A batching method and batching system for high-precision material weighing

Abstract

The batching method and batching system for high-precision material weighing provided by the present invention relate to the field of data processing; the method includes: obtaining the weighing values ​​of several standard weight pieces on a weighing device under a preset aging environment, wherein the preset aging environment includes sequentially setting several variables and their parameter values ​​that affect the weighing values; constructing a neural network learning model, dividing the training set and the result set according to the weighing values ​​and the parameter values ​​of several variables corresponding thereto for training, and obtaining a multi-factor weight correction model; obtaining the parameter values ​​of each variable under the working environment of the weighing device, calling the multi-factor weight correction model, and calibrating the weighing values ​​of the current weighing components of the weighing device; obtaining the first difference between the calibration value and the formula value, and when the first difference falls within the error range, recording the calibration value, otherwise, re-weighing. The present invention adopts the method of calibrating the weighing data to reduce the weighing error and ensure product quality in the batching stage.

Description

Technical Field

[0001] The invention relates to the technical field of data processing, and in particular to a batching method and a batching system for high-precision material weighing. Background Art

[0002] With the continuous development of productivity, the modern production process has gradually developed from manual to automated and intelligent. Traditional manual operations are generally not credible or data cannot be effectively verified in modern enterprises due to the large human influence. Therefore, how to integrate digitalization into manual operations to improve quality has become an important issue.

[0003] For enterprises that produce ingredients according to recipes, their traditional manual weighing and recording method of batching is the first object of automation and intelligent transformation. In the traditional production process of enterprises, due to the large number of production batches, in most cases, the recording of ingredient weighing data can only be recorded according to random inspection and self-inspection, and it is impossible to write and record the weight of each batch of materials in actual work; in addition, due to mechanical operation, employees cannot effectively ensure that the weight of each batch of materials is within the tolerance range, and the traceability of data is also poor. With the increasing quality requirements of customer groups, especially foreign companies such as Europe, the United States and Japan, they rely more on automation and dataization for production process control, and they can hardly accept traditional manual records or sampling records. In order to solve the above problems, some enterprises have replaced their production equipment by purchasing a series of automated and intelligent equipment. However, due to the high investment in production equipment, the burden on small enterprises is heavy. Therefore, some enterprises have improved the functions of existing weighing equipment so that the equipment can record weighing data in real time and trace data, which can achieve the purpose of intelligent data acquisition while reducing costs.

[0004] Although the method of improving weighing equipment can achieve the purpose of automatically recording weighing data and tracing it, the accuracy of weighing data gradually decreases during specific applications due to changes in the application environment and usage time of the weighing equipment. When making error judgments, although the displayed data meets the set error threshold, the actual data far exceeds the actual error; therefore, when tracing the quality of the finished product, the data appears to be accurate, but the root cause that actually affects the product quality cannot be found. Summary of the invention

[0005] The purpose of the present invention is to provide a batching method and a batching system for high-precision material weighing, which digitizes the impact of the environment on the weighing equipment into a multi-factor weight correction model through a machine learning method, and then calibrates the data of the weighing equipment in real time, providing accurate data support for finished product quality tracing.

[0006] To achieve the above object, the present invention proposes the following technical solution: a batching method for high-precision material weighing, comprising:

[0007] Obtaining weighing values ​​of a plurality of standard weight pieces on a weighing device under a preset aging environment, wherein the preset aging environment includes sequentially setting a plurality of variables and parameter values ​​that affect the weighing values;

[0008] Construct a neural network learning model, divide the training set and the result set into training sets according to the weighing values ​​and the parameter values ​​of several corresponding variables, so as to obtain a multi-factor weight correction model of each variable for the weighing value of the standard weight piece;

[0009] Obtain the parameter values ​​of various variables under the working environment of the weighing equipment, call the multi-factor weight correction model, calibrate the weighing value of the current weighing component of the weighing equipment, and obtain the calibration value;

[0010] Obtaining a first difference between the calibration value and the corresponding weighed component formula value data, and determining whether the first difference falls within a preset error range;

[0011] When the first difference is within the error range, record the calibration value and proceed to weighing the next formula component; otherwise, add or subtract the weighed components and weigh again until the first difference is within the error range.

