A sealing control method and system for intelligent manufacturing workshop

Abstract

The present invention is applicable to the field of sealing control technology, and in particular, relates to a sealing control method and system for an intelligent manufacturing workshop, the method comprising creating rows according to the numbers of production equipment, creating columns according to the serial numbers of microprocessors, obtaining a matrix, acquiring signals of each target microprocessor at different times based on the matrix, and obtaining a signal matrix containing time tags; adjusting the image acquisition accuracy and sealing trigger conditions of a visual inspection instrument installed at a sealing device according to the signal matrix; controlling the operation of the sealing device based on product detection by the adjusted visual inspection instrument; the present invention selects several relatively important microprocessors in the entire assembly line, regularly acquires the signals of the microprocessors, and performs periodic verification, and judges the stability of the entire assembly line according to the degree of periodicity. When the stability is high, the performance of the visual inspection instrument is appropriately reduced, thereby reducing the load of the visual inspection instrument and reducing resource loss.

Description

Technical Field

[0001] The present invention relates to the technical field of sealing control, and in particular to a sealing control method and system for an intelligent manufacturing workshop. Background Art

[0002] The workshop sealing device is usually used to package and seal products to ensure that they are always in a clean space during transportation and storage, while ensuring the safety of the products. In the unmanned intelligent manufacturing workshop, the sealing device is also automatically triggered. When the product is detected and identified as meeting the requirements, a trigger instruction is generated to control the operation of the sealing device.

[0003] Most of the existing product identification solutions rely on visual inspection instruments, which collect product images and then determine whether they meet the requirements. In this process, the working parameters of the visual inspection instrument are fixed. To ensure the accuracy of the test results, the visual inspection instrument is always in a high-performance working mode with a high load. How to reduce the load of the visual inspection instrument at the workshop sealing device and reduce resource loss is the technical problem raised by the present invention. Summary of the invention

[0004] The purpose of the present invention is to provide a sealing control method and system for an intelligent manufacturing workshop to solve the problem of "reducing the workload of the visual inspection instrument at the workshop sealing device to reduce resource loss" proposed in the above background technology.

[0005] To achieve the above object, the present invention provides the following technical solutions:

[0006] A sealing control method for an intelligent manufacturing workshop, the method comprising:

[0007] Query the serial numbers of all production equipment in the manufacturing line, and query the microprocessor with a data transmission module in each production equipment;

[0008] Selecting a preset number of microprocessors from all microprocessors with data transmission modules in each production device as target microprocessors, and sorting the target microprocessors;

[0009] Create rows according to the number of the production equipment, create columns according to the serial number of the microprocessor, and obtain a matrix. Based on the matrix, obtain the signals of each target microprocessor at different times to obtain a signal matrix containing time tags;

[0010] According to the signal matrix, the image acquisition accuracy and sealing triggering conditions of the visual inspection device installed at the sealing device are adjusted;

[0011] The product is inspected by the adjusted visual inspection instrument. When the inspection result meets the sealing trigger condition, a sealing execution instruction is generated to control the operation of the sealing device.

[0012] The independent variable of the adjustment process is the degree of obviousness of the periodic characteristics of the signal matrix. The higher the degree of obviousness, the lower the image acquisition accuracy and the lower the sealing triggering condition.

[0013] Furthermore, the step of selecting a preset number of microprocessors from all microprocessors with data transmission modules of each production device as target microprocessors and sorting the target microprocessors includes:

[0014] Query the priority of production equipment; the priority of production equipment is used to represent its impact on the production process. The higher the impact, the higher the priority.

[0015] Selecting production equipment in descending order of priority until the number of production equipment reaches a preset first quantity threshold as target production equipment;

[0016] Query the microprocessor with the data transmission module in the target production equipment to obtain the priority of the microprocessor; wherein the priority of the microprocessor is used to characterize the degree of its influence on the production equipment, and the higher the degree of influence, the higher the priority;

[0017] In each target production device, microprocessors are selected in descending order of priority until the number of microprocessors reaches a preset second number threshold as target microprocessors, and the target microprocessors are sorted according to the priority.

[0018] Furthermore, the steps of creating rows according to the numbers of the production equipment, creating columns according to the serial numbers of the microprocessors, obtaining a matrix, and acquiring signals of each target microprocessor at different times based on the matrix to obtain a signal matrix containing time tags include:

[0019] Create rows corresponding to the serial numbers of the production equipment one by one, create columns corresponding to the serial numbers of the microprocessors one by one, and obtain a matrix; wherein the row and column positions in the matrix correspond to the microprocessors in the production equipment one by one;

[0020] Acquire signals from each target microprocessor based on a preset data acquisition frequency, copy the matrix, and insert the signals into corresponding row and column positions in the copied matrix;

[0021] The acquisition time of the signal is used as the time tag of the matrix after the signal is inserted, and the signal matrix containing the time tag is obtained;

[0022] Among them, signal regularization is required before signal insertion. The signal regularization method is as follows:

[0023] Count historical signals, determine the majority signal, compare other signals with the majority signal, and output the comparison value as the regularized signal.

