Feed grease squeezing risk management and control system based on machine learning

Abstract

The invention discloses a feed grease squeezing risk management and control system based on machine learning, and relates to the field of feed grease squeezing risk management and control, and the system comprises a data division module, a preprocessing stage analysis module, a preliminary analysis module and a risk management and control module. Through carrying out mechanical learning on historical data of feed grease squeezing, a data comparison model corresponding to each crushing initial data, each cooking initial data and each dehydration initial data corresponding to each adipose tissue type is established, and a data basis is provided for feed grease squeezing risk management and control. And data comparison and related processing are performed on the working data of the to-be-evaluated sub-process, so that the practicability of risk management and control on feed grease squeezing quality improvement and yield improvement is improved, and the competitiveness and market status of enterprises are improved.

Description

Technical Field

[0001] The present application relates to the field of feed oil and fat pressing risk management and control, and in particular to a feed oil and fat pressing risk management and control system based on machine learning. Background Art

[0002] In the feed oil pressing industry, the competitiveness of enterprises depends not only on product quality and price, but also on the production efficiency and risk management capabilities of enterprises. Through risk control of the feed oil pressing process, enterprises can improve production efficiency and product quality, thereby enhancing the competitiveness and market position of enterprises, and thus stand out in the fierce market competition. The existing technology still has the following shortcomings: When conducting risk control warnings for feed oil and fat pressing, existing technologies usually only conduct risk control warnings for the pressing results at the end of each stage, which reduces the practicality and effectiveness of risk control. Specifically, it reduces the usability of risk control for improving the quality and output of feed oil and fat pressing, which is not conducive to improving the competitiveness and market position of enterprises. Summary of the invention

[0003] The purpose of the present invention is to provide a feed oil pressing risk management system based on machine learning to solve the problems raised in the above background technology.

[0004] To achieve the above purpose, the present invention provides the following technical solution: a feed oil pressing risk management and control system based on machine learning, comprising: Data division module: used to obtain historical data of feed oil pressing and comprehensively divide it according to the pressing process, so as to obtain historical data of the pretreatment stage of each type of fat tissue corresponding to each feed oil pressing; Pretreatment stage analysis module: used to analyze the historical data of the pretreatment stage of various types of fat tissues corresponding to each feed oil pressing, and obtain the crushing work data of each crushing process corresponding to each same crushing initial data, the cooking work data of each cooking process corresponding to each same cooking initial data, and the dehydration work data of each dehydration process corresponding to each same dehydration initial data.

[0005] Preliminary analysis module: used to analyze the crushing work data of each crushing process corresponding to each identical crushing initial data, the cooking work data of each cooking process corresponding to each identical cooking initial data, and the dehydration work data of each dehydration process corresponding to each identical dehydration initial data, so as to obtain data comparison models corresponding to each crushing initial data, each cooking initial data, and each dehydration initial data corresponding to each type of fat tissue; Risk management and control module: It is used to monitor and obtain the process data in the pretreatment stage, and to perform data model matching analysis with the data comparison models corresponding to the initial crushing data, initial cooking data and initial dehydration data of each type of fat tissue, so as to obtain the risk management and control results of the pretreatment stage.

[0006] In the preferred embodiment of this scheme, the specific implementation method of the data partitioning module is as follows: Establishing a data extraction relationship between the data partitioning module and the database, extracting the historical data of each feed oil pressing stored in the database, performing preliminary data acquisition on the historical data of each feed oil pressing, and obtaining the type of adipose tissue of each feed oil pressing, wherein the type of adipose tissue refers to different parts corresponding to different animals; By statistically dividing the types of adipose tissues corresponding to each feed oil pressing, the feed oil pressing corresponding to each type of adipose tissue is obtained. The feed oil pressing is divided according to the pressing stages corresponding to the feed oil pressing, and the pressing stages corresponding to the feed oil pressing are obtained. The pressing stages include a pretreatment stage, a pressing stage and a refining stage in sequence, and the pretreatment stage historical data of each type of adipose tissue corresponding to each feed oil pressing is obtained by matching.