[0012] Furthermore, the preset aging environment includes: setting any variable, and when other variables are set to fixed values, obtaining the weighing value of each standard weight piece on the weighing device; adjusting the value of the variable, and when other variables are set to fixed values, respectively obtaining the weighing value of each standard weight piece on the weighing device.

[0013] Furthermore, before the training is performed by dividing the training set and the result set according to the weighing values ​​and the parameter values ​​of the corresponding variables, the method further includes: preprocessing the weighing values ​​and the parameter values ​​of the corresponding variables;

[0014] The preprocessing includes using the standard deviation method to perform outlier processing on the parameter value of any variable and its corresponding weighing value; and using the z-score standardization method to standardize the parameter value of each variable and its corresponding weighing value after the outlier processing.

[0015] Furthermore, it also includes: using SPSS factor analysis method to extract several main variables of several variables that affect the weighing values, and dividing the training set and the result set according to the parameter values ​​of the main variables and their corresponding weighing values; wherein the several variables that affect the weighing values ​​include temperature, humidity, pressure, dust concentration in the environment, environmental acidity, equipment usage time and number of times the equipment is used.

[0016] Furthermore, it also includes: obtaining the second difference between the weighing value of the current weighing component of the weighing equipment and the corresponding weighing component formula value data, recording the weighing values ​​whose first difference belongs to the preset error range and the second difference does not belong to the preset error range, so as to facilitate data traceability.

[0017] Another technical solution of the present invention is to disclose a batching system for high-precision material weighing, the system comprising:

[0018] A first acquisition module is used to obtain the weighing values ​​of a plurality of standard weight pieces on a weighing device under a preset aging environment, wherein the preset aging environment includes sequentially setting a plurality of variables and parameter values ​​that affect the weighing values;

[0019] Construct a training module for constructing a neural network learning model, divide the training set and the result set into training sets according to the weighing values ​​and the parameter values ​​of several variables corresponding thereto, so as to obtain a multi-factor weight correction model of each variable for the weighing value of the standard weight piece;

[0020] The second acquisition module is used to obtain the parameter values ​​of various variables under the working environment of the weighing device, call the multi-factor weight correction model, calibrate the weighing value of the current weighing component of the weighing device, and obtain the calibration value;

[0021] A third acquisition module is used to obtain a first difference between the calibration value and the corresponding weighing component formula value data, and determine whether the first difference falls within a preset error range;

[0022] The judgment module is used to record the calibration value and enter the weighing of the next formula component when the first difference falls within the error range; otherwise, the weighed components are added and subtracted and then weighed again until the first difference falls within the error range.

[0023] Furthermore, it also includes: a preprocessing module, which is used to preprocess the weight value and the parameter values ​​of several variables corresponding thereto;

[0024] The preprocessing includes using the standard deviation method to perform outlier processing on the parameter value of any variable and its corresponding weighing value; and using the z-score standardization method to standardize the parameter value of each variable and its corresponding weighing value after the outlier processing.

[0025] Furthermore, it also includes: a variable analysis module, which is used to extract several main variables of several variables that affect the weighing values ​​using the SPSS factor analysis method, and divide the training set and the result set according to the parameter values ​​of the main variables and their corresponding weighing values; wherein the several variables that affect the weighing values ​​include temperature, humidity, pressure, dust concentration in the environment, environmental acidity, equipment usage time and number of times the equipment is used.

[0026] The present invention proposes a batching system based on MCGS, which includes an MCGS touch screen, a weighing device and a PLC; the MCGS touch screen is preset with the above-mentioned multi-factor weight correction model, and when it obtains the real-time weighing data of the weighing device, it calls the correction model to calibrate the data and displays the calibration data on its operation interface;

[0027] The MCGS touch screen is used to receive the control instruction issued by the PLC when the calibration value of the current weighing component enters the threshold range of the set weight, and record the material number of the current ingredient, the calibration data of each formula component, the total process weight, the package number, and the weighing time;

[0028] The MCGS touch screen is used to automatically record the material number of the current ingredient, the calibration data of each formula component, the total process weight, the package number, and the weighing time when the calibration value of the current weighing component enters the threshold range of the set weight and the actual weighing value does not fall within the threshold range of the set weight.