[0024] Furthermore, the step of adjusting the image acquisition accuracy of the visual inspection instrument installed at the sealing device and the sealing triggering condition thereof according to the signal matrix includes:

[0025] Arrange the signal matrix based on time labels;

[0026] Performing feature extraction on the signal matrix to convert the signal matrix into a single value; wherein there is only one feature extraction scheme, which is applicable to all signal matrices;

[0027] Based on time fitting, a single value corresponding to each signal matrix is ​​obtained to obtain a characteristic change function;

[0028] Perform Fourier transform on the characteristic change function to obtain the fundamental frequency amplitude and the harmonic wave amplitude, determine the probability of the existence of a cycle according to the fundamental frequency amplitude and the harmonic wave amplitude, and when the probability reaches a preset probability threshold, determine the cycle according to the fundamental frequency;

[0029] The image acquisition accuracy and the sealing trigger condition are determined based on the probability; the image acquisition accuracy is inversely proportional to the probability. When the recognition model adopts a convolutional recognition model, the sealing trigger condition adopts a convolution kernel matching number threshold, and the convolution kernel matching number threshold is inversely proportional to the probability.

[0030] Furthermore, the extracting features from the signal matrix and converting the signal matrix into a single value includes:

[0031] Querying the priority of the target microprocessor corresponding to each row and column position in the signal matrix, and determining the weight according to the priority;

[0032] All values ​​in the matrix are summed based on the weights to produce a single value.

[0033] Furthermore, the step of extracting features from the signal matrix and converting the signal matrix into a single value also includes:

[0034] Partitioning the signal matrix to obtain different sub-arrays, and recording the position number of each sub-array;

[0035] Querying the priority of the target microprocessor corresponding to each row and column position in the subarray, and determining the weight according to the priority;

[0036] All values ​​of the subarray are summed based on the weights to obtain a single value of the subarray; when fitting a single value, a single value corresponding to the same subarray is fitted according to the position number; when there is a probability of a period, the parameters use the mean of the fundamental frequency amplitude and the mean of the harmonic wave amplitude.

[0037] The technical solution of the present invention also provides a sealing control system for an intelligent manufacturing workshop, the system comprising:

[0038] A microprocessor query module is used to query the serial numbers of all production equipment in the manufacturing line and query the microprocessor with a data transmission module in each production equipment;

[0039] A microprocessor selection module is used to select a preset number of microprocessors from all microprocessors with data transmission modules of each production device as target microprocessors, and sort the target microprocessors;

[0040] The matrix establishment module is used to create rows according to the serial numbers of the production equipment and columns according to the serial numbers of the microprocessors to obtain a matrix, and to obtain the signals of each target microprocessor at different times based on the matrix timing to obtain a signal matrix containing time tags;

[0041] A matrix recognition module, used to adjust the image acquisition accuracy and sealing triggering conditions of a visual inspection instrument installed at the sealing device according to the signal matrix;

[0042] A control instruction generation module is used to detect products based on the adjusted visual inspection instrument. When the detection result meets the sealing trigger condition, a sealing execution instruction is generated to control the operation of the sealing device;

[0043] The independent variable of the adjustment process is the degree of obviousness of the periodic characteristics of the signal matrix. The higher the degree of obviousness, the lower the image acquisition accuracy and the lower the sealing triggering condition.

[0044] Furthermore, the microprocessor selection module includes:

[0045] A first query unit is used to query the priority of the production equipment; wherein the priority of the production equipment is used to represent the degree of its influence on the production process, and the higher the degree of influence, the higher the priority;

[0046] A first selection unit is used to select production equipment in descending order of priority until the number of production equipment reaches a preset first number threshold as target production equipment;

[0047] A second query unit is used to query the microprocessor with a data transmission module in the target production equipment to obtain the priority of the microprocessor; wherein the priority of the microprocessor is used to characterize the degree of its influence on the production equipment, and the higher the degree of influence, the higher the priority;

[0048] The second selection unit is used to select microprocessors in each target production device in descending order of priority until the number of microprocessors reaches a preset second number threshold as target microprocessors, and sort the target microprocessors according to priority.

