Product data processing method, device, electronic device and storage medium

Abstract

Embodiments of the present invention disclose a product data processing method, device, electronic device, and storage medium, including: calculating a weight discount coefficient path and a moisture change path for historical batches of products based on historical product production data; establishing a first correlation model based on the weight discount coefficient path for historical batches of products, and establishing a second correlation model based on the moisture change path for historical batches of products; determining a target product production sample data set based on the product historical production data; calculating a target weight discount coefficient path and a target moisture change path based on the target product production sample data set, the first correlation model, and the second correlation model; calculating a target process product weight based on the product feed weight and the target weight discount coefficient path, and calculating the target process product moisture based on the target finished product moisture and the target moisture change path. The technical solutions of the embodiments of the present invention can improve the accuracy, stability, and practicality of product standard data such as product weight and moisture.

Description

Technical Field

[0001] Embodiments of the present invention relate to the technical field of product process data processing, and in particular, to a product data processing method, device, electronic device, and storage medium. Background Art

[0002] In the product processing industry, especially for food processing, controlling product weight and moisture is crucial. Process engineers must establish weight and moisture standards for each step along the entire process path based on actual production conditions. These standards, combined with real-time process control, constrain parameters along the production path to ensure that final product quality remains within expectations.

[0003] Typically, when formulating product weight and moisture process standards, process personnel need to rely heavily on manual experience. That is, in the prior art, process personnel formulate product weight and moisture process standards based on their work experience.

[0004] In the process of realizing the present invention, the inventors discovered that the existing technology has the following defects: relying on manual experience to formulate product weight and moisture process standards, it is often impossible to efficiently analyze and utilize historical batch information of the product, and it also ignores the measurement differences in the moisture and weight measuring instrument calibration data, resulting in the formulated process standards being unstable and highly random, and losing the standard reference significance. Ultimately, the final product quality data obtained by the operator in accordance with the latest formulated process standards deviates from the expected value, resulting in the frequent re-adjustment of the process standards. Summary of the Invention

[0005] Embodiments of the present invention provide a product data processing method, device, electronic device, and storage medium, which can improve the accuracy, stability, and practicality of product standard data such as product weight and moisture.

[0006] According to one aspect of the present invention, a product data processing method is provided, comprising:

[0007] Calculate the weight loss coefficient path and moisture change path of historical batch products based on the historical production data of the product;

[0008] Establishing a first correlation model based on the weight reduction coefficient path of the historical batch products, and establishing a second correlation model based on the moisture change path of the historical batch products;

[0009] Determining a target product production sample data set based on the product historical production data; wherein the target product production sample data set includes a target weight reduction coefficient mean vector and a target moisture change mean vector;

[0010] Calculating a target weight reduction coefficient path and a target moisture change path according to the target product production sample data set, the first association model, and the second association model;

[0011] The target process product weight is calculated based on the product feed weight and the target weight reduction coefficient path, and the target process product moisture is calculated based on the target finished product moisture and the target moisture change path.

[0012] According to another aspect of the present invention, there is provided a product data processing device, comprising:

[0013] The first path calculation module is used to calculate the weight reduction coefficient path and moisture change path of historical batch products based on the historical production data of the product;

[0014] A correlation model establishment module, configured to establish a first correlation model based on the weight reduction coefficient path of the historical batch products, and to establish a second correlation model based on the moisture change path of the historical batch products;

[0015] A sample data set determination module is used to determine a target product production sample data set based on the product historical production data; wherein the target product production sample data set includes a target weight reduction coefficient mean vector and a target moisture change mean vector;

[0016] A second path calculation module is used to calculate a target weight reduction coefficient path and a target moisture change path according to the target product production sample data set, the first correlation model and the second correlation model;

[0017] The weight moisture calculation module is used to calculate the target process product weight based on the product feed weight and the target weight reduction coefficient path, and to calculate the target process product moisture based on the target finished product moisture and the target moisture change path.

[0018] According to another aspect of the present invention, an electronic device is provided, comprising:

[0019] at least one processor; and

[0020] a memory communicatively connected to the at least one processor; wherein,

[0021] The memory stores a computer program that can be executed by the at least one processor. The computer program is executed by the at least one processor to enable the at least one processor to execute the product data processing method according to any embodiment of the present invention.

[0022] According to another aspect of the present invention, a computer-readable storage medium is provided, wherein the computer-readable storage medium stores computer instructions, and the computer instructions are used to enable a processor to implement the product data processing method according to any embodiment of the present invention when executed.

[0023] The embodiment of the present invention calculates the weight loss coefficient path and moisture change path of historical batch products based on the historical production data of the product, so as to establish a first association model based on the weight loss coefficient path of the historical batch products, and establish a second association model based on the moisture change path of the historical batch products. After the model is established, a target product production sample data set including a target weight loss coefficient mean vector and a target moisture change mean vector is determined based on the historical production data of the product, so as to calculate the target weight loss coefficient path and the target moisture change path based on the target product production sample data set, the first association model and the second association model, and finally calculate the target process product weight in combination with the product feed weight and the target weight loss coefficient path, and calculate the target process product moisture in combination with the target finished product moisture and the target moisture change path, thereby solving the problems of low accuracy, poor stability and practicality in the existing determination of product weight and moisture standards based on manual experience, and improving the accuracy, stability and practicality of product standard data such as product weight and moisture.

[0024] It should be understood that the content described in this section is not intended to identify the key or important features of the embodiments of the present invention, nor is it intended to limit the scope of the present invention. Other features of the present invention will become readily understood through the following description. BRIEF DESCRIPTION OF THE DRAWINGS

[0025] In order to more clearly illustrate the technical solutions in the embodiments of the present invention, the following briefly introduces the drawings required for use in the description of the embodiments. Obviously, the drawings described below are only some embodiments of the present invention. For ordinary technicians in this field, other drawings can be obtained based on these drawings without creative work.

[0026] Figure 1 This is a flow chart of a product data processing method provided by the first embodiment of the present invention;

[0027] Figure 2 This is a flow chart of a product data processing method provided by the second embodiment of the present invention;

[0028] Figure 3 is a schematic diagram of a product data processing device provided by Embodiment 3 of the present invention;

[0029] Figure 4 This is a structural diagram of an electronic device provided in Example 4 of the present invention. DETAILED DESCRIPTION

[0030] In order to enable those skilled in the art to better understand the solutions of the present invention, the technical solutions in the embodiments of the present invention will be clearly and completely described below in conjunction with the drawings in the embodiments of the present invention. Obviously, the embodiments described are only part of the embodiments of the present invention, not all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by ordinary technicians in this field without making creative efforts should fall within the scope of protection of the present invention.

[0031] It should be noted that the terms "first", "second", etc. in the description and claims of the present invention and the above-mentioned drawings are used to distinguish similar objects and are not necessarily used to describe a specific order or sequence. It should be understood that the numbers used in this way can be interchanged where appropriate, so that the embodiments of the present invention described herein can be implemented in an order other than those illustrated or described herein. In addition, the terms "including" and "having" and any variations thereof are intended to cover non-exclusive inclusions. For example, a process, method, system, product or device that includes a series of steps or units is not necessarily limited to those steps or units clearly listed, but may include other steps or units that are not clearly listed or inherent to these processes, methods, products or devices.

[0032] Example 1

[0033] Figure 1 This is a flowchart of a product data processing method provided by the first embodiment of the present invention. This embodiment is applicable to the case where a correlation model is constructed based on the historical production data of a product to calculate the weight and moisture standard data of the product. The method can be executed by a product data processing device, which can be implemented by software and / or hardware and can generally be integrated into an electronic device. The electronic device can be a terminal device or a server device. The embodiment of the present invention does not limit the specific device type of the electronic device. Accordingly, if Figure 1 As shown, the method includes the following operations:

[0034] S110. Calculate the weight reduction coefficient path and moisture change path of historical batches of products based on historical production data of the products.

