Sock machine needle selection monitoring control system based on intelligent typesetting technology
Through the sock machine needle selection monitoring and control system with intelligent typesetting technology, data is collected and analyzed in real time, the accuracy of needle selection execution and structural similarity indicators are calculated, and the work of needle selection is automatically adjusted, which solves the problem of errors in the traditional sock machine needle selection. Improves sock quality and production efficiency.
Patent Information
- Application Number
- CN202411309816.2
- Authority / Receiving Office
- CN · China
- Patent Type
- Applications(China)
- Current Assignee / Owner
- Filing Date
- 2024-09-19
- Publication Date
- 2024-12-06
- Estimated Expiration
- Not applicable · inactive patent
AI Technical Summary
Traditional sock machines lack effective needle selection monitoring and control systems, resulting in errors in needle selection and affecting the quality of socks.
The sock machine needle selection monitoring and control system is adopted based on intelligent typesetting technology, including data acquisition module, status analysis module, control execution module and execution evaluation module, collect and analyze the operating status data of the sock machine in real time, and calculate the accuracy of the action sequence of the needle selection execution. The similarity index to the target pattern structure and automatically adjust the working condition of the needle selection.
Accurate evaluation and automatic adjustment of the sock machine needle selection performance is achieved, ensuring that the sock machine operates in the best condition and improving sock quality and production efficiency.
Smart Images

Figure CN119083015A_ABST
Abstract
Description
Technical Field
[0001] The invention relates to the technical field of needle selection monitoring for a sock machine, and in particular to a needle selection monitoring and control system for a sock machine based on intelligent typesetting technology. Background Art
[0002] With the development of industrial automation, the accuracy of needle selection has a vital impact on the quality of socks when socks are produced by sock machines. In the production process of traditional sock machines, due to the lack of effective needle selection monitoring and control systems, needle selection errors often occur, resulting in quality problems in the produced socks.
[0003] For example, Chinese patent publication number CN113737373A discloses a method for generating a pattern file for a sock machine, comprising: obtaining a pattern of a target product, the pattern of the target product including pattern, design and shape data; responding to a user's module selection instruction, the module including a shoelace hole module and a sock part module group, the module selection instruction being used to select the shoelace hole module and the sock part module group, or to select the sock part module group; converting the pattern, design and shape data of the target product into an action combination, needle selection arrangement and parameter data for knitting by the sock machine; storing the action combination, needle selection arrangement and parameter data for knitting by the sock machine, compiling the pattern of the target product according to the action combination, needle selection arrangement and parameter data for knitting by the sock machine, and forming a pattern file for the target product.
[0004] The prior art controls the sock machine by controlling the pattern of the generated product to adjust the relevant data of the target product when it is working. However, when controlling the sock machine, it is necessary to determine the current pattern based on the execution time of the corresponding needle selection on the socks and the accuracy of the relative command execution, so as to reduce the error generated at this time. Summary of the invention
[0005] In order to solve the above technical problems, the technical solution adopted by the present invention is: a sock machine needle selection monitoring and control system based on intelligent typesetting technology, including: a data acquisition module, used to collect the operating status data of the sock machine, and determine the position of the needle selection on the needle bed of the sock machine and the needle selection action sequence when the sock machine executes the target pattern, and the operating status data includes needle bed position information and needle selection instructions.
[0006] The state analysis module is used to process the position of the needle selection and the action sequence of the needle selection, determine the working status of the needle selection in the running state data, and calculate the accuracy index of the action sequence executed by the needle selection and the structural similarity index of the target pattern.
[0007] The control execution module is used to determine the execution time of the corresponding action sequence executed by the sock machine according to the output result of the state analysis module, and obtain the execution time index corresponding to the execution time of the sock machine.
[0008] The execution evaluation module is used to adjust the current working condition of the sock machine needle selection according to the accuracy index of the action sequence executed by the needle selection, the execution time index and the structural similarity index of the target pattern.
[0009] The beneficial effects of the present invention are as follows: 1. The present invention collects the operating status data of the sock machine in real time, including needle bed position information, needle selection instructions, etc., to provide data support for subsequent monitoring and control.