[0049] Furthermore, the matrix establishment module includes:

[0050] Establish an execution unit for creating rows corresponding to the serial numbers of the production equipment and creating columns corresponding to the serial numbers of the microprocessors, thereby obtaining a matrix; wherein the row and column positions in the matrix correspond to the microprocessors in the production equipment;

[0051] A signal insertion unit, used for acquiring signals of each target microprocessor based on a preset data acquisition frequency, copying the matrix, and inserting the signals into corresponding row and column positions in the copied matrix;

[0052] A label generating unit, used for taking the acquisition time of the signal as the time label of the matrix after the signal is inserted, so as to obtain a signal matrix containing the time label;

[0053] Among them, signal regularization is required before signal insertion. The signal regularization method is as follows:

[0054] Count historical signals, determine the majority signal, compare other signals with the majority signal, and output the comparison value as the regularized signal.

[0055] Furthermore, the matrix identification module includes:

[0056] an arranging unit, for arranging the signal matrix based on the time tags;

[0057] A feature extraction unit, used to extract features from the signal matrix and convert the signal matrix into a single value; wherein there is only one feature extraction scheme, which is applicable to all signal matrices;

[0058] A matrix fitting unit, used for fitting a single value corresponding to each signal matrix based on time to obtain a characteristic change function;

[0059] A cycle identification unit is used to perform Fourier transform on the characteristic change function to obtain the fundamental frequency amplitude and the harmonic wave amplitude, and determine the probability of the existence of the cycle according to the fundamental frequency amplitude and the harmonic wave amplitude. When the probability reaches a preset probability threshold, the cycle is determined according to the fundamental frequency;

[0060] An updating unit is used to determine the image acquisition accuracy and the sealing trigger condition based on the probability; the image acquisition accuracy is inversely proportional to the probability. When the recognition model adopts a convolutional recognition model, the sealing trigger condition adopts a convolution kernel matching number threshold, and the convolution kernel matching number threshold is inversely proportional to the probability.

[0061] Compared with the prior art, the beneficial effects of the present invention are: the present invention selects several relatively important microprocessors in the entire assembly line, obtains the signals of the microprocessors at a regular interval, and performs periodic verification, and judges the stability of the entire assembly line according to the degree of periodicity. When the stability is high, the performance of the visual inspection instrument is appropriately reduced, thereby reducing the load of the visual inspection instrument and reducing resource loss. BRIEF DESCRIPTION OF THE DRAWINGS

[0062] In order to more clearly illustrate the technical solutions in the embodiments of the present invention, the drawings required for use in the embodiments or the description of the prior art will be briefly introduced below. Obviously, the drawings in the following description are only some embodiments of the present invention.

[0063] Figure 1 A flowchart of a sealing control method for an intelligent manufacturing workshop provided in an embodiment of the present invention.

[0064] Figure 2 A block diagram of the composition of the intelligent manufacturing workshop sealing control system provided in an embodiment of the present invention. DETAILED DESCRIPTION

[0065] In order to make the purpose, technical solution and advantages of the present invention more clearly understood, the present invention is further described in detail below in conjunction with the accompanying drawings and embodiments. It should be understood that the specific embodiments described herein are only used to explain the present invention and are not intended to limit the present invention.

[0066] Figure 1 The implementation process of the sealing control method for an intelligent manufacturing workshop provided by an embodiment of the present invention is shown and described in detail below:

[0067] Step S100: querying the serial numbers of all production equipment in the manufacturing line, and querying the microprocessor with a data transmission module in each production equipment;

[0068] The intelligent manufacturing process is an intelligent assembly line. The sealing device is generally set at the end of the intelligent assembly line for sealing or packaging to ensure the safety of the contents and prevent leakage. Common sealing devices include heat sealing devices and mechanical sealing machines. In addition to sealing devices, other production equipment will be involved in the intelligent assembly line. There are a large number of intelligent control modules in these production equipment, which are used to complete corresponding work tasks without human operation. The intelligent assembly line is the manufacturing assembly line in the steps, and the intelligent control module is the microprocessor in the steps.

[0069] Step S200: selecting a preset number of microprocessors from all microprocessors with data transmission modules of each production device as target microprocessors, and sorting the target microprocessors;

[0070] What is needed in this application is a microprocessor with a data transmission function. For each production device, the same number of microprocessors are selected from the many microprocessors in the device as target microprocessors. After the selection is completed, the selected microprocessors in the same device can be sorted. In actual scenarios, the number of selected target microprocessors is generally not large. Three to five microprocessors are selected in each production device as target microprocessors. It should be noted that once the number is determined, the same number of microprocessors needs to be selected in all production devices.