[0029] The present invention also discloses an electronic device, which includes a memory, a processor, and a computer program stored in the memory and executable on the processor. When the processor runs the computer program, the batching method for high-precision material weighing as described above is implemented.

[0030] It can be seen from the above technical solutions that the technical solutions of the present invention have the following beneficial effects:

[0031] The batching method and batching system for high-precision material weighing disclosed in the present invention include: obtaining the weighing values ​​of several standard weight pieces on a weighing device under a preset aging environment, wherein the preset aging environment includes sequentially setting several variables and their parameter values ​​that affect the weighing values; constructing a neural network learning model, dividing the training set and the result set according to the weighing values ​​and the parameter values ​​of several variables corresponding thereto for training, and obtaining a multi-factor weight correction model; obtaining the parameter values ​​of each variable under the working environment of the weighing device, calling the multi-factor weight correction model, and calibrating the weighing values ​​of the current weighing components of the weighing device; obtaining the first difference between the calibration value and the formula value, and judging whether the first difference belongs to the preset error range; when the first difference belongs to the error range, recording the calibration value, and entering the weighing of the next formula component; otherwise, re-weighing. The present invention reduces the weighing error by calibrating the real-time weighing data, and ensures the product quality in the batching stage.

[0032] In addition, the present invention also records the weighing records of calibration data within the error range and weighing data not within the error range for later product quality analysis data tracing; and by analyzing such recorded data, the data accuracy of the current weighing equipment can be generally understood so that the equipment can be replaced in time.

[0033] It should be appreciated that all combinations of the foregoing concepts, as well as additional concepts described in greater detail below, may be considered to be part of the inventive subject matter of the present disclosure, provided such concepts are not mutually inconsistent.

[0034] The foregoing and other aspects, embodiments and features of the present invention can be more fully understood from the following description in conjunction with the accompanying drawings. Other additional aspects of the present invention, such as the features and / or beneficial effects of the exemplary embodiments, will be apparent from the following description or learned from the practice of the specific embodiments according to the present invention. BRIEF DESCRIPTION OF THE DRAWINGS

[0035] The accompanying drawings are not drawn to scale according to actual reference objects. In the accompanying drawings, each identical or nearly identical component shown in various figures may be represented by the same reference numeral. For the sake of clarity, not every component is labeled in each figure. Now, embodiments of various aspects of the present invention will be described by way of example and with reference to the accompanying drawings, in which:

[0036] Figure 1 The implementation process of the batching method of the present invention is Figure 1 ;

[0037] Figure 2 The implementation process of the batching method of the present invention is Figure 2 ;

[0038] Figure 3 It is a composition structure diagram of the batching system of the present invention;

[0039] Figure 4 This is a composition structure diagram of the MCGS-based batching system of the present invention. DETAILED DESCRIPTION

[0040] In order to make the purpose, technical solution and advantages of the embodiment of the present invention clearer, the technical solution of the embodiment of the present invention will be clearly and completely described below in conjunction with the drawings of the embodiment of the present invention. Obviously, the described embodiment is a part of the embodiment of the present invention, not all of the embodiments. Based on the described embodiment of the present invention, all other embodiments obtained by ordinary technicians in this field without creative work are within the scope of protection of the present invention. Unless otherwise defined, the technical terms or scientific terms used herein should be the common meaning understood by people with general skills in the field to which the present invention belongs.