[0007] In the preferred embodiment of this scheme, the specific implementation method of the pre-processing stage analysis module is as follows: Establishing a data extraction relationship between the pretreatment stage analysis module and the database, extracting the standard comparison data of various types of fat tissues corresponding to the pretreatment stage stored in the database, wherein the standard comparison data includes the average volume of standard fat tissue fragments, the standard first unsaturated fatty acid content, the standard second unsaturated fatty acid content and the terminal water content of standard fat tissue fragments; The pretreatment stage is divided into a crushing process, a cooking process and a dehydration process through the sub-processes corresponding to the pretreatment stage, and the historical data corresponding to the crushing process, the cooking process and the dehydration process are matched and obtained, wherein the historical data includes initial data, end data and working data; The initial crushing data corresponding to the crushing process is the total volume of fat tissue, the final crushing data is the average volume of fat tissue fragments, and the crushing working data is the crushing time and crushing power; The initial data of the cooking process corresponding to the cooking process is the average volume of the fat tissue fragments, the end data of the cooking process includes the content of the first unsaturated fatty acid, and the working data includes the cooking time and the cooking temperature; The dehydration initial data corresponding to the dehydration process include the average volume of the fat tissue fragments and the initial water content of the fat tissue fragments, the dehydration initial data are the final water content and the second unsaturated fatty acid content of the fat tissue fragments, and the dehydration working data include the centrifugal force and the centrifugal time; The average volume of the standard fat tissue fragments, the standard unsaturated fatty acid content and the terminal water content of the standard fat tissue fragments corresponding to the standard comparison data are matched with the crushing end data, the cooking end data and the dehydration initial data corresponding to each crushing process, each cooking process and each dehydration process, so as to obtain each crushing process, each cooking process and each dehydration process that are consistent with the standard comparison data, and each crushing process, each cooking process and each dehydration process that are consistent with the standard comparison data are recorded as each preliminary comparison crushing process, each preliminary comparison cooking process and each preliminary comparison dehydration process; Obtain the initial data and working data corresponding to each preliminary comparison crushing process, each preliminary comparison cooking process and each preliminary comparison dehydration process, perform data matching processing on the initial data and working data corresponding to each preliminary comparison crushing process, each preliminary comparison cooking process and each preliminary comparison dehydration process, and obtain the crushing working data of each preliminary comparison crushing process corresponding to each same crushing initial data, the cooking working data of each preliminary comparison cooking process corresponding to each same cooking initial data, and the dehydration working data of each preliminary comparison dehydration process corresponding to each same dehydration initial data.

[0008] In the preferred embodiment of this scheme, the specific implementation method of the preliminary analysis module is as follows: By calculating the formula , calculate the efficiency coefficient of each initial crushing data corresponding to each preliminary comparison crushing process, and record it as the efficiency coefficient of each initial crushing data corresponding to each preliminary comparison crushing process ; By calculating the formula , calculate the efficiency coefficient of each initial cooking data corresponding to each preliminary comparative cooking process work data, and record it as the efficiency coefficient of each initial cooking data corresponding to each preliminary comparative cooking process ; By calculating the formula , calculate the efficiency coefficient of the dehydration work data of each preliminary comparative dehydration process corresponding to each same dehydration initial data, and record it as the efficiency coefficient of each preliminary comparative dehydration process corresponding to each same dehydration initial data ,in and They are respectively represented as the crushing time and crushing power of each preliminary comparison crushing process corresponding to each same initial crushing data, and They are respectively represented by the preset crushing time weight and crushing power weight, and They are respectively represented as the cooking time and cooking temperature of each preliminary comparative cooking process corresponding to each same initial cooking data, and Respectively represented as the preset cooking time weight and cooking temperature weight, and They are respectively represented as the centrifugal force and centrifugal time of each preliminary comparative dehydration process corresponding to each same initial dehydration data, and They are respectively represented by the preset centrifugal force weight and centrifugal time weight, It is represented by the number of each identical broken initial data. It is represented by the number of the same initial data of cooking. It is represented by the number of the same dehydration initial data. It is represented by the number of each preliminary comparison crushing process corresponding to each same initial crushing data. It is represented by the number of each preliminary comparative cooking process corresponding to each same initial cooking data. It is represented by the number of each preliminary comparative dehydration process corresponding to each same dehydration initial data. Expressed as the natural light constant; Compare the efficiency coefficients of each preliminary comparative crushing process corresponding to each identical initial crushing data with the standard efficiency coefficients corresponding to the preset crushing process, select each preliminary comparative crushing process corresponding to each identical initial crushing data with an efficiency coefficient less than the standard efficiency coefficient, and record them as each effective comparative crushing process corresponding to each identical initial crushing data; compare the efficiency coefficients of each preliminary comparative cooking process corresponding to each identical initial steaming data with the standard efficiency coefficients corresponding to the preset steaming process, select each preliminary comparative cooking process corresponding to each identical initial steaming data with an efficiency coefficient greater than the standard efficiency coefficient, and record them as each effective comparative cooking process corresponding to each identical initial steaming data; compare the efficiency coefficients of each preliminary comparative dehydration process corresponding to each identical initial dehydration data with the standard efficiency coefficients corresponding to the preset dehydration process, select each preliminary comparative dehydration process corresponding to each identical initial dehydration data with an efficiency coefficient less than the standard efficiency coefficient, and record them as each effective comparative dehydration process corresponding to each identical initial dehydration data; A data model is established based on the crushing work data corresponding to each effective comparison crushing process corresponding to each identical crushing initial data, so as to obtain a data comparison model corresponding to each crushing initial data; a data model is established based on the cooking work data corresponding to each effective comparison cooking process corresponding to each identical cooking initial data, so as to obtain a data comparison model corresponding to each cooking initial data; a data model is established based on the dehydration work data corresponding to each effective comparison dehydration process corresponding to each identical dehydration initial data, so as to obtain a data comparison model corresponding to each dehydration initial data; The data comparison model corresponding to each crushing initial data, each cooking initial data and each dehydration initial data corresponding to each adipose tissue type is obtained statistically.