[0071] Step S300: Create rows according to the numbers of the production equipment, create columns according to the serial numbers of the microprocessors, obtain a matrix, and obtain signals of each target microprocessor at different times based on the matrix to obtain a signal matrix containing time tags;

[0072] The production equipment itself has a number. Since the production equipment has a number, it does not need to be sorted. Of course, it can also be sorted according to the relative position in the assembly line; create rows corresponding to the production equipment one by one, create columns corresponding to the serial numbers of the microprocessors one by one, and obtain a matrix as the basic data structure. At this time, each microprocessor of each production equipment can correspond to a row and column position; when the signals of each target microprocessor at different times are obtained, the signals are filled in the corresponding row and column positions to obtain the signal matrix at each time.

[0073] It is worth mentioning that the signal formats of different target microprocessors are different. Therefore, before filling, the signal needs to be regularized so that it falls within a preset numerical range, and then filled to make the data in the signal matrix more regular.

[0074] Step S400: adjusting the image acquisition accuracy and sealing triggering conditions of the visual inspection instrument installed at the sealing device according to the signal matrix;

[0075] Analyze the signal matrix at different times to determine its periodicity. Whether the assembly line is in a normal state is determined based on whether there is periodicity. If the periodicity is strong, it is considered that the intelligent production process is in a normal state. At this time, reduce the image acquisition accuracy of the visual inspection instrument and simultaneously reduce the sealing trigger conditions. In layman's terms, the stronger the periodicity, the more stable the intelligent production process. At this time, only the general characteristics of the contents need to be vaguely obtained, and then when certain lower conditions are met, the sealing instruction can be triggered to control the sealing device to work. In summary, the independent variable of the adjustment process is the degree of obviousness of the periodic characteristics of the signal matrix. The higher the degree of obviousness, the lower the image acquisition accuracy and the lower the sealing trigger conditions.

[0076] The working structure of the above content is: a camera is installed at the sealing device, the camera is used to obtain the image of the contents and feed it back to the sealing device. The sealing device recognizes the image, and when the image meets the preset sealing trigger conditions, a sealing instruction is generated to control the operation of the sealing device.

[0077] Step S500: inspecting the product based on the adjusted visual inspection instrument, and when the inspection result meets the sealing trigger condition, generating a sealing execution instruction to control the operation of the sealing device;

[0078] The product is inspected based on the adjusted visual inspection instrument. When the inspection results meet the requirements, an execution instruction is generated and sealing is performed.

[0079] It is worth mentioning that the present invention can also introduce abnormal judgment conditions, that is, to analyze the signal matrix at the current moment. If it meets the abnormal judgment conditions, for example, a certain value or several values ​​exceed the preset range, and the excess is large, at this time, the highest level of image acquisition accuracy and its sealing trigger conditions are directly used to identify the product and determine whether to perform the sealing task.

[0080] Regarding step S200, the step of selecting a preset number of microprocessors from all microprocessors with data transmission modules of each production device as target microprocessors, and sorting the target microprocessors includes:

[0081] Query the priority of production equipment; the priority of production equipment is used to represent its impact on the production process. The higher the impact, the higher the priority.

[0082] Selecting production equipment in descending order of priority until the number of production equipment reaches a preset first quantity threshold as target production equipment;

[0083] Query the microprocessor with the data transmission module in the target production equipment to obtain the priority of the microprocessor; wherein the priority of the microprocessor is used to characterize the degree of its influence on the production equipment, and the higher the degree of influence, the higher the priority;

[0084] In each target production device, microprocessors are selected in descending order of priority until the number of microprocessors reaches a preset second number threshold as target microprocessors, and the target microprocessors are sorted according to the priority.

[0085] The above content specifically describes the process of selecting the target microprocessor. First, the production equipment is selected based on the priority of the production equipment. The purpose of selecting the production equipment is to limit the number of production equipment, and then limit the number of rows in the matrix; then, for each production equipment, a preset number of microprocessors are selected based on the priority of the microprocessor.

[0086] Step S300, the steps of creating rows according to the numbers of the production equipment, creating columns according to the serial numbers of the microprocessors, obtaining a matrix, and acquiring signals of each target microprocessor at different times based on the matrix timing to obtain a signal matrix containing time tags include:

[0087] Create rows corresponding to the serial numbers of the production equipment one by one, create columns corresponding to the serial numbers of the microprocessors one by one, and obtain a matrix; wherein the row and column positions in the matrix correspond to the microprocessors in the production equipment one by one;

[0088] Acquire signals from each target microprocessor based on a preset data acquisition frequency, copy the matrix, and insert the signals into corresponding row and column positions in the copied matrix;

[0089] The acquisition time of the signal is used as the time tag of the matrix after the signal is inserted, and the signal matrix containing the time tag is obtained;

[0090] The rows of the obtained matrix correspond to the production equipment one by one, and the columns in the same row correspond to the microprocessors in the production equipment one by one; the signal of the microprocessor is obtained at preset time intervals (preset data acquisition frequency), the signal is regularized, inserted into the matrix, and the signal acquisition time is used as a time label to obtain the signal matrix; since each moment corresponds to a signal matrix, before inserting data, it is necessary to copy the matrix first and then insert it, that is, the originally created matrix is ​​used as a template. When data needs to be inserted, the template is copied and then inserted. The copy-insert process is a very common data processing method.