[0035] Among them, the historical production data of the product can be the production data of multiple historical batches of the product. Optionally, the historical production data of the product can include relevant data on the product weight and product moisture of each process in multiple historical batches of the product. The weight reduction coefficient can be the relative relationship between the product weights of each process in the historical production data of a batch of products. For example, assuming that the weight of the product of process 1 is A and the weight of the product of process 2 is B, the weight reduction coefficient between process 1 and process 2 can be B / A. The weight reduction coefficient path can be a path composed of the weight reduction coefficients between each process in the historical production data of a batch of products. The moisture change can be the moisture change between each process in the historical production data of a batch of products. For example, assuming that the moisture content of the product of process 1 is A and the moisture content of the product of process 2 is B, the moisture change between process 1 and process 2 can be AB or BA. The moisture change path can be a path composed of the moisture change between each process in the historical production data of a batch of products.

[0036] In embodiments of the present invention, production data from a sufficient number of historical batches of products can be obtained as historical product production data. Based on this historical product production data, the weight reduction coefficient path and moisture change path corresponding to each historical batch of products can be calculated. Optionally, product types can include, but are not limited to, tobacco, food, and other products requiring weight and moisture process standards. Embodiments of the present invention do not limit specific product types.

[0037] For example, suppose the historical production data for a batch of products involves three processes, where the weight loss coefficient between processes 1 and 2 is A, the weight loss coefficient between processes 2 and 3 is B, and the weight loss coefficient between processes 3 and 4 is C. The weight loss coefficient path corresponding to this historical batch of products can be ABC. Alternatively, the weight loss coefficient path can be represented in the form of a vector. For example, in the example above, the weight loss coefficient path corresponding to this historical batch of products can be (A, B, C).

[0038] For example, suppose the historical production data for a batch of products involves three processes. The moisture change between processes 1 and 2 is A, the moisture change between processes 2 and 3 is B, and the moisture change between processes 3 and 4 is C. The moisture change path corresponding to this historical batch of products can be ABC. Alternatively, the moisture change path can be represented as a vector. For example, in the example above, the moisture change path corresponding to this historical batch of products can be (A, B, C).

[0039] S120: Establish a first correlation model based on the weight reduction coefficient path of the historical batch products, and establish a second correlation model based on the moisture change path of the historical batch products.

[0040] The first correlation model can be used to calculate the weight of the product, and the second correlation model can be used to calculate the moisture change of the product.

[0041] Correspondingly, after arranging and calculating enough historical production data of the product to obtain the weight loss coefficient path and moisture change path of the historical batch products, the calculated weight loss coefficient path of the historical batch products can be used to establish a first association model, and the moisture change path of the historical batch products can be used to establish a second association model.

[0042] Optionally, the model types of the first association model and the second association model can be selected based on the amount of data of the weight coefficient path and the moisture change path. For example, if the number of samples of the weight coefficient path and the moisture change path is large enough, the first association model and the second association model can be established by training neural network models such as LSTM (Long Short Term Memory) and RNN (Recurrent Neural Networks). If the number of samples of the weight coefficient path and the moisture change path is small, the first association model and the second association model can be established by using models such as ARX (Auto-Regressive with Extra Inputs), linear regression or random forest. Therefore, by selecting different types of models to establish the first association model and the second association model according to the number of samples of the weight coefficient path and the moisture change path, the accuracy of the first association model and the second association model can be guaranteed.

[0043] S130. Determine a target product production sample data set based on the product historical production data; wherein the target product production sample data set includes a target weight reduction coefficient mean vector and a target moisture change mean vector.

[0044] The target product production sample dataset may be a product production sample dataset used to input into the association model to calculate product standard data such as product weight and product moisture. Optionally, the target product production sample dataset may include multiple samples. It is understandable that the weight loss coefficient path and moisture change path of a historical batch may be used as one of the product production sample datasets. The target weight loss coefficient mean vector may be the weight loss coefficient mean vector calculated from the weight loss coefficient paths corresponding to each product production sample dataset. The target moisture change mean vector may be the moisture change mean vector calculated from the moisture change paths corresponding to each product production sample dataset.

[0045] For example, when the historical production data of a product includes the weight loss coefficient paths and moisture change paths of 20 batches, the weight loss coefficient paths and moisture change paths of the last three batches can be selected as reference sample data sets, and the reference sample data sets can be sorted to generate the target product production sample data set. Assuming that the weight loss coefficient path of the first batch in the reference sample data set is: A1-B1-C1, the weight loss coefficient path of the second batch is: A2-B2-C2, and the weight loss coefficient path of the third batch is: A3-B3-C3, then the target weight loss coefficient mean vector can be: AM-BM-CM. Among them, AM is the mean of A1, A2 and A3, BM is the mean of B1, B2 and B3, and CM is the mean of C1, C2 and C3. Similarly, assuming that the moisture change path of the first batch in the reference sample dataset is: a1-b1-c1, the moisture change path of the second batch is: a2-b2-c2, and the moisture change path of the third batch is: a3-b3-c3, then the target moisture change mean vector can be: aM-bM-cM. Here, aM is the mean of a1, a2, and a3, bM is the mean of b1, b2, and b3, and cM is the mean of c1, c2, and c3.

[0046] S140: Calculate a target weight reduction coefficient path and a target moisture change path based on the target product production sample data set, the first association model, and the second association model.

[0047] The target weight reduction coefficient path may be a path formed by calculating the weight reduction coefficient standards of the product between each process. The target moisture change path may be a path formed by calculating the moisture change standards of the product between each process.

[0048] Accordingly, after determining the target product production sample dataset, the target weight loss coefficient mean vector in the target product production sample dataset can be brought into the first association model as an input parameter to automatically calculate and output the target weight loss coefficient path through the first association model. Simultaneously, the target moisture change mean vector in the target product production sample dataset can be brought into the second association model as an input parameter to automatically calculate and output the target moisture change path through the second association model.

[0049] S150, calculating the target process product weight according to the product feed weight and the target weight reduction coefficient path, and calculating the target process product moisture according to the target finished product moisture and the target moisture change path.

[0050] The product feed weight can be the planned feed weight for the latest batch of the product. The target process product weight can be the product weight standard corresponding to each process calculated for the latest batch of the product. The target finished product moisture content can be the planned finished product moisture content for the latest batch of the product. The finished product moisture content refers to the moisture content of the final product. The target process product moisture content can be the product moisture standard corresponding to each process calculated for the latest batch of the product.

[0051] It is understood that each batch of products will be planned with a corresponding product feed weight and target finished product moisture content. Therefore, after the target weight reduction coefficient path and target moisture content change path are calculated based on the first and second association models, the product weight in each production process can be calculated based on the generation pattern of the weight reduction coefficient path, combined with the product feed weight and the target weight reduction coefficient path. Furthermore, the product moisture content in each production process can be calculated based on the generation pattern of the moisture content change path, combined with the product feed weight and the target moisture content change path.

[0052] Accordingly, the product weight at each production step can be used as the target process weight, which can be used as the weight process standard for each step of the latest batch of the product. At the same time, the product moisture content at each production step can be used as the target process moisture content, which can be used as the moisture process standard for each step of the latest batch of the product.