[0041] The words "first", "second" and similar words used in the patent application specification and claims of the present invention do not indicate any order, quantity or importance, but are only used to distinguish different components. Similarly, unless the context clearly indicates otherwise, the singular forms of "a", "an" or "the" and other similar words do not indicate a quantitative limitation, but indicate the existence of at least one. Words such as "include" or "comprise" mean that the elements or objects appearing before "include" or "comprise" cover the features, wholes, steps, operations, elements and / or components listed after "include" or "comprise", and do not exclude the existence or addition of one or more other features, wholes, steps, operations, elements, components and / or their collections. "Up", "down", "left", "right" and the like are only used to indicate relative positional relationships. When the absolute position of the described object changes, the relative positional relationship may also change accordingly.

[0042] Based on the current method of improving the weighing equipment so that it can only record real-time weighing data within the allowable error range, although it can achieve the purpose of automatically recording weighing data and tracing, it does not take into account the influence of the working environment and the equipment itself on the weighing data, resulting in the historical record data not being able to provide any reference for the real reasons affecting product quality when tracing the product later. In view of the above problems, the present invention proposes a batching method and batching system for high-precision material weighing, which considers the influence of the environment and the equipment itself on the weighing data when the weighing equipment is working, pre-calculates a multi-factor weight correction model, corrects the real-time weighing data, and obtains accurate weighing results; in addition, through the synchronous recording of abnormal weighing data and calibration data, a reference basis is provided for historical tracing.

[0043] The following is a further detailed introduction to the batching method and batching system for high-precision material weighing disclosed in the present invention in conjunction with the embodiments shown in the accompanying drawings.

[0044] Combination Figure 1 and Figure 2 As shown, the batching method for high-precision material weighing disclosed in this embodiment includes the following steps:

[0045] Step S102, obtaining weighing values ​​of a plurality of standard weight pieces on a weighing device under a preset aging environment, wherein the preset aging environment includes sequentially setting a plurality of variables and parameter values ​​that affect the weighing values;

[0046] Factors that affect the accuracy of weighing equipment data include environmental factors and equipment factors, among which environmental factors include temperature, humidity, pressure, dust concentration in the environment, environmental acidity, etc., and equipment factors include equipment usage time and the number of times the equipment is used. The method of obtaining data in the preset aging environment is to determine the only variable, that is, including: setting any variable, and obtaining the weighing value of each standard weight on the weighing equipment when other variables are set to fixed values; adjusting the value of the variable, and obtaining the weighing value of each standard weight on the weighing equipment when other variables are set to fixed values. Each standard weight can be selected as a free combination of 50g, 100g, 500g or 1000g. Optionally, the weighing value can be the deviation between the real-time weighing value of each standard weight in each aging environment and the standard weight.

[0047] Step S104, constructing a neural network learning model, dividing the training set and the result set for training according to the weighing value and the parameter values ​​of several variables corresponding thereto, so as to obtain a multi-factor weight correction model of each variable for the weighing value of the standard weight piece;

[0048] During implementation, the neural network learning model selects the CNN neural network model for training, and the data obtained in step S102 are divided into a training set and a result using a 7:3 division ratio. Optionally, the obtained weighing values ​​and the corresponding parameter values ​​of several variables need to be preprocessed before being used as training data, such as using the standard deviation method to obtain the parameter value under any variable and its corresponding weighing value for outlier processing, and using the z-score standardization method to standardize the parameter values ​​of each variable after outlier processing and its corresponding weighing value, that is, step S1031. The purpose of the preprocessing process is to remove abnormal data and reduce the impact of erroneous data on the training model.

[0049] Step S106, obtaining the parameter values ​​of various variables under the working environment of the weighing device, calling the multi-factor weight correction model, calibrating the weighing value of the current weighing component of the weighing device, and obtaining the calibration value;

[0050] The acquisition of environmental variable parameter values ​​in the equipment working environment adopts the method of setting up sensors, such as setting up temperature and humidity sensors to measure environmental temperature and humidity, setting up barometers to measure environmental pressure, and setting up acid gas sensors to measure environmental acidity. The equipment use time and number of times in the equipment factors can be obtained based on the equipment use records. In this scheme, the calibration data after the weighing value is corrected by the multi-factor weight correction model is recorded as the accurate data of the current weighing component weight, and inevitably, the weighing equipment used for the current weighing component is the same model as the weighing equipment used for the above training to obtain the multi-factor weight correction model data.