[0010] 2. The present invention processes and analyzes the collected data, determines the working condition of the needle selection, and calculates the accuracy index of the action sequence executed by the needle selection and the structural similarity index of the target pattern, so as to accurately evaluate the needle selection performance of the sock machine.
[0011] 3. The present invention automatically adjusts the working condition of the needle selection of the sock machine according to the evaluation result, including adjusting the execution time of the needle selection and other parameters, so as to ensure that the sock machine can operate in the best state. BRIEF DESCRIPTION OF THE DRAWINGS
[0012] The present invention is further described below in conjunction with the accompanying drawings and embodiments.
[0013] Figure 1 It is a system framework diagram of a sock machine needle selection monitoring and control system based on intelligent typesetting technology.
[0014] Figure 2 The system diagram of a sock machine needle selection monitoring and control system based on intelligent typesetting technology.
[0015] Figure 3 It is a system flow chart of the needle selection monitoring and control system of a sock machine based on intelligent typesetting technology. DETAILED DESCRIPTION
[0016] The embodiments of the present invention are described in detail below. The embodiments described below are exemplary and are only used to explain the present invention, and should not be construed as limiting the present invention. If no specific techniques or conditions are specified in the embodiments, the techniques or conditions described in the literature in the art or the product specifications are used.
[0017] See also Figure 1 , Figure 2 , Figure 3A needle selection monitoring and control system for a sock machine based on intelligent typesetting technology includes: a data acquisition module, a state analysis module, a control execution module and an execution evaluation module. The data acquisition module is connected to the state analysis module. The data acquisition module sends the collected data to the state analysis module. The state analysis module determines whether there is an abnormality in the current operating state data, and sends the corresponding data to the control execution module. The control execution module judges the execution time of the sock machine according to the data of the state analysis module, and sends the judged comprehensive data to the execution evaluation module. The execution evaluation module adjusts the final output content according to the relevant data to determine the control of the current sock machine.
[0018] The data acquisition module is used to collect the operating status data of the sock machine, determine the position of the needle selection on the needle bed of the sock machine and the action sequence of the needle selection when the sock machine executes the target pattern, and the operating status data includes the needle bed position information and the needle selection instruction; at this time, sensors are used to detect the position and status of the moving parts of the machine, and an embedded system such as PLC is used to collect relevant operating status data, and determine how to select the needle and the corresponding position of the needle at this time.
[0019] The state analysis module is used to process the position of the needle selection and the action sequence of the needle selection, determine the working status of the needle selection in the running status data, and calculate the accuracy index of the action sequence executed by the needle selection and the structural similarity index of the target pattern; at this time, the running status data is analyzed to determine whether the current needle selection complies with the preset normal working mode when performing the task. At the same time, it is necessary to identify whether the current needle selection can generate the same style as the expected image under the set action sequence, and verify whether the currently set needle selection is bent, broken or missing through pattern recognition.
[0020] The control execution module is used to determine the execution time of the corresponding action sequence executed by the sock machine according to the output results of the state analysis module, and obtain the execution time indicator corresponding to the execution time of the sock machine; at this time, the execution time indicator is mainly to verify the time required for the needle selector to work normally when executing the current task, so as to determine whether the current control execution module is operating normally and whether the needle selector is working normally.
[0021] The execution evaluation module is used to adjust the working condition of the current sock machine needle selection according to the accuracy index of the action sequence executed by the needle selection, the execution time index and the structural similarity index of the target pattern; at this time, the monitoring status is determined to represent the problems and characteristics of the socks under the corresponding indicator-related status, so as to determine whether the current sock machine is in a normal working state and whether the needle selection is affected, resulting in work problems.
[0022] Determine the position of the needle selection on the needle bed of the sock machine and the implementation method of the needle selection action sequence when the sock machine executes the target pattern. By using the needle selection instructions in the operating status data, determine the command that needs to be executed for the needle selection when executing the target pattern. The command at this time is the needle selection action sequence, and determine the corresponding position of the needle selection when executing the action sequence from the needle selection action sequence and the corresponding needle bed position information.