[0053] The embodiment of the present invention calculates the weight loss coefficient path and moisture change path of historical batch products based on the historical production data of the product, so as to establish a first association model based on the weight loss coefficient path of the historical batch products, and establish a second association model based on the moisture change path of the historical batch products. After the model is established, a target product production sample data set including a target weight loss coefficient mean vector and a target moisture change mean vector is determined based on the historical production data of the product, so as to calculate the target weight loss coefficient path and the target moisture change path based on the target product production sample data set, the first association model and the second association model, and finally calculate the target process product weight in combination with the product feed weight and the target weight loss coefficient path, and calculate the target process product moisture in combination with the target finished product moisture and the target moisture change path, thereby solving the problems of low accuracy, poor stability and practicality in the existing determination of product weight and moisture standards based on manual experience, and improving the accuracy, stability and practicality of product standard data such as product weight and moisture.

[0054] Example 2

[0055] Figure 2This is a flowchart of a product data processing method provided by the second embodiment of the present invention. This embodiment is specific based on the above embodiment and provides multiple specific optional implementation methods for calculating the weight reduction coefficient path and the moisture change path, establishing the first correlation model and the second correlation model, calculating the target weight reduction coefficient path and the target moisture change path, and calculating the target process product weight and the target process product moisture. Figure 2 As shown, the method of this embodiment may include:

[0056] S210. Calculate the weight reduction coefficient path and moisture change path of historical batches of products based on historical production data of the products.

[0057] In an optional embodiment of the present invention, the calculation of the weight discount coefficient path and moisture change path of historical batch products based on the historical production data of the product may include: obtaining the feed weight and process product weight of the current historical batch product based on the historical production data of the product; calculating the ratio of the process product weights between two adjacent processes of the current historical batch product based on the feed weight and the process product weight, as the process weight discount coefficient of the two adjacent processes; combining the process weight discount coefficients according to the association order between the processes to obtain the weight discount coefficient path of the current historical batch product; obtaining the finished material moisture content and process product moisture content of the current historical batch product based on the historical production data of the product; calculating the difference in process product moisture content between two adjacent processes of the current historical batch product based on the finished material moisture content and the process product moisture content, as the process moisture change of the two adjacent processes; combining the process moisture change content according to the association order between the processes to obtain the moisture change path of the current historical batch product.

[0058] The process weight reduction coefficient can be the ratio of the product weights between two adjacent processes, for example, the ratio of the product weight of the previous process to the product weight of the next process, or the ratio of the product weight of the next process to the product weight of the previous process, and the embodiments of the present invention are not limited to this. The process product weight is the product weight corresponding to the process link, and the process product moisture is the product moisture corresponding to the process link. The process moisture change can be the difference in product moisture between two adjacent processes, for example, the difference in product moisture between the previous process and the product moisture of the next process, or the difference in product moisture between the next process and the product moisture of the previous process, and the embodiments of the present invention are not limited to this.

[0059] It is understandable that the product's historical production data includes production data for multiple batches of historical batch products. Therefore, the production data for each historical batch product can be processed simultaneously or sequentially to generate a matching weight reduction coefficient path and moisture change path. Accordingly, for the production data of the current historical batch product, the feed weight of the current historical batch product and the weight of the process product corresponding to each process can be obtained based on the product's historical production data, and the feed weight is used as the basis for the weight standard to calculate the weight variation coefficient of each adjacent process as the weight reduction coefficient, and the weight reduction coefficient is used to characterize the relative relationship between the weights of each process. Optionally, the ratio or difference of the process product weights between two adjacent processes can be used as the process weight reduction coefficient of the two adjacent processes. At the same time, for the production data of the current historical batch product, the finished product material moisture content of the final product of the current historical batch product and the process product moisture content corresponding to each process can be obtained based on the product's historical production data, and the finished product material moisture content is used as the basis for the moisture standard to calculate the moisture change of each adjacent process, and the moisture change is used to characterize the relative relationship between the moisture content of each process. Optionally, the difference in the process product moisture content between two adjacent processes can be used as the moisture change of the two adjacent processes.

[0060] Correspondingly, after obtaining the process weight discount coefficient and process moisture change between each process, the process weight discount coefficient and each process moisture change can be combined according to the flow order of the process, so as to obtain the weight discount coefficient path and moisture change path of the current historical batch of products.

[0061] It should be noted that the calculation methods for process weight reduction coefficients and process moisture changes can vary for different products. For example, for products whose moisture content decreases with the process flow, the moisture change can be calculated by taking the difference between the moisture content of the product in the previous process and the product in the next process. For products whose moisture content increases with the process flow, the moisture change can be calculated by taking the difference between the moisture content of the product in the next process and the product in the previous process.

[0062] S220. Acquire calibration-related data of the historical production data of the product; wherein the calibration-related data includes calibration time data, weight calibration amount data, and moisture calibration amount data.

[0063] Calibration time data can include time records for calibrating equipment related to product weight and moisture content, such as electronic scales and moisture meters. Weight calibration data can include calibration values ​​for electronic scales used in various processes. Moisture calibration data can include calibration values ​​for moisture meters used in various processes.

[0064] S230 , determining a first segmentation time point of a weight reduction coefficient path and a second segmentation time point of a moisture change path according to the adjustment time data.

[0065] The first segmentation time point can be used to divide the weight reduction coefficient path of historical batch products, and the second segmentation time point can be used to divide the moisture change path of historical batch products.

[0066] Because the measured values ​​of equipment like electronic scales and moisture meters gradually become inaccurate over time, they require periodic testing during actual production. When the deviation exceeds a certain value, they must be calibrated. Understandably, each calibration introduces a step-change in the equipment's subsequent measurements, disrupting the regularity of data between previous and subsequent processes. However, in practice, equipment calibration is infrequent, so the production data for historical batches of products between calibrations can be considered a stable operating condition.

[0067] To this end, the calibration records of relevant electronic scales and moisture meters for measuring weight and moisture in each process can be obtained to determine the calibration time data of the product's historical production data, and the time points for splitting the sample set can be determined based on the calibration time data.

[0068] It is understandable that the calibration operations for electronic scales and moisture meters may not be consistent. Therefore, the first split time point of the weight reduction coefficient path for historical batches of products can be determined based on the calibration records of the electronic scale, and the second split time point of the moisture change path for historical batches of products can be determined based on the calibration records of the moisture meter.

[0069] S240. Divide the weight reduction coefficient path of the historical batch products into a weight reduction coefficient sample set according to the first segmentation time point, and divide the moisture change path into a moisture change sample set according to the second segmentation time point.

[0070] The weight reduction coefficient sample set may be a data sample set including only weight reduction coefficient paths, and the water content change sample set may be a data sample set including only water content change paths.

[0071] In an embodiment of the present invention, the calibration time points of the relevant electronic scales and moisture meters for measuring weight and moisture in each process of the product can be used as the time points for dividing the sample sets, and the weight reduction coefficient paths and moisture change paths of historical batch products can be divided to form multiple weight reduction coefficient sample sets and moisture change sample sets containing different sample sizes. Each sample set retains the actual batch production sequence and is updated as the production batch is updated.

[0072] It should be noted that due to the different calibration conditions of related equipment such as electronic scales and moisture meters, the division of the weight reduction coefficient sample set and the moisture change sample set may also be different.

[0073] For example, assuming that the historical production data of the product is data of multiple batches within the time range of 2021.01.01-2021.10.01, and the relevant electronic scale is calibrated on 2021.05.01, then 2021.05.01 can be used as the split time point of the discount coefficient sample set, and the discount coefficient path before the equipment is calibrated between 2021.01.01-2021.05.01 can be divided into a discount coefficient sample set of a time zone, and the discount coefficient path after the equipment is calibrated between 2021.05.01-2021.10.01 can be divided into a discount coefficient sample set of a time zone.

[0074] For example, assuming that the historical production data of the product is data for multiple batches within the time range of 2021.01.01-2021.10.01, and the relevant moisture meter is calibrated on 2021.06.01, then 2021.06.01 can be used as the split time point for the moisture change sample set, and the moisture change path before the equipment is calibrated between 2021.01.01-2021.06.01 can be divided into a moisture change sample set in a time zone, and the moisture change path after the equipment is calibrated between 2021.06.01-2021.10.01 can be divided into a moisture change sample set in a time zone.