[0091] Specifically, signal regularization is required before signal insertion. The signal regularization method is as follows:

[0092] Count historical signals, determine the majority signal, compare other signals with the majority signal, and output the comparison value as the regularized signal.

[0093] The purpose of regularization is very simple, which is to convert signals of different formats into a numerical value. The signal is a higher-level concept, which is actually the output of each microprocessor. For a certain microprocessor, its output is recorded to obtain historical output, and the majority value is selected from the historical output. When a new output is obtained, the remaining majority values ​​are compared to obtain the regularized numerical value. The simplest comparison method is to first convert data of different formats into numerical values ​​(dimensionless processing) with the help of a data conversion model, and then calculate the ratio of the converted numerical values ​​to obtain the regularized data.

[0094] Step S400, the step of adjusting the image acquisition accuracy and sealing triggering conditions of the visual inspection instrument installed at the sealing device according to the signal matrix includes:

[0095] Arrange the signal matrix based on time labels;

[0096] Performing feature extraction on the signal matrix to convert the signal matrix into a single value; wherein there is only one feature extraction scheme, which is applicable to all signal matrices;

[0097] Based on time fitting, a single value corresponding to each signal matrix is ​​obtained to obtain a characteristic change function;

[0098] Perform Fourier transform on the characteristic change function to obtain the fundamental frequency amplitude and the harmonic wave amplitude, determine the probability of the existence of a cycle according to the fundamental frequency amplitude and the harmonic wave amplitude, and when the probability reaches a preset probability threshold, determine the cycle according to the fundamental frequency;

[0099] The image acquisition accuracy and the sealing trigger condition are determined based on the probability; the image acquisition accuracy is inversely proportional to the probability. When the recognition model adopts a convolutional recognition model, the sealing trigger condition adopts a convolution kernel matching number threshold, and the convolution kernel matching number threshold is inversely proportional to the probability.

[0100] Arrange the signal matrix in chronological order, extract the features of any of the arranged signal matrices according to the same rule, and obtain a single value corresponding to the signal matrix; fit a function with time as the independent variable and the single value as the dependent variable, which is called the characteristic change function; perform Fourier transform on the characteristic change function to obtain the fundamental frequency amplitude and the harmonic wave amplitude, and determine the probability of the existence of a period based on the fundamental frequency amplitude and the harmonic wave amplitude; Fourier transform itself is to convert the function into the frequency domain, and then obtain its frequency domain characteristics. It can convert any function into a sum of multiple periodic functions of different frequencies, such as trigonometric functions. If the signal is periodic, there will be an obvious fundamental frequency in its spectrum. Calculate the amplitude ratio of the fundamental frequency to the first harmonic, the second harmonic, etc. If the amplitude of the fundamental frequency is significantly larger than the amplitude of the harmonic, the signal is more likely to be periodic, that is, the probability of the existence of a period in the characteristic change function can be determined based on the fundamental frequency amplitude and the harmonic wave amplitude. When the probability is high enough, select the fundamental frequency, calculate the period based on the fundamental frequency, and regard it as the period of the characteristic change function.

[0101] Furthermore, the extracting features from the signal matrix and converting the signal matrix into a single value includes:

[0102] Querying the priority of the target microprocessor corresponding to each row and column position in the signal matrix, and determining the weight according to the priority;

[0103] All values ​​in the matrix are summed based on the weights to produce a single value.

[0104] The purpose of the feature extraction scheme is very simple, which is to convert the values ​​of all elements of the matrix into a numerical value. The above content provides a specific scheme to query the priority of the target microprocessor corresponding to each row and column position, determine the weight according to the priority, and multiply the weight by the numerical value and then sum them up to get a single numerical value.

[0105] Furthermore, the step of extracting features from the signal matrix and converting the signal matrix into a single value further includes:

[0106] Partitioning the signal matrix to obtain different sub-arrays, and recording the position number of each sub-array;

[0107] Querying the priority of the target microprocessor corresponding to each row and column position in the subarray, and determining the weight according to the priority;

[0108] All values ​​of the subarray are summed based on the weights to obtain a single value of the subarray; when fitting a single value, a single value corresponding to the same subarray is fitted according to the position number; when there is a probability of a period, the parameters use the mean of the fundamental frequency amplitude and the mean of the harmonic wave amplitude.