[0009] In the preferred solution of this scheme, the specific implementation method of the risk control module is as follows: By obtaining the types of fat tissue of the fat tissue of the fat pressed in the feed to be evaluated, screening is performed according to the types of fat tissue of the fat pressed in the feed to be evaluated, and a data comparison model corresponding to each initial crushing data, each initial cooking data, and each initial dehydration data corresponding to the fat pressed in the feed to be evaluated is obtained; By acquiring the data of the pre-processing stage corresponding to the fat tissue of the feed oil to be evaluated, the initial data and working data of the sub-process to be evaluated corresponding to the fat tissue of the feed oil to be evaluated are obtained, and the data comparison set corresponding to the sub-process to be evaluated is obtained by screening the initial data of the sub-process to be evaluated corresponding to the fat tissue of the feed oil to be evaluated; Compare the working data of the sub-process to be evaluated corresponding to the fat tissue of the feed to be evaluated with the data comparison set. If the working data is consistent with any working data in the data comparison set, it means that the sub-process to be evaluated corresponding to the fat tissue of the feed to be evaluated is working normally and the next process can be carried out. The normal working of the sub-process to be evaluated corresponding to the fat tissue of the feed to be evaluated is recorded as the risk warning result of fat pressing. If the working data is inconsistent with any working data in the data comparison set; Scenario 1: If the working data of the sub-process to be evaluated of the feed oil and fat pressing fat tissue can be changed to match any working data in the data comparison set, the sub-process to be evaluated of the feed oil and fat pressing fat tissue will be reprocessed, and the reprocessing of the sub-process to be evaluated of the feed oil and fat pressing fat tissue will be recorded as the feed oil and fat pressing risk warning result; Scenario 2: If the working data of the sub-process to be evaluated of feed oil and fat pressing fat tissue cannot be made consistent with any working data in the data comparison set by changing the working data of the sub-process to be evaluated of feed oil and fat pressing fat tissue, an alarm will be performed on the sub-process to be evaluated of feed oil and fat pressing fat tissue, and the alarm will be recorded as a risk warning result of feed oil and fat pressing.

[0010] Compared with the prior art, the present invention has the following beneficial effects: The present invention acquires historical data of feed oil pressing, performs machine learning on the historical data of feed oil pressing, establishes data comparison models corresponding to initial crushing data, initial cooking data and initial dehydration data of each type of fat tissue, thereby providing a data basis for risk control of feed oil pressing, and compares and processes the working data of the sub-process to be evaluated, thereby improving the practicability of risk control for improving the quality and output of feed oil pressing, which is beneficial to improving the competitiveness and market position of the enterprise. BRIEF DESCRIPTION OF THE DRAWINGS

[0011] The present invention is further described using the accompanying drawings, but the embodiments in the accompanying drawings do not constitute any limitation to the present invention. A person skilled in the art can obtain other drawings based on the following drawings without creative work.