[0075] It can be understood that since a depreciation coefficient sample set can include depreciation coefficient paths of multiple batches, and a moisture change sample set can include moisture change paths of multiple batches, for a sample set, the multiple paths of a sample set can be averaged in units of processes to obtain the depreciation coefficient mean vector and the moisture change mean vector.

[0076] In a specific example, the mean vector of the weight coefficient of a weight coefficient sample set can be expressed as Y i =[yi1,yi2,...,yi n-1 ]. Among them, Y i Represents the sample set of the weight reduction coefficient, yi n-1 Represents the mean of the n-1th weight loss coefficients in the weight loss paths of all batches in the weight loss sample set. For example, when a weight loss sample set includes weight loss paths of three batches, yi1 can be the mean of the first weight loss coefficients in the weight loss paths of the three batches.

[0077] S250: Establishing the first correlation model based on the weight adjustment data and the weight reduction coefficient sample set, and establishing the second correlation model based on the moisture adjustment data and the moisture change sample set.

[0078] In an optional embodiment of the present invention, the first association model and the second association model may include a linear regression model or a neural network model. In the embodiment of the present invention, the input of the first association model may be the weight reduction coefficient mean vector of the weight reduction coefficient sample set and the weight adjustment data of the electronic scale, and the output may be the weight reduction coefficient mean vector of the latest weight reduction coefficient sample set. The input of the second association model may be the moisture change mean vector of the moisture change sample set and the moisture adjustment data of the moisture meter, and the output may be the moisture change mean vector of the latest moisture change sample set. Optionally, the establishment process of the first association model and the second association model may be performed independently of each other without interfering with each other. Optionally, the specific structure and type of the first association model and the second association model may be selected and established according to the number of sample sets, and the embodiment of the present invention does not limit this.

[0079] For example, when the first correlation model and the second correlation model adopt linear regression models, the specific process of establishing the correlation model is described by taking the first correlation model as an example.

[0080] Step 1: The mean vector of the weight coefficient of a certain weight coefficient sample set can be expressed as Y i ,Y i =[yi1,yi2,...,yi n-1 ], the adjustment amount of the electronic scale can be expressed as X i ,X i =[xi1,xi2,...,xi n ]. Among them, xi n Represents the adjustment amount of the nth electronic scale. Optionally, each process can be equipped with an electronic scale for weight measurement. The mean vector of the next weight reduction coefficient sample set after the electronic scale is adjusted can be expressed as Y i+1 ,Y i+1 =[yi+11,yi+12,...,yi+1 n-1 ], where, yi+1 n-1 represents the mean of the n-1th weight reduction coefficients in the weight reduction coefficient paths of all batches in the next weight reduction coefficient sample set after the electronic scale is calibrated. In the above formula, n represents the number of electronic scales.

[0081] Step 2: Establish an ARX model for the above-mentioned weight reduction coefficient sample set. The structural expression of the ARX model is as follows:

[0082] y(t)=a1y(t-1)+a2y(t-2)+...+apy(tp)+b1x(t-1)+b2x(t-2)+...+bmx(tm)

[0083] +e(t)

[0084] Where y(t) is the output variable (dependent variable) of the system, y(t-1), y(t-2), ..., y(tp) are the delayed values ​​of the output variables, p is the AR order, x(t-1), x(t-2), ..., x(tm) are the input variables (external input) of the system, m is the X order, a1, a2, ..., ap are the autoregressive coefficients of the output variables, b1, b2, ..., bm are the coefficients of the external input variables, and e(t) is the error term of the model.

[0085] When applying the above-mentioned ARX model to the technical solution of the embodiment of the present invention, p=1 and m=1 can be used to establish a first-order ARX model as the first correlation model. That is, the model structure of the first correlation model can be y(t)=a1y(t-1)+b1x(t-1)+e(t).

[0086] Step 3: The mean vector of the weight coefficient based on all weight coefficient sample sets Y1-Y L , and the electronic scale adjustment amount X1-X L , using Y1-Y L Assign values ​​to y(t-1) and y(t), and use X1-X L Assign values ​​to x(t-1), for example, y(t) = Y2, y(t-1) = Y1, x(t-1) = X1, etc., to train the first association model until the accuracy of the first association model reaches a set value. During the training process, a recursive least squares method can be used to identify a1, b1, and e in the first-order ARX model, thereby obtaining the trained first association model.

[0087] Step 4: When a new weight reduction coefficient sample set is generated, the model parameters a1, b1, and e of the first correlation model can be updated using the latest electronic scale calibration data and the latest weight reduction coefficient mean vector data. Optionally, the recursive least squares method can also be used to update the parameters.

[0088] Similarly, the second correlation model can be trained using the same training method as the first correlation model until a fully trained second correlation model is obtained. When a new moisture change sample set is generated, the model parameters a1, b1, and e of the second correlation model can be updated using the latest moisture meter calibration data and the latest moisture change mean vector data.

[0089] S260: Determine a target product production sample data set based on the product historical production data.

[0090] The target product production sample data set includes a target weight reduction coefficient mean vector and a target moisture change mean vector.

[0091] S270: Calculate a target weight reduction coefficient path and a target moisture change path based on the target product production sample data set, the first association model, and the second association model.

[0092] In an optional embodiment of the present invention, the target product production sample data set includes a first target product production sample data set and a second target product production sample data set; accordingly, a first discount coefficient mean vector and a first moisture change mean vector of the first target product production sample data set, as well as a second discount coefficient mean vector and a second moisture change mean vector of the second target product production sample data set are calculated; the target adjustment amount sample data associated with the target product production sample data set is determined; the first discount coefficient mean vector and the weight adjustment amount sample data in the target adjustment amount sample data are brought into the first association model as input parameters to output a third discount coefficient mean vector through the first association model; the first moisture change mean vector and the moisture adjustment amount sample data in the target adjustment amount sample data are brought into the second association model as input parameters to output a third moisture change mean vector through the second association model; the second discount coefficient mean vector and the third discount coefficient mean vector are weighted to obtain the target discount coefficient path; the second moisture change mean vector and the third moisture change mean vector are weighted to obtain the target moisture change path.

[0093] The first target product production sample dataset may be the latest weight loss coefficient sample set and moisture change sample set, and the second target product production sample dataset may be the latest weight loss coefficient sample set and moisture change sample set. The first weight loss coefficient mean vector may be the weight loss coefficient mean vector of the latest weight loss coefficient sample set, and the first moisture change mean vector may be the moisture change mean vector of the latest moisture change sample set. The second weight loss coefficient mean vector may be the weight loss coefficient mean vector of the latest weight loss coefficient sample set, and the second moisture change mean vector may be the moisture change mean vector of the latest moisture change sample set. The target calibration sample data may be sample data obtained by calibrating equipment between the time periods of the first target product production sample dataset and the second target product production sample dataset. The third weight loss coefficient mean vector may be the weight loss coefficient mean vector calculated from the first weight loss coefficient mean vector using the first correlation model, and may be used as a reference for calculating the target weight loss coefficient path. The third moisture change mean vector may be the moisture change mean vector calculated from the first moisture change mean vector using the second correlation model, and may be used as a reference for calculating the target moisture change path.

[0094] When the sample data is small, considering that the accuracy or confidence of the association model (such as the linear programming model) may be difficult to meet the needs, the target weight discount coefficient path and the target moisture change path can be calculated in combination with the actual production data of the product, so as to determine the process standard data such as process product weight and process product moisture based on the target weight discount coefficient path and the target moisture change path.