[0023] The data acquisition module detects the rotation position and speed of the motor to determine the relative position of the current needle bed, and verifies whether the position of the currently selected needle is normal and whether the yarn set on the current selected needle is broken. It can also measure the change in pressure on the needle bed to indirectly determine the current state of the selected needle.
[0024] Needle selection instructions are a set of instructions generated by the control system, indicating which needles should be selected for knitting and which needles should be skipped; these instructions are based on pre-set patterns or pattern designs; needle selection instructions are the key to achieving specific patterns; by sending the correct needle selection instructions to actuators, such as electromagnets, stepper motors, etc., it can be ensured that only the required needles are activated each time knitting; this can form complex patterns and patterns while ensuring the uniformity and consistency of knitting.
[0025] Needle bed position information refers to the position information of the needle bed in the sock machine relative to the other parts of the machine; the needles on the needle bed are arranged in a certain shape, usually a circular or straight shape, for knitting socks; the position information of the needle bed is crucial to ensure the accuracy of needle selection; by knowing the exact position of the needle bed, the control system can accurately control the needle selection action to ensure that each needle selection occurs on the correct needle; in addition, the needle bed position information can also help the system track the progress of knitting, as well as locate and troubleshoot when errors occur.
[0026] Needle bed position information can usually be obtained in the following ways: Encoder: An encoder is a sensor that can measure the rotation angle and position of the shaft. In a sock machine, the encoder can be installed on the motor shaft that drives the needle bed to rotate to monitor the position changes of the needle bed. Limit switches: In some cases, limit switches can be used to determine whether the needle bed has reached a predetermined position. When the needle bed moves and touches the limit switch, the limit switch will send a signal to notify the control system of the position of the needle bed.
[0027] The needle selection instructions are not directly collected data, but a sequence of instructions generated by the control system according to the designed pattern or pattern logic; however, in order to monitor the needle selection process, the following devices can be used to record the actual needle selection status: Proximity Sensors: can be used to detect whether the needle selector has correctly activated the corresponding knitting needle; Current detection: by detecting the current changes of the solenoid valve or motor that performs the needle selection action, the status of the needle selector can be indirectly monitored.
[0028] Suppose there is a simple system in which the needle bed is driven by a stepper motor. Each step represents a certain displacement. The encoder can be set to correspond to one unit displacement of the needle bed per rotation, and the position information of the needle bed is updated once every rotation. At the same time, every time the PLC receives a needle selection instruction, it will record the current timestamp and the position information of the needle bed, so that the specific position and time of each needle selection action can be tracked.
[0029] The implementation method of processing the collected operating status data in the data acquisition module includes: checking the needle selection instructions currently issued, and determining that each needle selection instruction can comply with the expected pattern rules. The expected pattern rules are expressed by checking whether there are conflicting needle selection instructions, that is, different needle selection commands are issued to the same pin at the same time. Needle selection instructions without contradictions comply with the expected pattern rules, and conflicting needle selection instructions mark the relevant needle selection instructions.
[0030] The implementation method of processing the position of the needle selection and the action sequence of the needle selection in the state analysis module includes: generating a position change sequence for the needle selection instruction according to the change of the needle bed position information when the needle selection instruction is executed; setting the position change signal corresponding to each selected needle in the position change sequence according to the corresponding position of the selected needle, and then identifying the needle bed position information according to the position change of the corresponding selected needle on the needle bed, determining the working situation of the needle bed, and determining the change frequency of the position of each selected needle at this time, to generate a position change signal corresponding to each selected needle.
[0031] Determine the mutual correlation coefficient of the position change signals of adjacent needle selections based on the acquired position change signals; obtain a pair of needle selections corresponding to the maximum value of the needle selection mutual correlation coefficient, and use the area corresponding to the pair of needle selections corresponding to the maximum value of the needle selection mutual correlation coefficient as the target area for observation, obtain the position change signals of the needle selections at both ends of the target area, calculate the extreme values of the position change signals of the needle selections at both ends of the target area, perform parabolic interpolation on three points near the extreme values of the position change signals of the needle selections at both ends of the target area to obtain a parabolic function, and use the time corresponding to the extreme value points of the parabolic function as the time difference of the current target area, and determine the working condition of the needle selection based on the time difference of the target area.