[0051] Step S108, obtaining a first difference between the calibration value and the corresponding weighed component formula value data, and determining whether the first difference falls within a preset error range;

[0052] Step S110, when the first difference is within the error range, record the calibration value and proceed to weighing the next formula component; otherwise, add or subtract the weighed components and weigh again until the first difference is within the error range.

[0053] During implementation, the method allows recording of the calibration value only when the first difference falls within the calibration value within the error range; therefore, optionally, the method also includes: accepting a control instruction to determine the record, and when the first difference does not fall within the error range, issuing an alarm and not allowing recording; this process is intended to avoid problems such as employees failing to record truthfully as required, fabricating data, and recording inaccurate data.

[0054] As an optional implementation, although there are multiple factor variables that affect the weighing data, the degree of influence of each variable on the test results is different. For example, some variables contribute more than 99% of the influence on the test results, and the influence of some variables on the test results is almost negligible. Therefore, they should be excluded in the variable analysis; therefore, after obtaining and preprocessing the data, this method uses SPSS factor analysis to extract several main variables of several variables that affect the weighing values, and divides the training set and the result set according to the parameter values ​​of the main variables and their corresponding weighing values, that is, step S1032; in this scheme, the main variables obtained by SPSS factor analysis include 5, temperature, humidity, dust concentration in the environment, equipment usage time and number of times equipment is used.

[0055] As an optional implementation, the batching method also includes: obtaining the second difference between the weighing value of the current weighing component of the weighing device and the corresponding weighing component formula value data, and recording the weighing values ​​whose first difference belongs to the preset error range and whose second difference does not belong to the preset error range, so as to facilitate data traceability. Such data whose calibration values ​​meet the weighing requirements and whose actual weighing data do not meet the weighing requirements are the main data that cause batching errors and may affect product quality. Therefore, the method clearly records such data, and can provide effective data support for later product quality traceability and quality control analysis.

[0056] As an optional implementation, the batching method disclosed in the embodiment further includes: constructing a cloud storage, and when the calibration value is recorded, the recorded data is synchronously uploaded to the cloud storage. The cloud storage data is used as a backup when the local storage data cannot be found.

[0057] The batching method for high-precision material weighing disclosed in the present invention solves the problem that weighing data is inaccurate and affects product quality due to changes in the working environment and equipment usage of the weighing equipment. A multi-factor weight correction model of the weighing equipment is trained through machine learning, and the model is used to calibrate the weighing value in real time when the equipment is in use to obtain a calibration value. Although there is an error between the calibration value and the true value, the error is significantly smaller than the error between the weighing value and the true value. Batching is performed based on the calibration value, effectively ensuring product quality in the batching stage.

[0058] In an embodiment of the present invention, an electronic device is also provided, which includes a memory, a processor, and a computer program stored in the memory and executable on the processor. When the processor runs the computer program, the batching method for high-precision material weighing in the above embodiment is implemented.

[0059] The above program can be run in a processor, or it can be stored in a memory, that is, a computer-readable medium, which includes permanent and non-permanent, removable and non-removable media. Information storage can be achieved by any method or technology. Information can be computer-readable instructions, data structures, program modules or other data. Examples of computer storage media include, but are not limited to, phase change memory (PRAM), static random access memory (SRAM), dynamic random access memory (DRAM), other types of random access memory (RAM), read-only memory (ROM), electrically erasable programmable read-only memory (EEPROM), flash memory or other memory technology, read-only compact disk read-only memory (CD-ROM), digital versatile disk (DVD) or other optical storage, magnetic cassettes, magnetic tape magnetic disk storage or other magnetic storage devices or any other non-transmission media that can be used to store information that can be accessed by a computing device. As defined in this article, computer-readable media does not include temporary computer-readable media, such as modulated data signals and carrier waves.