[0095] Specifically, the latest weight loss coefficient sample set and moisture content change sample set can be used as the first target product production sample data set, and the latest weight loss coefficient sample set and moisture content change sample set can be used as the second target product production sample data set. Furthermore, the weight loss coefficient mean vector in the latest weight loss coefficient sample set is calculated as the first weight loss coefficient mean vector, the moisture content change mean vector in the latest moisture content change sample set is calculated as the first moisture content change mean vector, the weight loss coefficient mean vector in the latest weight loss coefficient sample set is calculated as the second weight loss coefficient mean vector, and the moisture content change mean vector in the latest moisture content change sample set is calculated as the second moisture content change mean vector. At the same time, the latest electronic scale calibration value is obtained as the weight calibration value sample data, and the latest moisture meter calibration value is obtained as the moisture content adjustment value sample data. Furthermore, the first weight loss coefficient mean vector and the latest weight adjustment sample data are brought into the first association model as input parameters. The first association model outputs a third weight loss coefficient mean vector, and the second and third weight loss coefficient mean vectors are weighted to obtain a target weight loss coefficient path. Simultaneously, the first moisture change mean vector and the latest moisture adjustment sample data are brought into the second association model as input parameters. The second association model outputs a third moisture change mean vector, and the second and third moisture change mean vectors are weighted to obtain a target moisture change path.

[0096] Optionally, when weighting the second and third weighting coefficient mean vectors, and when weighting the second and third moisture change mean vectors, the weights corresponding to the respective vectors can be set based on actual needs. For example, if the actual production data is considered to be highly credible, the weights for the second and third moisture change mean vectors can be set relatively high.

[0097] In a specific example, Table 1 shows the product weight data for each process in a specific weight reduction coefficient sample set for a product. As shown in Table 1, the weight reduction coefficient sample set includes product weight data for three batches of the product, each of which involves four processes. Accordingly, the weight reduction coefficient path for batch 1 might be m12 / m11-m13 / m12-m14 / m13; the weight reduction coefficient path for batch 2 might be m22 / m21-m23 / m22-m24 / m23; and the weight reduction coefficient path for batch 3 might be m32 / m31-m33 / m32-m34 / m33. Calculating the mean of each weight coefficient in this weight coefficient sample set yields: m1 = (m12 / m11 + m22 / m21 + m32 / m31) / 3, m2 = (m13 / m12 + m23 / m22 + m33 / m32) / 3, and m3 = (m14 / m13 + m24 / m23 + m34 / m33) / 3. Accordingly, the mean weight coefficient vector for this weight coefficient sample set can be: (m1, m2, m3).

[0098] Table 1 Product weight of each process

[0099] Process 1 Process 2 Process 3 Process 4 Batch 1 m11 m12 m13 m14 Batch 2 m21 m22 m23 m24 Batch 3 m31 m32 m33 m34

[0100] In a specific example, Table 2 shows the moisture data for each process in a moisture variation sample set for a product. As shown in Table 2, the moisture variation sample set includes moisture data for three batches of the product, each involving four processes. Accordingly, the moisture variation path for batch 1 can be (w11-w12)-(w12-w13)-(w13-w14); the moisture variation path for batch 2 can be (w21-w22)-(w22-w23)-(w23-w24); and the moisture variation path for batch 3 can be (w31-w32)-(w32-w33)-(w33-w34). Calculating the mean of each moisture change in this moisture change sample set yields: w1 = ((w11-w12) + (w21-w22) + (w31-w32)) / 3. Similarly, w2 = ((w12-w13) + (w22-w23) + (w32-w33)) / 3, and w3 = ((w13-w14) + (w23-w24) + (w33-w34)) / 3. Accordingly, the mean moisture change vector for this moisture change sample set can be: (w1, w2, w3).

[0101] Table 2 Product moisture content of each process

[0102] Process 1 Process 2 Process 3 Process 4 Batch 1 w11 w12 w13 w14 Batch 2 w21 w22 w23 w24 Batch 3 w31 w32 w33 w34

[0103] Exemplarily, according to the calculation method of the above-mentioned mean vector, the weight loss coefficient mean vector a1 and the moisture change mean vector b1 in the newest sample set can be calculated, and the weight loss coefficient mean vector a2 and the moisture change mean vector b2 in the latest sample set can be calculated. Among them, the latest sample set includes all batches of data since the last calibration of the quality measurement instrument, and the newest sample set includes data from the last calibration of the last measurement instrument to the last calibration of the last measurement instrument. Further, the weight loss coefficient mean vector a1 in the newest sample set and the calibration amount of the electronic scale are brought into the first association model to obtain a set of new weight loss coefficient mean vectors a3, and the moisture change mean vector b1 in the newest sample set and the moisture meter calibration amount are brought into the second association model S to obtain a set of new moisture change mean vectors b3. Further, a2 and a3 are weighted to obtain the target weight loss coefficient path, and b2 and b3 are weighted to obtain the target moisture change path.

[0104] In an optional embodiment of the present invention, the target product production sample data set may include a second target product production sample data set; accordingly, the calculation of the target discount coefficient path and the target moisture change path based on the target product production sample data set, the first association model and the second association model may include: calculating the second discount coefficient mean vector and the second moisture change mean vector of the second target product production sample data set; determining the target adjustment amount sample data associated with the target product production sample data set; bringing the second discount coefficient mean vector and the weight adjustment amount sample data in the target adjustment amount sample data as input parameters into the first association model to output a fourth discount coefficient mean vector through the first association model; bringing the second moisture change amount mean vector and the moisture adjustment amount sample data in the target adjustment amount sample data as input parameters into the second association model to output a fourth moisture change amount mean vector through the second association model; generating the target discount coefficient path according to the fourth discount coefficient mean vector, and generating the target moisture change amount path according to the fourth moisture change amount mean vector.

[0105] The fourth weight coefficient mean vector may be a weight coefficient mean vector calculated from the second weight coefficient mean vector using the first correlation model, and may be used as a reference for calculating the target weight coefficient path. The fourth moisture change mean vector may be a moisture change mean vector calculated from the second moisture change mean vector using the second correlation model, and may be used as a reference for calculating the target moisture change path.

[0106] When there is a large amount of sample data, considering that the accuracy or confidence of the association model (such as the neural network model) can meet the requirements, the target weight discount coefficient path and the target moisture change path can be calculated through the association model, and the process standard data such as process product weight and process product moisture can be determined based on the target weight discount coefficient path and the target moisture change path.

[0107] Specifically, the latest weight loss coefficient sample set and moisture change sample set can be directly used as the second target product production sample data set. The weight loss coefficient mean vector in the latest weight loss coefficient sample set is calculated as the second weight loss coefficient mean vector, and the moisture change mean vector in the latest moisture change sample set is calculated as the second moisture change mean vector. Simultaneously, the latest electronic scale calibration values ​​are obtained as weight calibration sample data, and the latest moisture meter calibration values ​​are obtained as moisture calibration sample data. Furthermore, the second weight loss coefficient mean vector and the latest weight calibration sample data are used as input parameters for the first association model, and a fourth weight loss coefficient mean vector is outputted from the first association model. Simultaneously, the second moisture change mean vector and the latest moisture adjustment sample data are used as input parameters for the second association model, and a fourth moisture change mean vector is outputted from the second association model. Accordingly, after obtaining the fourth weight loss coefficient mean vector and the fourth moisture change mean vector, a target weight loss coefficient path can be directly generated based on the fourth weight loss coefficient mean vector, and a target moisture change path can be generated based on the fourth moisture change mean vector.

[0108] For example, when the fourth weight loss coefficient mean vector is (m1, m2, m3), the target weight loss coefficient path can be generated as m1-m2-m3. When the fourth moisture change mean vector is (w1, w2, w3), the target weight loss coefficient path can be generated as w1-w2-w3.