[0032] If the time difference is within an expected range, the needle selection is working properly. If the time difference is outside the expected range, it indicates that the following problems may exist: Synchronization problem: There is no synchronization between the needle selections, which may be due to mechanical failure or control system problems. Mechanical problem: The needle selection or its related mechanical parts may be worn or damaged. Control system problem: There may be problems with the control system, resulting in inconsistent needle selection actions. Sensor problem: The sensor may have false alarms or damage, resulting in inaccurate signal acquisition.
[0033] The purpose of generating a position change signal is to determine whether the current target needle selection performs related work as time changes, and whether the corresponding needle selection has a high correlation coefficient, to determine whether the current needle selection is working normally. Because when using the needle selection to work, the needle selection on the same needle bed will have relatively similar working conditions. Although the needle selection may be different when working, the selection will be highly similar in frequency. At this time, it is necessary to determine whether there is corresponding correlation in the needle selection at the same time point, and when there is correlation, whether there is a synchronization problem or a control problem.
[0034] For example, the calculation method for the correlation coefficient is expressed as: Among them, r represents the correlation coefficient, represents the average value of the position change signal x, x i represents the value of the position change signal x at the i-th time point, Represents the average value of the position change signal y, y i Represents the value of the position change signal y at the i-th time point, n represents the number of time points, i=1,2,,,n.
[0035] The correlation coefficient at this time indicates whether there is a relationship between the position change signals of two adjacent needle selectors. If the correlation coefficient of the position change signals of the two needle selectors is close to 1, it means that the two signals change almost completely synchronously, that is, the change of one signal can almost completely predict the change of the other signal; if the two needle selectors should work independently, but their signals show extremely high correlation, this may indicate a synchronization problem, that is, the two needle selectors do not work independently in actual operation, but affect each other; if a sensor fails, causing it to always output the same data as another normally operating sensor, it will also lead to high correlation.
[0036] The method for obtaining the accuracy index of the executed action sequence includes: sorting the action sequence according to the compiled action combination, needle selection arrangement and parameter data, and the sorting method at this time is to sort the action sequence based on the first executed action; determine whether the sorted action combination, needle selection arrangement and parameter data are the same as the content in the preset action sequence, calculate the error rate, compliance rate and consistency index of the errors, and use the weighted sum of the error rate, compliance rate and consistency index as the output accuracy index.
[0037] The error rate is calculated by comparing the action combinations, needle selection arrangements and parameter data in turn, comparing the corresponding data with the action sequence that has been prepared in the database in advance, determining whether the relevant data under each step is accurate, and treating the differences as errors, and comparing the errors with the overall execution steps to obtain the error rate. The compliance rate is calculated by calculating whether the corresponding data of the current action combination, needle selection arrangement and parameter data are the same as those of the preset action sequence. For the consistency coefficient, the time synchronization error and position deviation in the execution of the action sequence are calculated in turn.
[0038] The time synchronization error is the average value of the error between the expected action time and the actual action time. The position deviation is the average value of the error between the expected action position and the actual action position. The weighted average of the normalized time synchronization error and position deviation is used as the consistency coefficient of the setting.
[0039] The purpose of verifying the action sequence at this time is to determine whether the current action queue has caused the needle selection that should be changed to perform corresponding actions due to communication or current problems, so as to identify the anomalies caused by the needle selection problem.
[0040] The method for obtaining the structural similarity index of the target pattern includes: decomposing the target pattern into multiple pattern decomposition diagrams corresponding to the target pattern according to the generated action sequence, wherein the pattern decomposition diagram is different according to the position of the action sequence execution, and decomposing the originally generated pattern into multiple images according to the corresponding position on the socks. The images obtained at this time can assist in identifying the images that need to be formed in the socks at different times. These images can represent the process of socks knitting. Recognizing these images can assist in determining whether there is an abnormality in the currently used needle selection, such as misalignment, missing or duplication of the pattern of the socks, uneven tightness on the surface of the socks, small holes or broken threads on the socks, and verifying the relevant needle selection positions to realize the management and control of the corresponding needle selection.