[0109] It is worth mentioning that if the matrix is ​​large, that is, the number of selected production equipment and the number of selected microprocessors are large, the signal matrix can be partitioned. For example, a 16*8 matrix can be divided into four 8*4 matrices. Then, each sub-array is analyzed separately, and the single value of the sub-array is calculated using a scheme similar to the original scheme, and four single values ​​can be obtained.

[0110] After obtaining four single values, the subsequent characteristic change function fitting process will also change, and four characteristic change functions can be obtained. When performing periodic identification on the four characteristic change functions, the mean of the fundamental frequency amplitude and the mean of the harmonic amplitude are calculated first, and then a periodic analysis is performed. This split calculation scheme is equivalent to scaling the matrix. Although it increases the calculation time, it can reduce the performance requirements of computing resources.

[0111] Figure 2 The following is a structural block diagram of the intelligent manufacturing workshop sealing control system provided by an embodiment of the present invention. The intelligent manufacturing workshop sealing control system 10 includes:

[0112] A microprocessor query module 11 is used to query the serial numbers of all production equipment in the manufacturing line and query the microprocessor with a data transmission module in each production equipment;

[0113] A microprocessor selection module 12 is used to select a preset number of microprocessors from all microprocessors with data transmission modules of each production device as target microprocessors, and sort the target microprocessors;

[0114] The matrix establishment module 13 is used to create rows according to the serial numbers of the production equipment and columns according to the serial numbers of the microprocessors to obtain a matrix, and to obtain signals of each target microprocessor at different times based on the matrix timing to obtain a signal matrix containing time tags;

[0115] A matrix recognition module 14, used to adjust the image acquisition accuracy and sealing triggering conditions of the visual inspection instrument installed at the sealing device according to the signal matrix;

[0116] The control instruction generation module 15 is used to detect the product based on the adjusted visual inspection instrument, and when the detection result meets the sealing trigger condition, generate a sealing execution instruction to control the operation of the sealing device;

[0117] The independent variable of the adjustment process is the degree of obviousness of the periodic characteristics of the signal matrix. The higher the degree of obviousness, the lower the image acquisition accuracy and the lower the sealing triggering condition.

[0118] Furthermore, the microprocessor selection module 12 includes:

[0119] A first query unit is used to query the priority of the production equipment; wherein the priority of the production equipment is used to represent the degree of its influence on the production process, and the higher the degree of influence, the higher the priority;

[0120] A first selection unit is used to select production equipment in descending order of priority until the number of production equipment reaches a preset first number threshold as target production equipment;

[0121] A second query unit is used to query the microprocessor with a data transmission module in the target production equipment to obtain the priority of the microprocessor; wherein the priority of the microprocessor is used to characterize the degree of its influence on the production equipment, and the higher the degree of influence, the higher the priority;

[0122] The second selection unit is used to select microprocessors in each target production device in descending order of priority until the number of microprocessors reaches a preset second number threshold as target microprocessors, and sort the target microprocessors according to priority.

[0123] Specifically, the matrix establishment module 13 includes:

[0124] Establish an execution unit for creating rows corresponding to the serial numbers of the production equipment and creating columns corresponding to the serial numbers of the microprocessors, thereby obtaining a matrix; wherein the row and column positions in the matrix correspond to the microprocessors in the production equipment;

[0125] A signal insertion unit, used for acquiring signals of each target microprocessor based on a preset data acquisition frequency, copying the matrix, and inserting the signals into corresponding row and column positions in the copied matrix;

[0126] A label generating unit, used for taking the acquisition time of the signal as the time label of the matrix after the signal is inserted, so as to obtain a signal matrix containing the time label;

[0127] Among them, signal regularization is required before signal insertion. The signal regularization method is as follows:

[0128] Count historical signals, determine the majority signal, compare other signals with the majority signal, and output the comparison value as the regularized signal.

[0129] Furthermore, the matrix identification module 14 includes:

[0130] an arranging unit, for arranging the signal matrix based on the time tags;

[0131] A feature extraction unit, used to extract features from the signal matrix and convert the signal matrix into a single value; wherein there is only one feature extraction scheme, which is applicable to all signal matrices;

[0132] A matrix fitting unit, used for fitting a single value corresponding to each signal matrix based on time to obtain a characteristic change function;

[0133] A cycle identification unit is used to perform Fourier transform on the characteristic change function to obtain the fundamental frequency amplitude and the harmonic wave amplitude, and determine the probability of the existence of the cycle according to the fundamental frequency amplitude and the harmonic wave amplitude. When the probability reaches a preset probability threshold, the cycle is determined according to the fundamental frequency;

[0134] An updating unit is used to determine the image acquisition accuracy and the sealing trigger condition based on the probability; the image acquisition accuracy is inversely proportional to the probability. When the recognition model adopts a convolutional recognition model, the sealing trigger condition adopts a convolution kernel matching number threshold, and the convolution kernel matching number threshold is inversely proportional to the probability.