[0109] S280: Taking the product feed weight as the initial weight reference, multiplying the weight discount coefficients of each process in the target weight discount coefficient path in sequence to obtain the target process product weight that matches each process.

[0110] Among them, the initial weight basis is also the weight basis put into product production.

[0111] In a specific example, assuming that the target weight discount coefficient path is m1-m2-m3, the product input weight is m0, and the corresponding process product weight of the first process is m0, then by multiplying the weight discount coefficients of each process in the target weight discount coefficient path in sequence, it can be obtained that the process product weight of the second process is m0*m1, the process product weight of the third process is m0*m1*m2, and the process product weight of the fourth process is m0*m1*m2*m3.

[0112] S290: Taking the target finished product moisture as the finished product moisture target value, sequentially accumulating the moisture changes of each process in the target moisture change path to obtain the target process product moisture matching each process.

[0113] The target moisture value of the finished product is also the moisture value of the final product.

[0114] In a specific example, assuming that the target moisture change path is w1-w2-w3, the target moisture value of the finished product is w0, and the corresponding moisture value of the process product of the fourth process is w0, then by accumulating the moisture changes of each process in the target moisture change path in sequence, it can be obtained that the moisture value of the process product of the third process is w0+w3, the moisture value of the process product of the second process is w0+w3+w2, and the moisture value of the process product of the first process is w0+w3+w2+w1.

[0115] Understandably, during the production process, equipment measurement inaccuracies accumulate gradually, while equipment calibration is a one-time process. Therefore, it's important to consider the impact of instrument inaccuracies on data patterns. The above technical solution establishes correlation models between electronic scale calibration and weight reduction coefficient, and between moisture meter calibration and moisture change, to quantitatively characterize the extent to which instrument inaccuracies disrupt data patterns. This allows for the development of a new standard based on the previous version.

[0116] This technical solution fully utilizes historical production data to develop process standard data for each product. This processing results in stable and highly interpretable standard data. Furthermore, correlation models established using batch data sample sets, electronic scale calibration values, and weight reduction coefficients, as well as correlation models using batch data sample sets, moisture meter calibration values, and moisture variation, address the problem of instrument inaccuracy disrupting data patterns. Furthermore, these correlation models can be updated online, enhancing the practicality of this method for processing process standard data.

[0117] It should be noted that any arrangement and combination of the technical features in the above embodiments also falls within the protection scope of the present invention.

[0118] Example 3

[0119] Figure 3 Schematic diagram of a product data processing device provided by the third embodiment of the present invention. Figure 3 As shown, the device includes: a first path calculation module 310, an association model establishment module 320, a sample data set determination module 330, a second path calculation module 340 and a weight moisture calculation module 350, wherein:

[0120] A first path calculation module 310 is used to calculate the weight reduction coefficient path and moisture change path of historical batch products based on historical production data of the product;

[0121] The correlation model establishment module 320 is used to establish a first correlation model based on the weight reduction coefficient path of the historical batch products, and to establish a second correlation model based on the moisture change path of the historical batch products;

[0122] The sample data set determination module 330 is configured to determine a target product production sample data set based on the product historical production data; wherein the target product production sample data set includes a target weight reduction coefficient mean vector and a target moisture change mean vector;

[0123] A second path calculation module 340 is configured to calculate a target weight reduction coefficient path and a target moisture change path based on the target product production sample dataset, the first correlation model, and the second correlation model;

[0124] The weight moisture calculation module 350 is used to calculate the target process product weight based on the product input weight and the target weight reduction coefficient path, and calculate the target process product moisture based on the target finished product moisture and the target moisture change path.

[0125] The embodiment of the present invention calculates the weight loss coefficient path and moisture change path of historical batch products based on the historical production data of the product, so as to establish a first association model based on the weight loss coefficient path of the historical batch products, and establish a second association model based on the moisture change path of the historical batch products. After the model is established, a target product production sample data set including a target weight loss coefficient mean vector and a target moisture change mean vector is determined based on the historical production data of the product, so as to calculate the target weight loss coefficient path and the target moisture change path based on the target product production sample data set, the first association model and the second association model, and finally calculate the target process product weight in combination with the product feed weight and the target weight loss coefficient path, and calculate the target process product moisture in combination with the target finished product moisture and the target moisture change path, thereby solving the problems of low accuracy, poor stability and practicality in the existing determination of product weight and moisture standards based on manual experience, and improving the accuracy, stability and practicality of product standard data such as product weight and moisture.

[0126] Optionally, the first path calculation module 310 is specifically used to: obtain the feed weight and process product weight of the current historical batch of products based on the product historical production data; calculate the ratio of the process product weights between two adjacent processes of the current historical batch of products based on the feed weight and the process product weight, as the process weight discount coefficient of the two adjacent processes; combine the process weight discount coefficients according to the association order between the processes to obtain the weight discount coefficient path of the current historical batch of products; obtain the finished material moisture content and process product moisture content of the current historical batch of products based on the product historical production data; calculate the difference in process product moisture content between two adjacent processes of the current historical batch of products based on the finished material moisture content and the process product moisture content, as the process moisture change of the two adjacent processes; combine the process moisture change content according to the association order between the processes to obtain the moisture change path of the current historical batch of products.

[0127] Optionally, the association model establishment module 320 is specifically used to: obtain the adjustment association data of the historical production data of the product; wherein the adjustment association data includes adjustment time data, weight adjustment amount data and moisture adjustment amount data; determine the first segmentation time point of the discount coefficient path and the second segmentation time point of the moisture change amount path according to the adjustment time data; divide the discount coefficient path of the historical batch product into a discount coefficient sample set according to the first segmentation time point, and divide the moisture change amount path into a moisture change amount sample set according to the second segmentation time point; establish the first association model based on the weight adjustment amount data and the discount coefficient sample set, and establish the second association model based on the moisture adjustment amount data and the moisture change amount sample set.

[0128] Optionally, the first association model and the second association model include a linear regression model or a neural network model.

[0129] Optionally, the target product production sample data set includes a first target product production sample data set and a second target product production sample data set; the second path calculation module 340 is specifically used to: calculate the first discount coefficient mean vector and the first moisture change mean vector of the first target product production sample data set, and the second discount coefficient mean vector and the second moisture change mean vector of the second target product production sample data set; determine the target adjustment amount sample data associated with the target product production sample data set; bring the first discount coefficient mean vector and the weight adjustment amount sample data in the target adjustment amount sample data as input parameters into the first association model to output a third discount coefficient mean vector through the first association model; bring the first moisture change amount mean vector and the moisture adjustment amount sample data in the target adjustment amount sample data as input parameters into the second association model to output a third moisture change amount mean vector through the second association model; perform weighted processing on the second discount coefficient mean vector and the third discount coefficient mean vector to obtain the target discount coefficient path; perform weighted processing on the second moisture change amount mean vector and the third moisture change amount mean vector to obtain the target moisture change amount path.

[0130] Optionally, the target product production sample data set includes a second target product production sample data set; the second path calculation module 340 is specifically used to: calculate the second discount coefficient mean vector and the second moisture change mean vector of the second target product production sample data set; determine the target adjustment amount sample data associated with the target product production sample data set; bring the second discount coefficient mean vector and the weight adjustment amount sample data in the target adjustment amount sample data as input parameters into the first association model to output a fourth discount coefficient mean vector through the first association model; bring the second moisture change mean vector and the moisture adjustment amount sample data in the target adjustment amount sample data as input parameters into the second association model to output a fourth moisture change mean vector through the second association model; generate the target discount coefficient path according to the fourth discount coefficient mean vector, and generate the target moisture change amount path according to the fourth moisture change mean vector.

