An algorithm template acquisition method, system and storage medium for piecework of a sewing machine

Abstract

The application discloses a kind of algorithm template acquisition method, system and storage medium for sewing machine piecework, it is related to mechanical equipment technical field, comprising the following steps: according to the source setting flag of template data, template data is stored classifiedly after;Cloud platform calls stored template data;Template data is respectively processed based on data flag;According to the result of data processing, generate piecework algorithm template, and piecework algorithm template is stored classifiedly;The acquisition of algorithm template is carried out before piecework algorithm carries out piecework, provides basis for piecework algorithm, improves the efficiency and accuracy of piecework algorithm, obtains a variety of different algorithm templates, can be better applied to a variety of different working conditions, improves the universality of algorithm template use.

Description

Technical Field

[0001] This invention relates to the field of mechanical equipment technology, and in particular to a method, system and storage medium for obtaining algorithm templates for piece counting in sewing machines. Background Technology

[0002] Currently, IoT-enabled sewing equipment has largely replaced ordinary sewing machines, and factories have shifted from manual piecework to automated piecework methods using work order software, hanging systems, and piecework algorithms. To design piecework algorithms, a large amount of sewing machine motion data (such as stitch count and thread trimming times) is needed to build and optimize the algorithm model. Different algorithm templates result in significant differences in the accuracy of piecework algorithms. Most algorithms on the market currently require labeling the data generated during the sewing process (at least the data from the first three pieces sewn) to obtain a matching template. Piecework algorithms designed based on this method have poor accuracy. Therefore, obtaining templates more conveniently and accurately through technological means is a pressing problem that piecework algorithm design needs to solve.

[0003] For example, the "Method for Automatic Piece Counting of a Sewing Machine" provided in Chinese Patent CN104018300A adds a piece counter connected to the sewing machine via a signal, and this piece counter is connected to a computer through a repeater to ultimately realize the automatic piece counting function of the sewing machine. This method still requires data template matching during the piece counting process, resulting in a certain delay and a tendency for piece counting errors. Furthermore, the piece counting algorithm template used in this method is too simplistic and cannot meet the requirements of different working conditions. Summary of the Invention

[0004] This invention mainly addresses the problems existing in the prior art; it provides a method, system, and storage medium for obtaining algorithm templates for piece counting in sewing machines, and sets up multiple different piece counting algorithm templates so that the piece counting algorithm does not need to obtain template data during the piece counting process, thereby improving piece counting efficiency and accuracy.

[0005] The above-mentioned technical problem of the present invention is mainly solved by the following technical solution: a method for obtaining an algorithm template for piece counting in sewing machines, comprising the following steps:

[0006] After setting a flag based on the source of the template data, the template data is classified and stored.

[0007] The cloud platform retrieves the stored template data;

[0008] The template data is processed based on the data flag bits;

[0009] Based on the data processing results, piece-counting algorithm templates are generated and then categorized and stored.

[0010] Preferably, the source of the template data includes at least one of a sewing machine and an instruction input terminal.

[0011] Preferably, the specific method for classifying and storing the template data is as follows:

[0012] When the source of the template data is identified as a sewing machine, the template data is stored in the sewing machine subset;

[0013] When the source of the template data is identified as the instruction input terminal, the template data is stored in the instruction input terminal set.

[0014] Preferably, when the template data comes from a sewing machine, the template data includes: a sewing machine start signal, several sewing machine action signals, and a sewing machine completion signal.

[0015] Preferably, when the template data comes from the instruction input terminal, the template data includes: a garment design drawing and sewing machine action signals based on the garment design drawing.

[0016] Preferably, the sewing machine action signals include: the number of stitches and the number of thread cuts during the garment sewing process.

[0017] The present invention also provides an algorithm template acquisition system for piece counting of sewing machines, comprising: a data acquisition module for acquiring template data and setting flag bits according to the source of the template data; and an algorithm server for classifying and storing the template data according to the set flag bits, processing the template data based on the flag bits, and generating piece counting algorithm templates based on the data processing results.

[0018] The present invention also provides a storage medium storing program instructions that execute the above-described method when the program instructions are run.

[0019] The beneficial effects of this invention are:

[0020] 1. Obtain the algorithm template before the piece-counting algorithm performs the piece-counting calculation. This provides a foundation for the piece-counting algorithm and improves its efficiency and accuracy.

[0021] 2. Obtaining multiple different algorithm templates allows for better application to various working conditions, improving the universality of algorithm template usage. Attached Figure Description

[0022] Figure 1 This is a flowchart of the method for obtaining the piece-counting algorithm template according to an embodiment of the present invention.

[0023] Figure 2 This is a template data diagram obtained from a sewing machine according to an embodiment of the present invention. Detailed Implementation

[0024] To make the objectives, technical solutions, and advantages of the present invention clearer, the technical solutions in the embodiments of the present invention will be further described in detail below with reference to the accompanying drawings. It should be understood that the specific embodiments described herein are only for explaining the present invention and are not intended to limit the invention.

