Section data analysis method and system based on working condition grid | |
| CN114971599B (en) | Food manufacturing processing technological parameter setting method, device, equipment and medium |
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 |