[0131] Optionally, the weight moisture calculation module 350 is specifically used to: take the product feed weight as the initial weight reference, and sequentially multiply the weight discount coefficient of each process in the target weight discount coefficient path to obtain the target process product weight matching each process; take the target finished product moisture as the finished product moisture target value, and sequentially accumulate the moisture change of each process in the target moisture change path to obtain the target process product moisture matching each process.

[0132] The above-mentioned product data processing device can execute the product data processing method provided by any embodiment of the present invention, and has the corresponding functional modules and beneficial effects of the execution method. For technical details not fully described in this embodiment, please refer to the product data processing method provided by any embodiment of the present invention.

[0133] Since the product data processing device described above is a device that can execute the product data processing method in the embodiment of the present invention, and based on the product data processing method described in the embodiment of the present invention, those skilled in the art will be able to understand the specific implementation and various variations of the product data processing device of this embodiment. Therefore, how the product data processing device implements the product data processing method in the embodiment of the present invention will not be described in detail here. As long as those skilled in the art can implement the device used by the product data processing method in the embodiment of the present invention, it falls within the scope of protection of this application.

[0134] Example 4

[0135] Figure 4 A schematic diagram of the structure of an electronic device 10 that can be used to implement an embodiment of the present invention is shown. The electronic device is intended to represent various forms of digital computers, such as laptop computers, desktop computers, workstations, personal digital assistants, servers, blade servers, mainframe computers, and other suitable computers. The electronic device can also represent various forms of mobile devices, such as personal digital processing, cellular phones, smart phones, wearable devices (such as helmets, glasses, watches, etc.) and other similar computing devices. The components shown herein, their connections and relationships, and their functions are merely examples and are not intended to limit the implementation of the present invention described and / or claimed herein.

[0136] like Figure 4 As shown, the electronic device 10 includes at least one processor 11 and a memory, such as a read-only memory (ROM) 12, a random access memory (RAM) 13, etc., which is communicatively connected to the at least one processor 11. The memory stores a computer program that can be executed by the at least one processor. The processor 11 can perform various appropriate actions and processes according to the computer program stored in the read-only memory (ROM) 12 or the computer program loaded from the storage unit 18 into the random access memory (RAM) 13. Various programs and data required for the operation of the electronic device 10 can also be stored in the RAM 13. The processor 11, ROM 12, and RAM 13 are connected to each other via a bus 14. An input / output (I / O) interface 15 is also connected to the bus 14.

[0137] Multiple components in the electronic device 10 are connected to the I / O interface 15, including an input unit 16, such as a keyboard, a mouse, etc.; an output unit 17, such as various types of displays, speakers, etc.; a storage unit 18, such as a magnetic disk, an optical disk, etc.; and a communication unit 19, such as a network card, a modem, a wireless communication transceiver, etc. The communication unit 19 allows the electronic device 10 to exchange information / data with other devices via a computer network such as the Internet and / or various telecommunication networks.

[0138] The processor 11 can be any general-purpose and / or specialized processing component with processing and computing capabilities. Some examples of the processor 11 include, but are not limited to, a central processing unit (CPU), a graphics processing unit (GPU), various specialized artificial intelligence (AI) computing chips, various processors that run machine learning model algorithms, a digital signal processor (DSP), and any suitable processor, controller, microcontroller, etc. The processor 11 executes the various methods and processes described above, such as the product data processing method.

[0139] In some embodiments, the product data processing method can be implemented as a computer program tangibly embodied in a computer-readable storage medium, such as storage unit 18. In some embodiments, part or all of the computer program can be loaded and / or installed on electronic device 10 via ROM 12 and / or communication unit 19. When the computer program is loaded into RAM 13 and executed by processor 11, one or more steps of the product data processing method described above can be performed. Alternatively, in other embodiments, processor 11 can be configured to execute the product data processing method in any other suitable manner (e.g., via firmware).

[0140] Optionally, the product data processing method may include: calculating the weight discount coefficient path and moisture change path of historical batch products based on the historical production data of the product; establishing a first association model based on the weight discount coefficient path of the historical batch products, and establishing a second association model based on the moisture change path of the historical batch products; determining a target product production sample data set based on the historical production data of the product; wherein the target product production sample data set includes a target weight discount coefficient mean vector and a target moisture change mean vector; calculating the target weight discount coefficient path and target moisture change path based on the target product production sample data set, the first association model and the second association model; calculating the target process product weight based on the product feed weight and the target weight discount coefficient path, and calculating the target process product moisture based on the target finished product moisture and the target moisture change path.

[0141] Various embodiments of the systems and techniques described herein can be implemented in digital electronic circuit systems, integrated circuit systems, field programmable gate arrays (FPGAs), application specific integrated circuits (ASICs), application specific standard products (ASSPs), system-on-chip systems (SOCs), programmable logic devices (CPLDs), computer hardware, firmware, software, and / or combinations thereof. These various embodiments can include being implemented in one or more computer programs that are executable and / or interpreted on a programmable system that includes at least one programmable processor, which can be a special purpose or general purpose programmable processor that can receive data and instructions from a storage system, at least one input device, and at least one output device, and transmit data and instructions to the storage system, the at least one input device, and the at least one output device.

[0142] Computer programs for implementing the methods of the present invention may be written in any combination of one or more programming languages. These computer programs may be provided to a processor of a general-purpose computer, a special-purpose computer, or other programmable data processing device, such that when the computer program is executed by the processor, the functions / operations specified in the flowcharts and / or block diagrams are implemented. The computer program may be executed entirely on the machine, partially on the machine, as a stand-alone software package, partially on the machine and partially on a remote machine, or entirely on a remote machine or server.

[0143] In the context of the present invention, computer-readable storage media can be tangible media that can contain or store a computer program for use with an instruction execution system, device or equipment or used in combination with an instruction execution system, device or equipment. Computer-readable storage media can include but are not limited to electronic, magnetic, optical, electromagnetic, infrared or semiconductor systems, devices or equipment, or any suitable combination of the foregoing. Alternatively, computer-readable storage media can be machine-readable signal media. More specific examples of machine-readable storage media can include electrical connections based on one or more lines, portable computer disks, hard disks, random access memories (RAM), read-only memories (ROM), erasable programmable read-only memories (EPROM or flash memory), optical fibers, portable compact disk read-only memories (CD-ROM), optical storage devices, magnetic storage devices, or any suitable combination of the foregoing.

[0144] To provide interaction with a user, the systems and techniques described herein can be implemented on an electronic device having: a display device (e.g., a CRT (cathode ray tube) or LCD (liquid crystal display) monitor) for displaying information to the user; and a keyboard and pointing device (e.g., a mouse or trackball) through which the user can provide input to the electronic device. Other types of devices can also be used to provide interaction with the user; for example, the feedback provided to the user can be any form of sensory feedback (e.g., visual feedback, auditory feedback, or tactile feedback); and the input from the user can be received in any form (including acoustic input, voice input, or tactile input).

[0145] The systems and techniques described herein can be implemented in a computing system that includes back-end components (e.g., as a data server), or a computing system that includes middleware components (e.g., an application server), or a computing system that includes front-end components (e.g., a user computer with a graphical user interface or web browser through which a user can interact with implementations of the systems and techniques described herein), or a computing system that includes any combination of such back-end components, middleware components, or front-end components. The components of the system can be interconnected by any form or medium of digital data communication (e.g., a communication network). Examples of communication networks include: a local area network (LAN), a wide area network (WAN), a blockchain network, and the Internet.