[0135] The above are only preferred embodiments of the present invention, and are not intended to limit the patent scope of the present invention. Any equivalent structure or equivalent process transformation made using the contents of the present invention specification and drawings, or directly or indirectly applied in other related technical fields, are also included in the patent protection scope of the present invention.

Claims

1. A sealing control method for an intelligent manufacturing workshop, characterized in that: The method comprises: Query the serial numbers of all production equipment in the manufacturing line, and query the microprocessor with a data transmission module in each production equipment; Selecting a preset number of microprocessors from all microprocessors with data transmission modules in each production device as target microprocessors, and sorting the target microprocessors; Create rows according to the number of the production equipment, create columns according to the serial number of the microprocessor, and obtain a matrix. Based on the matrix, obtain the signals of each target microprocessor at different times to obtain a signal matrix containing time tags; According to the signal matrix, the image acquisition accuracy and sealing triggering conditions of the visual inspection device installed at the sealing device are adjusted; The product is inspected by the adjusted visual inspection instrument. When the inspection result meets the sealing trigger condition, a sealing execution instruction is generated to control the operation of the sealing device. The independent variable of the adjustment process is the degree of obviousness of the periodic characteristics of the signal matrix. The higher the degree of obviousness, the lower the image acquisition accuracy and the lower the sealing triggering condition.

2. The sealing control method of the intelligent manufacturing workshop according to claim 1 is characterized in that: The step of selecting a preset number of microprocessors from all microprocessors with data transmission modules in each production device as target microprocessors and sorting the target microprocessors comprises: Query the priority of production equipment; the priority of production equipment is used to represent its impact on the production process. The higher the impact, the higher the priority. Selecting production equipment in descending order of priority until the number of production equipment reaches a preset first number threshold as target production equipment; Query the microprocessor with the data transmission module in the target production equipment to obtain the priority of the microprocessor; wherein the priority of the microprocessor is used to characterize the degree of its influence on the production equipment, and the higher the degree of influence, the higher the priority; In each target production device, microprocessors are selected in descending order of priority until the number of microprocessors reaches a preset second number threshold as target microprocessors, and the target microprocessors are sorted according to the priority.

3. The sealing control method of the intelligent manufacturing workshop according to claim 1 is characterized in that: The steps of creating rows according to the numbers of the production equipment, creating columns according to the serial numbers of the microprocessors, obtaining a matrix, and acquiring signals of each target microprocessor at different times based on the matrix to obtain a signal matrix containing time tags include: Create rows corresponding to the serial numbers of the production equipment one by one, create columns corresponding to the serial numbers of the microprocessors one by one, and obtain a matrix; wherein the row and column positions in the matrix correspond to the microprocessors in the production equipment one by one; Acquire signals from each target microprocessor based on a preset data acquisition frequency, copy the matrix, and insert the signals into corresponding row and column positions in the copied matrix; The acquisition time of the signal is used as the time tag of the matrix after the signal is inserted, and the signal matrix containing the time tag is obtained; Among them, signal regularization is required before signal insertion. The signal regularization method is as follows: Count historical signals, determine the majority signal, compare other signals with the majority signal, and output the comparison value as the regularized signal.

4. The sealing control method of the intelligent manufacturing workshop according to claim 1 is characterized in that: The step of adjusting the image acquisition accuracy and sealing triggering conditions of the visual inspection device installed at the sealing device according to the signal matrix includes: Arrange the signal matrix based on time labels; Performing feature extraction on the signal matrix to convert the signal matrix into a single value; wherein there is only one feature extraction scheme, which is applicable to all signal matrices; Based on time fitting, a single value corresponding to each signal matrix is ​​obtained to obtain a characteristic change function; Perform Fourier transform on the characteristic change function to obtain the fundamental frequency amplitude and the harmonic wave amplitude, determine the probability of the existence of a cycle according to the fundamental frequency amplitude and the harmonic wave amplitude, and when the probability reaches a preset probability threshold, determine the cycle according to the fundamental frequency; The image acquisition accuracy and the sealing trigger condition are determined based on the probability; the image acquisition accuracy is inversely proportional to the probability. When the recognition model adopts a convolutional recognition model, the sealing trigger condition adopts a convolution kernel matching number threshold, and the convolution kernel matching number threshold is inversely proportional to the probability.

5. The sealing control method for an intelligent manufacturing workshop according to claim 4 is characterized in that: The extracting features from the signal matrix and converting the signal matrix into a single value comprises: Querying the priority of the target microprocessor corresponding to each row and column position in the signal matrix, and determining the weight according to the priority; All values ​​in the matrix are summed based on the weights to produce a single value.