[0025] Example:

[0026] A method for obtaining algorithm templates for piece counting in sewing machines, such as... Figure 1 As shown, it includes the following steps:

[0027] S1: After setting a flag based on the source of the template data, the template data is classified and stored. The data header of the template data displays a data source identifier. The template data is marked according to the data source identifier, and after setting a flag, it is classified and stored. The source of the template data includes at least one of the following: sewing machine and instruction input terminal. When the template data is provided from the sewing machine, the template data is transmitted to the cloud platform through the gateway device. The algorithm server is set in the cloud platform, and storage devices are set in the cloud platform to store the data. The algorithm server can quickly call the template data in the storage device.

[0028] When the template data comes from a sewing machine, the template data includes: a sewing machine start signal, several sewing machine action signals, and a sewing machine completion signal. The specific method for obtaining template data from the sewing machine is as follows: Each garment is first sewn using the sewing machine. The sewing machine start signal, several sewing machine action signals, and the sewing machine completion signal are used as the sewing template data for each garment, and this data is sent to the cloud platform's algorithm server via a gateway device.

[0029] The sewing machine action signals mentioned in this invention include the number of stitches and the number of thread cuts during the sewing process. The number of thread cuts corresponds to the sewing process, and one thread cut marks the completion of one process. The number of stitches corresponds to the length of the sewing thread in each process. When setting the matching template based on the number of stitches and the number of thread cuts, the number of thread cuts is set with a specific value, and the number of stitches is set with a threshold range. When the number of thread cuts is the same as the set value and the number of stitches is within the threshold range, one piece count is completed; otherwise, the piece count is discarded and an error is reported.

[0030] For example, the workshop foreman first sews a garment to obtain... Figure 2 The template data shown on the left, Figure 2 The data in the middle and right columns represents worker sewing data. As shown in the graph, the worker in the middle column only had one motor start, which was split into two. The stitch count differed from the template by only one stitch (11+13), and the algorithm counted this as completing one piece. However, the stitch count in the right column was 54, which was significantly different from the template, so it was discarded. This method is used to match the worker's data with a standard template for piece counting (this is just a simple example using stitch count; actual calculations will use more features).

[0031] When the source of the template data is identified as the command input terminal, the template data is stored in the command input terminal set. When the template data comes from the command input terminal, the template data includes: a garment design drawing and sewing machine action signals based on the garment design drawing. The specific method for obtaining template data from the command input terminal is as follows: when the garment leaves the factory, the garment design drawing from the garment factory is obtained. The garment design drawing includes a garment structure drawing and garment dimensions. Based on the dimensions of different edges and corners of the garment, the number of stitches and thread cuts required for the current edge or corner is calculated. The garment structure drawing, garment dimensions, and the number of stitches and thread cuts required for the corresponding position of the garment are stored in the cloud platform.

[0032] For example, importing a garment design drawing includes three steps: sewing the neckline, sewing the shoulder edge, and sewing the cuffs. Completing these three steps counts as one garment. The design drawing shows that the neckline is circular and 20 inches long (virtual data, for example), the shoulder edge is straight and 10 inches long, and the cuffs are circular and 10 inches long. Using these known conditions (the sewing machine's stitch length is also known and stored in the database), we can deduce the sewing machine's action sequence and the number of stitches. When sewing the circular edge, the worker cannot complete it in one go and will start and stop the motor multiple times, resulting in an action sequence like 2323232323236. Assuming a stitch length of 1 inch, this requires 20 stitches. Similarly, the cuffs also follow the 232323236 sequence, requiring 10 stitches. Sewing the shoulders will be completed in one or two steps using 23236 or 236, requiring 10 stitches. The design drawing clearly shows the characteristics of the sewing process, namely the worker's actions and the number of stitches. Based on these conditions, an algorithm template is obtained for piece counting.

[0033] The piece-rate algorithm template at the instruction input end incorporates garment design drawings, which compensates for the shortcomings of the sewing machine piece-rate algorithm template. It also eliminates the need for workers to annotate the process, but the presence of garment design drawings allows for process annotation, which reduces potential errors caused by manual intervention and ensures the integrity of the piece-rate process, making piece-rate more accurate.

[0034] The specific method for classifying and storing template data is as follows: when the source of the template data is identified as a sewing machine, the template data is stored in the sewing machine subset; when the source of the template data is identified as an instruction input terminal, the template data is stored in the instruction input terminal set, and the sewing machine subset and the instruction input terminal set are stored in the storage device.

[0035] Furthermore, the sewing machine action signals mentioned in this invention also include the number of presser foot lifts, the stitch count interval, and the thread trimming interval. The number of stitches and thread trimmings determines whether the current stitch is complete; the stitch count interval and thread trimming interval determine whether the current sewing is smooth and whether sewing malfunctions have occurred; and comparing the presser foot lift count with the stitch count improves the piece counting accuracy of the algorithm template.