[0060] These computer programs can also be loaded onto a computer or other programmable data processing device so that a series of operating steps are executed on the computer or other programmable device to produce a computer-implemented process, so that the instructions executed on the computer or other programmable device provide instructions for implementing the process. Figure 1 A process or multiple processes and / or boxes Figure 1 The steps of the functions specified in a block or multiple blocks may be implemented by different modules corresponding to different method steps.

[0061] In this embodiment, such a device or system is provided, which can be called a batching system for high-precision material weighing, such as Figure 3As shown, the system includes: a first acquisition module, which is used to obtain the weighing values ​​of several standard weight pieces on the weighing equipment under a preset aging environment, wherein the preset aging environment includes setting several variables and their parameter values ​​that affect the weighing values ​​in sequence; a training module, which is used to construct a neural network learning model, and divide the training set and the result set according to the weighing values ​​and the parameter values ​​of several variables corresponding to them for training, so as to obtain a multi-factor weight correction model for the weighing values ​​of the standard weight pieces of each variable; a second acquisition module, which is used to obtain the parameter values ​​of each variable under the working environment of the weighing equipment, call the multi-factor weight correction model, calibrate the weighing values ​​of the current weighing components of the weighing equipment, and obtain the calibration value; a third acquisition module, which is used to obtain the first difference between the calibration value and the corresponding weighing component formula value data, and judge whether the first difference belongs to the preset error range; a judgment module, which is used to record the calibration value when the first difference belongs to the error range, and enter the weighing of the next formula component; otherwise, the weighing component is re-weighed after addition and subtraction until the first difference belongs to the error range.

[0062] The steps of the system for implementing the batching method of high-precision material weighing disclosed in the above embodiment have been described and will not be repeated here.

[0063] For example, the process of the first acquisition module presetting the aging environment is to set any variable, and when other variables are set to fixed values, obtain the weighing values ​​of each standard weight piece on the weighing device; adjust the value of the variable, and when other variables are set to fixed values, respectively obtain the weighing values ​​of each standard weight piece on the weighing device. That is, in this process, a single variable is maintained.

[0064] As an optional embodiment, the batching system also includes: a preprocessing module for preprocessing the weighing value and the parameter values ​​of several variables corresponding thereto; the preprocessing process includes using the standard deviation method to process the outliers of the parameter values ​​obtained under any variable and its corresponding weighing value; and using the z-score standardization method to standardize the parameter values ​​of each variable and its corresponding weighing value after the outlier processing. A variable analysis module is used to extract several main variables of several variables that affect the weighing value by using the SPSS factor analysis method, and divide the training set and the result set by the parameter values ​​of the main variables and their corresponding weighing values; wherein the several variables that affect the weighing value include temperature, humidity, pressure, dust concentration in the environment, environmental acidity, equipment usage time and equipment usage times.

[0065] Combination Figure 4As shown, the present invention also proposes a batching system based on MCGS, which includes a MCGS touch screen, a weighing device and a PLC; when applied, when the above-mentioned batching method is applied in the system, there is no need to perform steps S102 and S104, and a multi-factor weight correction model pre-trained for the same type of weighing device is directly set in the MCGS touch screen, and then when the MCGS touch screen obtains the real-time weighing data of the weighing device, the correction model is directly called to perform data calibration in combination with the real-time main variable parameter value collected by the data acquisition device, and then the calibration data is displayed on its operation interface.

[0066] Specifically, the weighing equipment uses the RS232 communication interface to communicate with the MCGS touch screen to obtain the weighing value of the weighing equipment in real time; the staff pre-enters the composition of each formula in the parameter setting interface of the MCGS touch screen, that is, the operation interface, including the component name, the corresponding weight of each component, the threshold range corresponding to the weight of each component, the number of packages, and the planned number. The preset script program calculates, records and determines the obtained weighing values; the preset determination conditions are used for result determination. For example, for component A in formula 1, the calibration data of component A can only be recorded when the difference between the "current component A calibration value" and the "set weight value" is less than 5g. Otherwise, recording is not allowed and an out-of-tolerance alarm pop-up window will appear on the MCGS touch screen. The MCGS touch screen realizes human-computer interaction, and its built-in recipe configuration storage function can be used by staff to add, delete, modify and call recipes on the screen; the above recipe editing function is realized by adopting the built-in "recipe configuration design" function of the MCGS touch screen. Staff store commonly used recipes in the memory of the MCGS touch screen in advance. When editing, they can call the "recipe component" through the "recipe management" button on the operation interface, and can also use the "edit recipe" button to add, delete and modify the recipe stored internally.