[0012] Figure 1 This is a schematic diagram of module connection according to an embodiment of the present invention. DETAILED DESCRIPTION

[0013] The following will be combined with the drawings in the embodiments of the present invention to clearly and completely describe the technical solutions in the embodiments of the present invention. Obviously, the described embodiments 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 creative work are within the scope of protection of the present invention.

[0014] See also Figure 1 The present invention provides a feed oil squeezing risk management and control system based on machine learning, which includes a data partitioning module, a preprocessing stage analysis module, a preliminary analysis module and a risk management and control module: The data partitioning module is connected to the preprocessing stage analysis module, the preprocessing stage analysis module is connected to the preliminary analysis module, and the preliminary analysis module is connected to the risk management and control module.

[0015] The data division module is used to obtain the historical data of feed oil pressing and comprehensively divide it according to the pressing process, so as to obtain the historical data of the pretreatment stage of each type of fat tissue corresponding to each feed oil pressing; Furthermore, the specific implementation of the data partitioning module is as follows: Establishing a data extraction relationship between the data partitioning module and the database, extracting the historical data of each feed oil pressing stored in the database, performing preliminary data acquisition on the historical data of each feed oil pressing, and obtaining the type of adipose tissue of each feed oil pressing, wherein the type of adipose tissue refers to different parts corresponding to different animals; By statistically dividing the types of adipose tissues corresponding to each feed oil pressing, the feed oil pressing corresponding to each type of adipose tissue is obtained. The feed oil pressing is divided according to the pressing stages corresponding to the feed oil pressing, and the pressing stages corresponding to the feed oil pressing are obtained. The pressing stages include a pretreatment stage, a pressing stage and a refining stage in sequence, and the pretreatment stage historical data of each type of adipose tissue corresponding to each feed oil pressing is obtained by matching.

[0016] Pretreatment stage analysis module: used to analyze the historical data of the pretreatment stage of various types of fat tissues corresponding to each feed oil pressing, and obtain the crushing work data of each crushing process corresponding to each same crushing initial data, the cooking work data of each cooking process corresponding to each same cooking initial data, and the dehydration work data of each dehydration process corresponding to each same dehydration initial data.

[0017] Furthermore, the specific execution method of the preprocessing stage analysis module is as follows: Establishing a data extraction relationship between the pretreatment stage analysis module and the database, extracting the standard comparison data of various types of fat tissues corresponding to the pretreatment stage stored in the database, wherein the standard comparison data includes the average volume of standard fat tissue fragments, the standard first unsaturated fatty acid content, the standard second unsaturated fatty acid content and the terminal water content of standard fat tissue fragments; The pretreatment stage is divided into a crushing process, a cooking process and a dehydration process through the sub-processes corresponding to the pretreatment stage, and the historical data corresponding to the crushing process, the cooking process and the dehydration process are matched and obtained, wherein the historical data includes initial data, end data and working data; The initial crushing data corresponding to the crushing process is the total volume of fat tissue, the final crushing data is the average volume of fat tissue fragments, and the crushing working data is the crushing time and crushing power; The initial data of the cooking process corresponding to the cooking process is the average volume of the fat tissue fragments, the end data of the cooking process includes the content of the first unsaturated fatty acid, and the working data includes the cooking time and the cooking temperature; The dehydration initial data corresponding to the dehydration process include the average volume of the fat tissue fragments and the initial water content of the fat tissue fragments, the dehydration initial data are the final water content and the second unsaturated fatty acid content of the fat tissue fragments, and the dehydration working data include the centrifugal force and the centrifugal time; The average volume of the standard fat tissue fragments, the standard unsaturated fatty acid content and the terminal water content of the standard fat tissue fragments corresponding to the standard comparison data are matched with the crushing end data, the cooking end data and the dehydration initial data corresponding to each crushing process, each cooking process and each dehydration process, so as to obtain each crushing process, each cooking process and each dehydration process that are consistent with the standard comparison data, and each crushing process, each cooking process and each dehydration process that are consistent with the standard comparison data are recorded as each preliminary comparison crushing process, each preliminary comparison cooking process and each preliminary comparison dehydration process; Obtain the initial data and working data corresponding to each preliminary comparison crushing process, each preliminary comparison cooking process and each preliminary comparison dehydration process, perform data matching processing on the initial data and working data corresponding to each preliminary comparison crushing process, each preliminary comparison cooking process and each preliminary comparison dehydration process, and obtain the crushing working data of each preliminary comparison crushing process corresponding to each same crushing initial data, the cooking working data of each preliminary comparison cooking process corresponding to each same cooking initial data, and the dehydration working data of each preliminary comparison dehydration process corresponding to each same dehydration initial data.