[0041] The obtained pattern decomposition diagram is compared with the preset image, the pixel average and variance of the pattern decomposition diagram and the preset image are determined, and the pixel covariance between the pattern decomposition diagram and the preset image is calculated to obtain the corresponding structural similarity index.
[0042] Among them, SSIM represents the structural similarity index, μ 1 Represents the average pixel value of the pattern decomposition graph, μ 2 Represents the average pixel value of the preset image, σ 1 represents the pixel variance of the pattern decomposition graph, σ 2 Represents the pixel variance of the preset image, σ 12 represents the pixel covariance between the pattern decomposition graph and the preset image, C 1 and C 2 represents the regulation constant, usually C 1 =(k 1 L) 2 , C 2 =(k 2 L) 2 ; where is the maximum possible value of the pixel intensity, such as L = 255 in an 8-bit grayscale image; and k 1 =0.01, k 2 =0.03.
[0043] At this time, the structural similarity index of each pattern decomposition diagram is judged in turn. If the structural similarity index reaches the threshold, the output is qualified. If the structural similarity index does not reach the threshold, the output is unqualified, and the position of the needle selection for executing the corresponding action sequence is determined according to the identified position.
[0044] At this time, the image decomposition diagram is realized by following the needle selection instruction to determine the position of each pattern on the socks and how many clusters are required to compile the corresponding pattern when executing sock production. The currently collected image is compared with the needle selection instruction to divide and obtain the corresponding image decomposition diagram.
[0045] The implementation method of determining the execution time of the sock machine when executing the corresponding action sequence in the control execution module includes: determining the time estimation parameter of the target needle selection according to the execution time of the current action sequence, determining the adjusted execution time of the target needle selection according to the time estimation parameter of the current action sequence, and using the ratio of the adjusted execution time to the preset execution time as the output execution time indicator.
[0046] The method for obtaining the duration estimation parameters includes: obtaining the associated duration of each needle selection in the target needle selection area according to the execution time of each needle selection action in the target needle selection area, the number of times the needle selection needs to be associated, and the preset association weight; obtaining the estimated duration of the target needle selection area according to the total number of execution actions in the target needle selection area, the frequency of needle selection work, and the preset work weight.
[0047] The associated duration of each needle selection in the target needle selection area and the estimated duration of the target needle selection area are combined to obtain a corresponding duration estimation parameter.
[0048] The above-mentioned association duration is expressed as the product of the execution time of each needle selection in the area where the target needle selection is located, the number of times the needle selection needs to be associated, and the preset association weight; assuming that there are 3 needle selections in an area, the time required for each to execute an action is 1 second, 1.2 seconds and 1.5 seconds respectively, the number of times required to be associated is 2 times, 3 times and 1 time respectively, and the preset association weight is 1.5; then the calculated association duration is finally 10.65 seconds.
[0049] The above-mentioned estimated duration is the total number of actions performed multiplied by the preset work weight, and then divided by the frequency of needle selection. At this time, assuming that a total of 100 actions need to be performed in an area, the frequency of needle selection is 2 times / second, and the preset work weight is 1.5, then the calculated estimated duration is 75 seconds. The final duration estimation parameter obtained at this time is the sum of the associated duration and the estimated duration.
[0050] The implementation method of determining the adjusted execution duration of the target needle selection according to the duration estimation parameters of the current action sequence includes: inputting the duration estimation parameters into a data model according to the acquired duration estimation parameters, and using the output result of the data model as the adjusted execution duration.
[0051] The implementation method of the data model is to use the duration estimation parameters as input data, divide the historical data of the sock machine into a sample set and a training set in a ratio of seven to three, obtain the associated weights and work weights corresponding to the current duration estimation parameters as environmental variables, and adjust the current output duration estimation parameters. At this time, the duration estimation parameters in the corresponding situation are determined by adjusting the associated weights and work weights, and judge whether the execution time after each adjustment can meet the production requirements. The production requirements here indicate that the currently set execution time needs to meet a value. For example, it can be judged that an execution time is 60 seconds. At this time, the execution time closest to the production requirements is output to determine the current adjusted execution time.