[0067] When in use, when the staff calls a formula and weighs each component according to the formula requirements, when the calibration value calculated by the specified component weighing enters the threshold range of the set weight, the staff can record the current weighing component results through the control command issued by the PLC through the "Component Confirmation" button on the MCGS touch screen, that is, the above-mentioned first difference is within the threshold range of the set weight. The record content includes the material number, calibration data of each formula component, total process weight, package number, and weighing time; if it does not meet the requirements, an alarm prompts correction. When a batch is weighed, all weighing data is stored on the MCGS touch screen and automatically enters the weighing of the next batch.

[0068] In addition, the MCGS touch screen also stores data in situations where the calibration value of the component currently being weighed by the staff member enters the threshold range of the set weight, and the actual weighing value does not fall within the threshold range of the set weight, that is, the above-mentioned first difference is within the threshold range of the set weight, and the second difference is outside the threshold range of the set weight. The MCGS touch screen automatically records the material number of the current ingredient, the calibration data of each formula component, the total process weight, the package number, and the weighing time; this process is automatically implemented without confirmation by the staff.

[0069] When the operator completes the weighing of the last component in the formula, the script program will automatically reset the "Current Component Real-time Weighing" information on the operation interface to zero, and automatically add 1 to the "Current Number of Packages"; at the same time, the material number of the previous package, the actual weight of each component, the package number, the weighing time and other information will be automatically recorded.

[0070] The data recorded and stored on the MCGS touch screen can be viewed directly or exported via a USB flash drive. When viewing, the current display interface of the MCGS touch screen can be operated to the historical data interface, and the staff can view the historical record data based on a selected formula or a certain time period. Similarly, when exporting, the staff can export data based on a selected formula or a certain time period, and execute several controls for the export function. The export format is an EXCLE document, which is convenient for viewing on a computer, or can be imported into other data management software for data analysis.

[0071] Although the present invention has been disclosed as above with preferred embodiments, it is not intended to limit the present invention. A person with ordinary knowledge in the technical field to which the present invention belongs may make various changes and modifications without departing from the spirit and scope of the present invention. Therefore, the protection scope of the present invention shall be determined by the definition of the claims.

Claims

1. A batching method for high-precision material weighing, characterized in that: include: Obtaining the weighing values ​​of several standard weight pieces on the weighing device under a preset aging environment, wherein the preset aging environment includes sequentially setting several variables and parameter values ​​that affect the weighing values, specifically including: setting any variable, and when other variables are set to fixed values, obtaining the weighing values ​​of each standard weight piece on the weighing device; adjusting the value of the variable, and when other variables are set to fixed values, respectively obtaining the weighing values ​​of each standard weight piece on the weighing device; Construct a neural network learning model, divide the training set and the result set into training sets according to the weighing values ​​and the parameter values ​​of several corresponding variables, so as to obtain a multi-factor weight correction model of each variable for the weighing value of the standard weight piece; Obtain the parameter values ​​of various variables under the working environment of the weighing equipment, call the multi-factor weight correction model, calibrate the weighing value of the current weighing component of the weighing equipment, and obtain the calibration value; Obtaining a first difference between the calibration value and the corresponding weighed component formula value data, and determining whether the first difference falls within a preset error range; When the first difference is within the error range, record the calibration value and proceed to weigh the next formula component; otherwise, add or subtract the weighed components and weigh again until the first difference is within the error range; Before the training is performed by dividing the training set and the result set according to the weighing values ​​and the parameter values ​​of the corresponding variables, the method further includes: preprocessing the weighing values ​​and the parameter values ​​of the corresponding variables; The preprocessing includes using the standard deviation method to perform outlier processing on the parameter value of any variable and its corresponding weighing value; and using the z-score standardization method to standardize the parameter value of each variable and its corresponding weighing value after the outlier processing.