[0018] Preliminary analysis module: used to analyze the crushing work data of each crushing process corresponding to each identical crushing initial data, the cooking work data of each cooking process corresponding to each identical cooking initial data, and the dehydration work data of each dehydration process corresponding to each identical dehydration initial data, so as to obtain data comparison models corresponding to each crushing initial data, each cooking initial data, and each dehydration initial data corresponding to each type of fat tissue; Furthermore, the specific execution method of the preliminary analysis module is as follows: By calculating the formula , calculate the efficiency coefficient of each initial crushing data corresponding to each preliminary comparison crushing process, and record it as the efficiency coefficient of each initial crushing data corresponding to each preliminary comparison crushing process ; By calculating the formula , calculate the efficiency coefficient of each initial cooking data corresponding to each preliminary comparative cooking process work data, and record it as the efficiency coefficient of each initial cooking data corresponding to each preliminary comparative cooking process ; By calculating the formula , calculate the efficiency coefficient of the dehydration work data of each preliminary comparative dehydration process corresponding to each same dehydration initial data, and record it as the efficiency coefficient of each preliminary comparative dehydration process corresponding to each same dehydration initial data ,in and They are respectively represented as the crushing time and crushing power of each preliminary comparison crushing process corresponding to each same initial crushing data, and They are respectively represented by the preset crushing time weight and crushing power weight, and They are respectively represented as the cooking time and cooking temperature of each preliminary comparative cooking process corresponding to each same initial cooking data, and Respectively represented as the preset cooking time weight and cooking temperature weight, and They are respectively represented as the centrifugal force and centrifugal time of each preliminary comparative dehydration process corresponding to each same initial dehydration data, and They are respectively represented by the preset centrifugal force weight and centrifugal time weight, It is represented by the number of each identical broken initial data. It is represented by the number of the same initial data of cooking. It is represented by the number of the same dehydration initial data. It is represented by the number of each preliminary comparison crushing process corresponding to each same initial crushing data. It is represented by the number of each preliminary comparative cooking process corresponding to each same initial cooking data. It is represented by the number of each preliminary comparative dehydration process corresponding to each same dehydration initial data. Expressed as the natural light constant; Compare the efficiency coefficients of each preliminary comparative crushing process corresponding to each identical initial crushing data with the standard efficiency coefficients corresponding to the preset crushing process, select each preliminary comparative crushing process corresponding to each identical initial crushing data with an efficiency coefficient less than the standard efficiency coefficient, and record them as each effective comparative crushing process corresponding to each identical initial crushing data; compare the efficiency coefficients of each preliminary comparative cooking process corresponding to each identical initial steaming data with the standard efficiency coefficients corresponding to the preset steaming process, select each preliminary comparative cooking process corresponding to each identical initial steaming data with an efficiency coefficient greater than the standard efficiency coefficient, and record them as each effective comparative cooking process corresponding to each identical initial steaming data; compare the efficiency coefficients of each preliminary comparative dehydration process corresponding to each identical initial dehydration data with the standard efficiency coefficients corresponding to the preset dehydration process, select each preliminary comparative dehydration process corresponding to each identical initial dehydration data with an efficiency coefficient less than the standard efficiency coefficient, and record them as each effective comparative dehydration process corresponding to each identical initial dehydration data; A data model is established based on the crushing work data corresponding to each effective comparison crushing process corresponding to each identical crushing initial data, so as to obtain a data comparison model corresponding to each crushing initial data; a data model is established based on the cooking work data corresponding to each effective comparison cooking process corresponding to each identical cooking initial data, so as to obtain a data comparison model corresponding to each cooking initial data; a data model is established based on the dehydration work data corresponding to each effective comparison dehydration process corresponding to each identical dehydration initial data, so as to obtain a data comparison model corresponding to each dehydration initial data; The data comparison model corresponding to each crushing initial data, each cooking initial data and each dehydration initial data corresponding to each adipose tissue type is obtained statistically.