[0052] In the execution evaluation module, according to the accuracy index of the action sequence executed by the needle selection, the execution time index and the structural similarity index of the target pattern, the implementation method of adjusting the current working condition of the needle selection of the sock machine includes: taking the weighted average of the accuracy index, the execution time index and the structural similarity index to obtain a comprehensive index value; judging whether the current comprehensive index value meets the preset conditions, and adjusting the working condition of the needle selection of the sock machine according to the set preset conditions.
[0053] The preset condition is expressed as follows: when the comprehensive index value is within the preset range, the current sock machine needle selection is working normally, and the preset range is used to determine whether the sock machine is currently in a normal state; when the comprehensive index value is greater than the upper limit of the preset range, the position of the needle selection corresponding to the action sequence at this time is determined, and the median value of the execution time of the needle selection at the corresponding position of the action sequence at this time is used as the adjusted execution time to adjust the current working condition of the sock machine needle selection; when the comprehensive index value is less than the lower limit of the preset range, the maximum value of the corresponding needle selection execution time is used as the adjusted execution time to adjust the current working condition of the sock machine needle selection.
[0054] Although the embodiments of the present invention have been shown and described above, it is to be understood that the above embodiments are exemplary and are not to be construed as limitations on the present invention. A person skilled in the art may make changes, modifications, substitutions and variations to the above embodiments within the scope of the present invention and they are still covered by the protection scope of the present invention.
Claims
1. A sock machine needle selection monitoring and control system based on intelligent typesetting technology, It is characterized in that include: A data acquisition module is used to collect the running status data of the sock machine, and determine the position of the needle selection on the needle bed of the sock machine and the action sequence of the needle selection when the sock machine executes the target pattern. The running status data includes the needle bed position information and the needle selection instruction; The state analysis module is used to process the position of the needle selection and the action sequence of the needle selection, determine the working condition of the needle selection in the running state data, and calculate the accuracy index of the action sequence executed by the needle selection and the structural similarity index of the target pattern; A control execution module is used to determine the execution time of the corresponding action sequence executed by the sock machine according to the output result of the state analysis module, and obtain the execution time index corresponding to the execution time of the sock machine; The execution evaluation module is used to adjust the current working condition of the sock machine needle selection according to the accuracy index of the action sequence executed by the needle selection, the execution time index and the structural similarity index of the target pattern.
2. According to claim 1, a sock machine needle selection monitoring and control system based on intelligent typesetting technology, It is characterized in that The implementation method of processing the collected running status data in the data collection module includes: checking the currently issued needle selection instructions to determine whether each needle selection instruction can meet the expected pattern rules.
3. According to claim 1, a sock machine needle selection monitoring and control system based on intelligent typesetting technology, It is characterized in that The implementation method of processing the position of the needle selection and the action sequence of the needle selection in the state analysis module includes generating a position change sequence according to the change of the needle bed position information when the needle selection instruction is executed; setting the position change signal corresponding to each needle selection according to the position of the corresponding needle selection in the position change sequence; Determine the mutual correlation coefficient of the position change signals of adjacent needle selections based on the acquired position change signals; obtain a pair of needle selections corresponding to the maximum value of the needle selection mutual correlation coefficient, and use the area corresponding to the pair of needle selections corresponding to the maximum value of the needle selection mutual correlation coefficient as the target area for observation, obtain the position change signals of the needle selections at both ends of the target area, calculate the extreme values of the position change signals of the needle selections at both ends of the target area, perform parabolic interpolation on three points near the extreme values of the position change signals of the needle selections at both ends of the target area to obtain a parabolic function, and use the time corresponding to the extreme value points of the parabolic function as the time difference of the current target area, and determine the working condition of the needle selection based on the time difference of the target area.
4. According to claim 3, a sock machine needle selection monitoring and control system based on intelligent typesetting technology, It is characterized in that The calculation method for the correlation coefficient is expressed as: Among them, r represents the correlation coefficient, represents the average value of the position change signal x, x i represents the value of the position change signal x at the i-th time point, Represents the average value of the position change signal y, y i Represents the value of the position change signal y at the i-th time point, n represents the number of time points, i=1,2,,,n.