6. The sealing control method of the intelligent manufacturing workshop according to claim 5 is characterized in that: The step of extracting features from the signal matrix and converting the signal matrix into a single value also includes: Partitioning the signal matrix to obtain different sub-arrays, and recording the position number of each sub-array; Querying the priority of the target microprocessor corresponding to each row and column position in the subarray, and determining the weight according to the priority; All values ​​of the subarray are summed based on the weights to obtain a single value of the subarray; when fitting a single value, a single value corresponding to the same subarray is fitted according to the position number; when there is a probability of a period, the parameters use the mean of the fundamental frequency amplitude and the mean of the harmonic wave amplitude.

7. A sealing control system for an intelligent manufacturing workshop, characterized in that: The system comprises: A microprocessor query module is used to query the serial numbers of all production equipment in the manufacturing line and query the microprocessor with a data transmission module in each production equipment; A microprocessor selection module is used to select a preset number of microprocessors from all microprocessors with data transmission modules of each production device as target microprocessors, and sort the target microprocessors; The matrix establishment module is used to create rows according to the serial numbers of the production equipment and columns according to the serial numbers of the microprocessors to obtain a matrix, and to obtain the signals of each target microprocessor at different times based on the matrix timing to obtain a signal matrix containing time tags; A matrix recognition module, used to adjust the image acquisition accuracy and sealing triggering conditions of a visual inspection instrument installed at the sealing device according to the signal matrix; A control instruction generation module is used to detect products based on the adjusted visual inspection instrument. When the detection result meets the sealing trigger condition, a sealing execution instruction is generated to control the operation of the sealing device; The independent variable of the adjustment process is the degree of obviousness of the periodic characteristics of the signal matrix. The higher the degree of obviousness, the lower the image acquisition accuracy and the lower the sealing triggering condition.

8. The intelligent manufacturing workshop sealing control system according to claim 7 is characterized in that: The microprocessor selection module includes: A first query unit is used to query the priority of the production equipment; wherein the priority of the production equipment is used to represent the degree of its influence on the production process, and the higher the degree of influence, the higher the priority; A first selection unit is used to select production equipment in descending order of priority until the number of production equipment reaches a preset first number threshold as target production equipment; A second query unit is used to query the microprocessor with a data transmission module in the target production equipment to obtain the priority of the microprocessor; wherein the priority of the microprocessor is used to characterize the degree of its influence on the production equipment, and the higher the degree of influence, the higher the priority; The second selection unit is used to select microprocessors in each target production device in descending order of priority until the number of microprocessors reaches a preset second number threshold as target microprocessors, and sort the target microprocessors according to priority.

9. The intelligent manufacturing workshop sealing control system according to claim 7, characterized in that: The matrix establishment module comprises: Establish an execution unit for creating rows corresponding to the serial numbers of the production equipment and creating columns corresponding to the serial numbers of the microprocessors, thereby obtaining a matrix; wherein the row and column positions in the matrix correspond to the microprocessors in the production equipment; A signal insertion unit, used for acquiring signals of each target microprocessor based on a preset data acquisition frequency, copying the matrix, and inserting the signals into corresponding row and column positions in the copied matrix; A label generating unit, used for taking the acquisition time of the signal as the time label of the matrix after the signal is inserted, so as to obtain a signal matrix containing the time label; Among them, signal regularization is required before signal insertion. The signal regularization method is as follows: Count historical signals, determine the majority signal, compare other signals with the majority signal, and output the comparison value as the regularized signal.

10. The intelligent manufacturing workshop sealing control system according to claim 9, characterized in that: The matrix recognition module comprises: an arranging unit, for arranging the signal matrix based on the time tags; A feature extraction unit, used to extract features from the signal matrix and convert the signal matrix into a single value; wherein there is only one feature extraction scheme, which is applicable to all signal matrices; A matrix fitting unit, used for fitting a single value corresponding to each signal matrix based on time to obtain a characteristic change function; A cycle identification unit is used to perform Fourier transform on the characteristic change function to obtain the fundamental frequency amplitude and the harmonic wave amplitude, and determine the probability of the existence of the cycle according to the fundamental frequency amplitude and the harmonic wave amplitude. When the probability reaches a preset probability threshold, the cycle is determined according to the fundamental frequency; An updating unit is used to determine the image acquisition accuracy and the sealing trigger condition based on the probability; the image acquisition accuracy is inversely proportional to the probability. When the recognition model adopts a convolutional recognition model, the sealing trigger condition adopts a convolution kernel matching number threshold, and the convolution kernel matching number threshold is inversely proportional to the probability.

Images