[0019] Risk management and control module: It is used to monitor and obtain the process data in the pretreatment stage, and to perform data model matching analysis with the data comparison models corresponding to the initial crushing data, initial cooking data and initial dehydration data of each type of fat tissue, so as to obtain the risk management and control results of the pretreatment stage.

[0020] Furthermore, the specific implementation of the risk management module is as follows: By obtaining the types of fat tissue of the fat tissue of the fat pressed in the feed to be evaluated, screening is performed according to the types of fat tissue of the fat pressed in the feed to be evaluated, and a data comparison model corresponding to each initial crushing data, each initial cooking data, and each initial dehydration data corresponding to the fat pressed in the feed to be evaluated is obtained; By acquiring the data of the pre-processing stage corresponding to the fat tissue of the feed oil to be evaluated, the initial data and working data of the sub-process to be evaluated corresponding to the fat tissue of the feed oil to be evaluated are obtained, and the data comparison set corresponding to the sub-process to be evaluated is obtained by screening the initial data of the sub-process to be evaluated corresponding to the fat tissue of the feed oil to be evaluated; Compare the working data of the sub-process to be evaluated corresponding to the fat tissue of the feed to be evaluated with the data comparison set. If the working data is consistent with any working data in the data comparison set, it means that the sub-process to be evaluated corresponding to the fat tissue of the feed to be evaluated is working normally and the next process can be carried out. The normal working of the sub-process to be evaluated corresponding to the fat tissue of the feed to be evaluated is recorded as the risk warning result of fat pressing. If the working data is inconsistent with any working data in the data comparison set; Scenario 1: If the working data of the sub-process to be evaluated of the feed oil and fat pressing fat tissue can be changed to match any working data in the data comparison set, the sub-process to be evaluated of the feed oil and fat pressing fat tissue will be reprocessed, and the reprocessing of the sub-process to be evaluated of the feed oil and fat pressing fat tissue will be recorded as the feed oil and fat pressing risk warning result; Scenario 2: If the working data of the sub-process to be evaluated of feed oil and fat pressing fat tissue cannot be made consistent with any working data in the data comparison set by changing the working data of the sub-process to be evaluated of feed oil and fat pressing fat tissue, an alarm will be performed on the sub-process to be evaluated of feed oil and fat pressing fat tissue, and the alarm will be recorded as a risk warning result of feed oil and fat pressing.

[0021] The above are all preferred embodiments of the present application, and the protection scope of the present application is not limited thereto. Therefore, any equivalent changes made according to the structure, shape, and principle of the present application should be included in the protection scope of the present application.

Claims

1. A feed oil pressing risk management and control system based on machine learning, characterized in that: include: Data division module: used to obtain historical data of feed oil pressing and comprehensively divide it according to the pressing process, so as to obtain historical data of the pretreatment stage of each type of fat tissue corresponding to each feed oil pressing; Pretreatment stage analysis module: used to analyze the historical data of the pretreatment stage of various types of fat tissues corresponding to each feed oil pressing, and obtain the crushing work data of each crushing process corresponding to each same crushing initial data, the cooking work data of each cooking process corresponding to each same cooking initial data, and the dehydration work data of each dehydration process corresponding to each same dehydration initial data; Preliminary analysis module: used to analyze the crushing work data of each crushing process corresponding to each identical crushing initial data, the cooking work data of each cooking process corresponding to each identical cooking initial data, and the dehydration work data of each dehydration process corresponding to each identical dehydration initial data, so as to obtain data comparison models corresponding to each crushing initial data, each cooking initial data, and each dehydration initial data corresponding to each type of fat tissue; Risk management and control module: It is used to monitor and obtain the process data in the pretreatment stage, and to perform data model matching analysis with the data comparison models corresponding to the initial crushing data, initial cooking data and initial dehydration data of each type of fat tissue, so as to obtain the risk management and control results of the pretreatment stage.

2. The feed oil pressing risk management and control system based on machine learning according to claim 1, characterized in that: The specific implementation method of the data partitioning module is as follows: Establishing a data extraction relationship between the data partitioning module and the database, extracting the historical data of each feed oil pressing stored in the database, performing preliminary data acquisition on the historical data of each feed oil pressing, and obtaining the type of adipose tissue of each feed oil pressing, wherein the type of adipose tissue refers to different parts corresponding to different animals; By statistically dividing the types of adipose tissues corresponding to each feed oil pressing, the feed oil pressing corresponding to each type of adipose tissue is obtained. The feed oil pressing is divided according to the pressing stages corresponding to the feed oil pressing, and the pressing stages corresponding to the feed oil pressing are obtained. The pressing stages include a pretreatment stage, a pressing stage and a refining stage in sequence, and the pretreatment stage historical data of each type of adipose tissue corresponding to each feed oil pressing is obtained by matching.