[0146] A computing system may include clients and servers. The clients and servers are typically remote from each other and typically interact via a communication network. This client-server relationship arises through computer programs running on the respective computers, creating a client-server relationship. The server may be a cloud server, also known as a cloud computing server or cloud host. This server is a hosting product within the cloud computing service ecosystem that addresses the management difficulties and limited scalability of traditional physical hosting and VPS services.

Claims

1. A product data processing method, characterized in that: include: Calculating a weight loss coefficient path and a moisture change path for historical batches of products based on historical product production data; wherein the weight loss coefficient path is a path formed by weight loss coefficients between various processes in the historical production data for a batch of products, and the weight loss coefficients are the relative relationship between product weights between various processes in the historical production data for a batch of products; and the moisture change path is a path formed by moisture change amounts between various processes in the historical production data for a batch of products, and the moisture change amounts are the moisture change amounts between various processes in the historical production data for a batch of products; Acquire calibration-related data of the historical production data of the product; wherein the calibration-related data includes weight calibration data and moisture calibration data; Establishing a first correlation model based on the weight adjustment data and the weight reduction coefficient path of the historical batch products, and establishing a second correlation model based on the moisture adjustment data and the moisture change path of the historical batch products; Determining a target product production sample data set based on the product historical production data; wherein the target product production sample data set includes a target weight reduction coefficient mean vector and a target moisture change mean vector; Calculating a target weight reduction coefficient path and a target moisture change path according to the target product production sample data set, the first association model, and the second association model; The target process product weight is calculated based on the product feed weight and the target weight reduction coefficient path, and the target process product moisture is calculated based on the target finished product moisture and the target moisture change path.

2. The method according to claim 1, characterized in that The calculation of the weight reduction coefficient path and moisture change path of historical batch products based on the historical production data of the product includes: Obtain the material weight and process product weight of the current historical batch of products based on the historical production data of the product; Calculate the ratio of the process product weights between two adjacent processes of the current historical batch of products based on the input weight and the process product weight as the process weight discount coefficient of the two adjacent processes; Combining the weight discount coefficients of each process according to the association sequence between the processes to obtain the weight discount coefficient path of the current historical batch product; Obtain the finished product moisture and process product moisture of the current historical batch of products based on the historical production data of the product; Calculate the difference in process product moisture between two adjacent processes of the current historical batch of products based on the finished product material moisture and the process product moisture as the process moisture change of the two adjacent processes; The moisture change amount of each process is combined according to the association sequence between the processes to obtain the moisture change amount path of the current historical batch of products.

3. The method according to claim 1, characterized in that The first correlation model is established according to the weight reduction coefficient path of the historical batch products, and the second correlation model is established according to the moisture change path of the historical batch products, including: Acquire calibration-related data of the historical production data of the product; wherein the calibration-related data includes calibration time data, weight calibration amount data, and moisture calibration amount data; Determine a first segmentation time point of the weight reduction coefficient path and a second segmentation time point of the moisture change path according to the adjustment time data; Dividing the weight reduction coefficient path of the historical batch product into a weight reduction coefficient sample set according to the first segmentation time point, and dividing the moisture change path into a moisture change sample set according to the second segmentation time point; The first correlation model is established based on the weight adjustment data and the weight reduction coefficient sample set, and the second correlation model is established based on the moisture adjustment data and the moisture change sample set.

4. The method according to any one of claims 1 to 3, characterized in that: The first association model and the second association model include a linear regression model or a neural network model.

5. The method according to claim 1, wherein The target product production sample data set includes a first target product production sample data set and a second target product production sample data set; The calculating of the target weight reduction coefficient path and the target moisture change path according to the target product production sample data set, the first correlation model, and the second correlation model includes: Calculating a first weight reduction coefficient mean vector and a first moisture change mean vector of the first target product production sample data set, and a second weight reduction coefficient mean vector and a second moisture change mean vector of the second target product production sample data set; Determining target adjustment quantity sample data associated with the target product production sample data set; Bringing the first weight reduction coefficient mean vector and the weight adjustment amount sample data in the target adjustment amount sample data as input parameters into the first association model, so as to output a third weight reduction coefficient mean vector through the first association model; Bringing the first moisture change amount mean vector and the moisture adjustment amount sample data in the target adjustment amount sample data as input parameters into the second correlation model, so as to output a third moisture change amount mean vector through the second correlation model; Performing weighted processing on the second weight reduction coefficient mean vector and the third weight reduction coefficient mean vector to obtain the target weight reduction coefficient path; The second moisture change mean vector and the third moisture change mean vector are weighted to obtain the target moisture change path.

6. The method according to claim 1, characterized in that The target product production sample data set includes a second target product production sample data set; The calculating of the target weight reduction coefficient path and the target moisture change path according to the target product production sample data set, the first correlation model, and the second correlation model includes: Calculating a second weight reduction coefficient mean vector and a second moisture change mean vector of the second target product production sample data set; Determining target adjustment quantity sample data associated with the target product production sample data set; Bringing the second weight reduction coefficient mean vector and the weight adjustment amount sample data in the target adjustment amount sample data as input parameters into the first association model, so as to output a fourth weight reduction coefficient mean vector through the first association model; Bringing the second moisture change amount mean vector and the moisture adjustment amount sample data in the target adjustment amount sample data as input parameters into the second correlation model, so as to output a fourth moisture change amount mean vector through the second correlation model; The target weight reduction coefficient path is generated according to the fourth weight reduction coefficient mean vector, and the target moisture change path is generated according to the fourth moisture change mean vector.

7. The method according to claim 1, characterized in that The target process product weight is calculated based on the product feed weight and the target weight reduction coefficient path, and the target process product moisture is calculated based on the target finished product moisture and the target moisture change path, including: Taking the product feed weight as the initial weight reference, the weight reduction coefficients of each process in the target weight reduction coefficient path are multiplied in sequence to obtain the target process product weight matched by each process; The target finished product moisture is used as the finished product moisture target value, and the moisture changes of each process in the target moisture change path are accumulated in sequence to obtain the target process product moisture matching each process.

8. A product data processing device, characterized in that: include: A first path calculation module is configured to calculate a weight reduction coefficient path and a moisture change path for historical batches of products based on historical product production data; wherein the weight reduction coefficient path is a path formed by weight reduction coefficients between various processes in a batch of product historical production data, wherein the weight reduction coefficients are the relative relationship between product weights between various processes in a batch of product historical production data; and the moisture change path is a path formed by moisture change amounts between various processes in a batch of product historical production data, wherein the moisture change amounts are the moisture change amounts between various processes in a batch of product historical production data; a correlation model establishment module, configured to obtain calibration correlation data of the historical production data of the product, wherein the calibration correlation data includes weight calibration data and moisture calibration data; establish a first correlation model based on the weight calibration data and the weight reduction coefficient path of the historical batch product, and establish a second correlation model based on the moisture calibration data and the moisture variation path of the historical batch product; A sample data set determination module is used to determine a target product production sample data set based on the product historical production data; wherein the target product production sample data set includes a target weight reduction coefficient mean vector and a target moisture change mean vector; A second path calculation module is used to calculate a target weight reduction coefficient path and a target moisture change path according to the target product production sample data set, the first correlation model and the second correlation model; The weight moisture calculation module is used to calculate the target process product weight based on the product feed weight and the target weight reduction coefficient path, and to calculate the target process product moisture based on the target finished product moisture and the target moisture change path.

9. An electronic device, characterized in that: The electronic device comprises: at least one processor; and a memory communicatively connected to the at least one processor; wherein, The memory stores a computer program executable by the at least one processor, and the computer program is executed by the at least one processor so that the at least one processor can execute the product data processing method according to any one of claims 1 to 7.

10. A computer-readable storage medium, characterized in that The computer-readable storage medium stores computer instructions, and the computer instructions are used to enable a processor to implement the product data processing method according to any one of claims 1 to 7 when executed.

Images