5. According to claim 1, a sock machine needle selection monitoring and control system based on intelligent typesetting technology, It is characterized in that The method for obtaining the accuracy index of the executed action sequence includes: sorting the action sequence according to the compiled action combination, needle selection arrangement and parameter data, determining whether the sorted action combination, needle selection arrangement and parameter data are the same as the content in the preset action sequence, calculating the error rate, compliance rate and consistency index of the errors, and taking the weighted sum of the error rate, compliance rate and consistency index as the output accuracy index.
6. According to claim 1, a sock machine needle selection monitoring and control system based on intelligent typesetting technology, It is characterized in that Methods for obtaining the structural similarity index of the target pattern include: Decomposing the target pattern into a plurality of pattern decomposition graphs corresponding to the target pattern according to the generated action sequence, wherein the pattern decomposition graphs are different according to the positions where the action sequence is executed, and decomposing the originally generated pattern into a plurality of images according to the corresponding positions on the socks; The obtained pattern decomposition diagram is compared with the preset image, the pixel average and variance of the pattern decomposition diagram and the preset image are determined, and the pixel covariance between the pattern decomposition diagram and the preset image is calculated to obtain the corresponding structural similarity index. Among them, SSIM represents the structural similarity index, μ 1 Represents the average pixel value of the pattern decomposition graph, μ 2 Represents the average pixel value of the preset image, σ 1 represents the pixel variance of the pattern decomposition graph, σ 2 Represents the pixel variance of the preset image, σ 12 represents the pixel covariance between the pattern decomposition graph and the preset image, C 1 and C 2 Represents the adjustment constant.
7. According to claim 1, a sock machine needle selection monitoring and control system based on intelligent typesetting technology, It is characterized in that The implementation method of determining the execution time of the sock machine when executing the corresponding action sequence in the control execution module includes: determining the time estimation parameter of the target needle selection according to the execution time of the current action sequence, determining the adjusted execution time of the target needle selection according to the time estimation parameter of the current action sequence, and using the ratio of the adjusted execution time to the preset execution time as the output execution time indicator.
8. According to claim 7, a sock machine needle selection monitoring and control system based on intelligent typesetting technology, It is characterized in that The method for obtaining the duration estimation parameter includes: obtaining the associated duration of each needle selection in the target needle selection area according to the execution time of each needle selection action in the target needle selection area, the number of times the needle selection needs to be associated, and the preset associated weight; obtaining the estimated duration of the target needle selection area according to the total number of execution actions in the target needle selection area, the frequency of needle selection work, and the preset work weight; The associated duration of each needle selection in the target needle selection area and the estimated duration of the target needle selection area are combined to obtain a corresponding duration estimation parameter.
9. According to claim 1, a sock machine needle selection monitoring and control system based on intelligent typesetting technology, It is characterized in that In the execution evaluation module, according to the accuracy index of the action sequence executed by the needle selection, the execution time index and the structural similarity index of the target pattern, the implementation method of adjusting the current working condition of the needle selection of the sock machine includes: taking the weighted average of the accuracy index, the execution time index and the structural similarity index to obtain a comprehensive index value; judging whether the current comprehensive index value meets the preset conditions, and adjusting the working condition of the needle selection of the sock machine according to the set preset conditions.
10. The sock machine needle selection monitoring and control system based on intelligent typesetting technology according to claim 9, It is characterized in that The preset condition indicates that when the comprehensive index value is within the preset range, the current needle selection of the sock machine is normal. The preset range is used to determine whether the sock machine is currently in a normal state. When the comprehensive index value is greater than the upper limit value of the preset range, the needle selection position corresponding to the action sequence at this time is determined, and the median value of the execution time of the needle selection at the corresponding position of the action sequence at this time is used as the adjusted execution time to adjust the current working condition of the needle selection of the sock machine; when the comprehensive index value is less than the lower limit value of the preset range, the maximum value of the corresponding needle selection execution time is used as the adjusted execution time to adjust the current working condition of the needle selection of the sock machine.
Citation Information
Patent Citations
Method and system for generating flower type file for hosiery machine, control system and hosiery machine
CN113737373A