3. A feed oil pressing risk management and control system based on machine learning according to claim 2, characterized in that: The specific execution mode of the pre-processing stage analysis module is as follows: Establishing a data extraction relationship between the pretreatment stage analysis module and the database, extracting the standard comparison data of various types of fat tissues corresponding to the pretreatment stage stored in the database, wherein the standard comparison data includes the average volume of standard fat tissue fragments, the standard first unsaturated fatty acid content, the standard second unsaturated fatty acid content and the terminal water content of standard fat tissue fragments; The pretreatment stage is divided into a crushing process, a cooking process and a dehydration process through the sub-processes corresponding to the pretreatment stage, and the historical data corresponding to the crushing process, the cooking process and the dehydration process are matched and obtained, wherein the historical data includes initial data, end data and working data; The initial crushing data corresponding to the crushing process is the total volume of fat tissue, the final crushing data is the average volume of fat tissue fragments, and the crushing working data is the crushing time and crushing power; The initial data of the cooking process corresponding to the cooking process is the average volume of the fat tissue fragments, the end data of the cooking process includes the content of the first unsaturated fatty acid, and the working data includes the cooking time and the cooking temperature; The dehydration initial data corresponding to the dehydration process include the average volume of the fat tissue fragments and the initial water content of the fat tissue fragments, the dehydration initial data are the final water content and the second unsaturated fatty acid content of the fat tissue fragments, and the dehydration working data include the centrifugal force and the centrifugal time; The average volume of the standard fat tissue fragments, the standard unsaturated fatty acid content and the terminal water content of the standard fat tissue fragments corresponding to the standard comparison data are matched with the crushing end data, the cooking end data and the dehydration initial data corresponding to each crushing process, each cooking process and each dehydration process, so as to obtain each crushing process, each cooking process and each dehydration process that are consistent with the standard comparison data, and each crushing process, each cooking process and each dehydration process that are consistent with the standard comparison data are recorded as each preliminary comparison crushing process, each preliminary comparison cooking process and each preliminary comparison dehydration process; Obtain the initial data and working data corresponding to each preliminary comparison crushing process, each preliminary comparison cooking process and each preliminary comparison dehydration process, perform data matching processing on the initial data and working data corresponding to each preliminary comparison crushing process, each preliminary comparison cooking process and each preliminary comparison dehydration process, and obtain the crushing working data of each preliminary comparison crushing process corresponding to each same crushing initial data, the cooking working data of each preliminary comparison cooking process corresponding to each same cooking initial data, and the dehydration working data of each preliminary comparison dehydration process corresponding to each same dehydration initial data.