2. The batching method for high-precision material weighing according to claim 1, characterized in that: Also includes: SPSS factor analysis method is used to extract several main variables of several variables that affect the weighing values, and the training set and the result set are divided according to the parameter values ​​of the main variables and their corresponding weighing values; among them, several variables that affect the weighing values ​​include temperature, humidity, pressure, dust concentration in the environment, environmental acidity, equipment usage time and number of times the equipment is used.

3. The batching method for high-precision material weighing according to claim 2, characterized in that: Also includes: Obtain the second difference between the weighing value of the current weighing component of the weighing device and the corresponding weighing component formula value data, and record the weighing values ​​whose first difference belongs to the preset error range and whose second difference does not belong to the preset error range for data traceability.

4. A batching system for high-precision material weighing, characterized in that: include: The first acquisition module is used to obtain the weighing values ​​of several standard weight pieces on the weighing device under a preset aging environment, wherein the preset aging environment includes sequentially setting several variables and parameter values ​​that affect the weighing values, specifically including: setting any variable, and when other variables are set to fixed values, obtaining the weighing values ​​of each standard weight piece on the weighing device; adjusting the value of the variable, and when other variables are set to fixed values, respectively obtaining the weighing values ​​of each standard weight piece on the weighing device; Construct a training module for constructing a neural network learning model, divide the training set and the result set into training sets according to the weighing values ​​and the parameter values ​​of several variables corresponding thereto, so as to obtain a multi-factor weight correction model of each variable for the weighing value of the standard weight piece; The second acquisition module is used to obtain the parameter values ​​of various variables under the working environment of the weighing device, call the multi-factor weight correction model, calibrate the weighing value of the current weighing component of the weighing device, and obtain the calibration value; A third acquisition module is used to obtain a first difference between the calibration value and the corresponding weighing component formula value data, and determine whether the first difference falls within a preset error range; A judgment module is used to record the calibration value and enter the weighing of the next formula component when the first difference value falls within the error range; otherwise, the weighed components are added and subtracted and then weighed again until the first difference value falls within the error range; Also includes: A preprocessing module, used for preprocessing the weight value and the parameter values ​​of several variables corresponding thereto; The preprocessing includes using the standard deviation method to perform outlier processing on the parameter value of any variable and its corresponding weighing value; and using the z-score standardization method to standardize the parameter value of each variable and its corresponding weighing value after the outlier processing.

5. The batching system for high-precision material weighing according to claim 4 is characterized in that: Also includes: The variable analysis module is used to extract several main variables of several variables that affect the weighing values ​​using the SPSS factor analysis method, and divide the training set and the result set according to the parameter values ​​of the main variables and their corresponding weighing values; among them, the several variables that affect the weighing values ​​include temperature, humidity, pressure, dust concentration in the environment, environmental acidity, equipment usage time and number of times the equipment is used.

6. A batching system based on MCGS, characterized in that: It includes an MCGS touch screen, a weighing device and a PLC; the MCGS touch screen is preset with the batching method of high-precision material weighing according to any one of claims 1 to 3, and when it obtains the real-time weighing data of the weighing device, it calls the multi-factor weight correction model to calibrate the data, and displays the calibration data on its operation interface; The MCGS touch screen is used to receive the control instruction issued by the PLC when the calibration value of the current weighing component enters the threshold range of the set weight, and record the material number of the current ingredient, the calibration data of each formula component, the total process weight, the package number, and the weighing time; The MCGS touch screen is used to automatically record the material number of the current ingredient, the calibration data of each formula component, the total process weight, the package number, and the weighing time when the calibration value of the current weighing component enters the threshold range of the set weight and the actual weighing value does not fall within the threshold range of the set weight.

7. An electronic device comprising a memory, a processor, and a computer program stored in the memory and executable on the processor, wherein when the processor runs the computer program, the batching method for high-precision material weighing as described in any one of claims 1 to 3 is implemented.

Images