[0036] S2: The cloud platform retrieves the stored template data; the cloud platform sends a query command to the storage device to check whether the storage device stores template data. When template data exists in the storage device, it sends a corresponding signal back to the cloud platform, which then retrieves the template data. When multiple template data exist, the template data is retrieved sequentially based on the time information carried by the template time.

[0037] S3: Perform data processing on the template data based on the data flag bits; the specific methods for data processing include processing the sewing machine subset data and processing the instruction input terminal set data.

[0038] The processing method for the sewing machine subset data is as follows: the algorithm server directly calls the sewing machine subset data in the storage device as the algorithm template. If the sewing process matches the sewing machine subset data, it means that a garment has been sewn and a piece count is performed.

[0039] The method for processing instruction input terminal set data is as follows: the process is broken down according to the garment design drawing and garment size, the number of sewing stitches and the number of thread cuts required for each process are obtained, the number of stitches and the number of thread cuts for each process are matched, and when they are completely the same, it means that a garment is sewn and one piece count is performed.

[0040] S4: Generate piece-rate algorithm templates based on the data processing results, and categorize and store these templates. Two types of piece-rate algorithm templates are generated: a sewing machine piece-rate algorithm template and a command input piece-rate algorithm template. The sewing machine piece-rate algorithm template eliminates worker involvement, directly acquiring sewing machine work signals without requiring process annotation, thus reducing potential errors caused by human intervention. However, the sewing machine start signal, sewing machine action signal, and sewing machine completion signal are repetitive throughout the sewing process. Therefore, it is difficult to directly determine the completion and integrity of the process based on the sewing machine action signals, potentially leading to piece-rate errors.

[0041] This invention also provides an algorithm template acquisition system for piecework calculation of sewing machines, comprising: a data acquisition module for acquiring template data and setting flag bits according to the source of the template data; and an algorithm server for classifying and storing the template data according to the set flag bits, processing the template data based on the flag bits, and generating piecework calculation algorithm templates based on the data processing results. The algorithm server of this invention is located within a cloud platform, which is equipped with a storage device. The storage device stores the data information acquired by the data acquisition module. The algorithm server retrieves data from the storage device and stores the generated piecework calculation algorithm templates in the storage device, which are then retrieved by the piecework server for piecework calculation of sewing machines.

[0042] The present invention also provides a storage medium storing program instructions that execute the above-described method when the program instructions are run.

[0043] This invention obtains an algorithm template before the piece-counting algorithm performs piece-counting, providing a foundation for the piece-counting algorithm and improving its efficiency and accuracy. By obtaining multiple different algorithm templates, it can be better applied to various different working conditions, improving the universality of the algorithm templates.

[0044] The embodiments described above are merely preferred embodiments of the present invention and are not intended to limit the present invention in any way. Other variations and modifications are possible without departing from the technical solutions described in the claims.

Claims

1. A method for obtaining an algorithm template for piecework calculation in sewing machines, characterized in that, Includes the following steps: After setting a flag based on the source of the template data, the template data is classified and stored. The template data comes from the instruction input terminal, which obtains the garment design drawings from the garment factory. The garment design drawings include garment structure diagrams and garment dimensions. Based on the size of different edges and corners of the garment and the preset sewing machine stitch length, the number of stitches and thread cuts required for the current edge or corner are calculated, and a sewing machine action signal containing the action sequence is generated accordingly. The cloud platform retrieves the stored template data; The template data is processed based on the data flag bits; Based on the data processing results, piece-counting algorithm templates are generated and then categorized and stored.

2. The method for obtaining an algorithm template for piece counting in sewing machines according to claim 1, characterized in that, The template data can be sourced from at least one of a sewing machine or an instruction input terminal.

3. The method for obtaining an algorithm template for piece counting in sewing machines according to claim 2, characterized in that, The specific method for classifying and storing template data is as follows: When the source of the template data is identified as a sewing machine, the template data is stored in the sewing machine subset; When the source of the template data is identified as the instruction input terminal, the template data is stored in the instruction input terminal set.

4. The method for obtaining an algorithm template for piece counting in sewing machines according to claim 2, characterized in that, When the template data comes from a sewing machine, the template data includes: The sewing machine start signal, several sewing machine action signals, and the sewing machine completion signal.

5. The method for obtaining an algorithm template for piece counting in sewing machines according to claim 2, characterized in that, When the template data comes from the instruction input terminal, the template data includes: Clothing design drawings and sewing machine motion signals based on the clothing design drawings.

6. A method for obtaining an algorithm template for piece counting in sewing machines according to claim 4 or 5, characterized in that, The sewing machine action signals include: The number of stitches and the number of times the thread is cut during the sewing process of a sewing machine.

7. A system for obtaining an algorithm template for piece counting in sewing machines, employing the method described in any one of claims 1-6, characterized in that, include: The acquisition module is used to acquire template data and set flags according to the source of the template data; The algorithm server classifies and stores template data according to the set flag bits, processes the template data based on the flag bits, and generates piece-rate algorithm templates based on the data processing results.

8. A storage medium, characterized in that, The storage medium stores program instructions, which, when executed, perform the method as described in any one of claims 1-6.

Images