4. The feed oil pressing risk management and control system based on machine learning according to claim 3 is characterized in that: The specific implementation method of the preliminary analysis module is as follows: By calculating the formula , calculate the efficiency coefficient of each initial crushing data corresponding to each preliminary comparison crushing process, and record it as the efficiency coefficient of each initial crushing data corresponding to each preliminary comparison crushing process ; By calculating the formula , calculate the efficiency coefficient of each initial cooking data corresponding to each preliminary comparative cooking process work data, and record it as the efficiency coefficient of each initial cooking data corresponding to each preliminary comparative cooking process ; By calculating the formula , calculate the efficiency coefficient of the dehydration work data of each preliminary comparative dehydration process corresponding to each same dehydration initial data, and record it as the efficiency coefficient of each preliminary comparative dehydration process corresponding to each same dehydration initial data ,in and They are respectively represented as the crushing time and crushing power of each preliminary comparison crushing process corresponding to each same initial crushing data, and They are respectively represented by the preset crushing time weight and crushing power weight, and They are respectively represented as the cooking time and cooking temperature of each preliminary comparative cooking process corresponding to each same initial cooking data, and Respectively represented as the preset cooking time weight and cooking temperature weight, and They are respectively represented as the centrifugal force and centrifugal time of each preliminary comparative dehydration process corresponding to each same initial dehydration data, and They are respectively represented by the preset centrifugal force weight and centrifugal time weight, It is represented by the number of each identical broken initial data. It is represented by the number of the same initial data of cooking. It is represented by the number of the same dehydration initial data. It is represented by the number of each preliminary comparison crushing process corresponding to each same initial crushing data. It is represented by the number of each preliminary comparative cooking process corresponding to each same initial cooking data. It is represented by the number of each preliminary comparative dehydration process corresponding to each same dehydration initial data. Expressed as the natural light constant; Compare the efficiency coefficients of each preliminary comparative crushing process corresponding to each identical initial crushing data with the standard efficiency coefficients corresponding to the preset crushing process, select each preliminary comparative crushing process corresponding to each identical initial crushing data with an efficiency coefficient less than the standard efficiency coefficient, and record them as each effective comparative crushing process corresponding to each identical initial crushing data; compare the efficiency coefficients of each preliminary comparative cooking process corresponding to each identical initial steaming data with the standard efficiency coefficients corresponding to the preset steaming process, select each preliminary comparative cooking process corresponding to each identical initial steaming data with an efficiency coefficient greater than the standard efficiency coefficient, and record them as each effective comparative cooking process corresponding to each identical initial steaming data; compare the efficiency coefficients of each preliminary comparative dehydration process corresponding to each identical initial dehydration data with the standard efficiency coefficients corresponding to the preset dehydration process, select each preliminary comparative dehydration process corresponding to each identical initial dehydration data with an efficiency coefficient less than the standard efficiency coefficient, and record them as each effective comparative dehydration process corresponding to each identical initial dehydration data; A data model is established based on the crushing work data corresponding to each effective comparison crushing process corresponding to each identical crushing initial data, so as to obtain a data comparison model corresponding to each crushing initial data; a data model is established based on the cooking work data corresponding to each effective comparison cooking process corresponding to each identical cooking initial data, so as to obtain a data comparison model corresponding to each cooking initial data; a data model is established based on the dehydration work data corresponding to each effective comparison dehydration process corresponding to each identical dehydration initial data, so as to obtain a data comparison model corresponding to each dehydration initial data; The data comparison model corresponding to each crushing initial data, each cooking initial data and each dehydration initial data corresponding to each adipose tissue type is obtained statistically.

5. The feed oil pressing risk management and control system based on machine learning according to claim 4, characterized in that: The specific implementation method of the risk management module is as follows: By obtaining the types of fat tissue of the fat tissue of the fat pressed in the feed to be evaluated, screening is performed according to the types of fat tissue of the fat pressed in the feed to be evaluated, and a data comparison model corresponding to each initial crushing data, each initial cooking data, and each initial dehydration data corresponding to the fat pressed in the feed to be evaluated is obtained; By acquiring the data of the pre-processing stage corresponding to the fat tissue of the feed oil to be evaluated, the initial data and working data of the sub-process to be evaluated corresponding to the fat tissue of the feed oil to be evaluated are obtained, and the data comparison set corresponding to the sub-process to be evaluated is obtained by screening the initial data of the sub-process to be evaluated corresponding to the fat tissue of the feed oil to be evaluated; Compare the working data of the sub-process to be evaluated corresponding to the fat tissue of the feed to be evaluated with the data comparison set. If the working data is consistent with any working data in the data comparison set, it means that the sub-process to be evaluated corresponding to the fat tissue of the feed to be evaluated is working normally and the next process can be carried out. The normal working of the sub-process to be evaluated corresponding to the fat tissue of the feed to be evaluated is recorded as the risk warning result of fat pressing. If the working data is inconsistent with any working data in the data comparison set; Scenario 1: If the working data of the sub-process to be evaluated of the feed oil and fat pressing fat tissue can be changed to match any working data in the data comparison set, the sub-process to be evaluated of the feed oil and fat pressing fat tissue will be reprocessed, and the reprocessing of the sub-process to be evaluated of the feed oil and fat pressing fat tissue will be recorded as the feed oil and fat pressing risk warning result; Scenario 2: If the working data of the sub-process to be evaluated of feed oil and fat pressing fat tissue cannot be made consistent with any working data in the data comparison set by changing the working data of the sub-process to be evaluated of feed oil and fat pressing fat tissue, an alarm will be performed on the sub-process to be evaluated of feed oil and fat pressing fat tissue, and the alarm will be recorded as a risk warning result of feed oil and fat pressing.

Images