A method and system for optimizing protein extraction process of macadamia nuts

Through the combination of the Internet of Things and hierarchical analysis method, the parameters of macadamia protein extraction equipment are monitored and analyzed in real time, semantic network images are generated, abnormal unit modules are traced and analyzed and adjusted and optimized, which solves the problem of parameter coordinated adjustment in the existing technology and improves the problem of improved parameters Extraction efficiency and product quality.

CN118800346BActive Publication Date: 2024-11-08SOUTH SUBTROPICAL CROP RES INST CHINA ACAD OF TROPICAL AGRI SCI
View PDF 0 Cites 0 Cited by

Patent Information

Application Number
CN202411281615.6
Authority / Receiving Office
CN · China
Patent Type
Patents(China)
Current Assignee / Owner
Filing Date
2024-09-13
Publication Date
2024-11-08
Estimated Expiration
2044-09-13

AI Technical Summary

Technical Problem

The existing intelligent regulation methods are difficult to achieve coordinated regulation between parameters in the macadamia protein extraction process, resulting in large fluctuations in extraction efficiency and product quality and lack of global optimization.

Method used

The Internet of Things and hierarchical analysis method are used to monitor and compare the parameters of protein extraction equipment in real time, generate semantic network images, trace the source and analyze abnormal unit modules, and achieve coordinated regulation of parameters and steady-state operation of the equipment through adjustment optimization processing.

Benefits of technology

Through real-time monitoring, data analysis and intelligent decision-making, macadamia protein extraction efficiency, product quality and production safety are improved, and labor costs and production losses caused by equipment failures are reduced.

✦ Generated by Eureka AI based on patent content.

Smart Images

  • Figure CN118800346B_ABST
    Figure CN118800346B_ABST
Patent Text Reader

Abstract

The present invention relates to the technical field of protein extraction process optimization, and in particular to a method and system for optimizing protein extraction process of macadamia nuts. Compare and analyze the actual parameter value sets of various protein extraction parameters within a preset time period with the preset parameter value sets; perform correlation analysis on each unit module and various protein extraction parameters based on the hierarchical analysis method, and generate a semantic network image between each unit module and various protein extraction parameters; based on the semantic network image, and in combination with the protein extraction parameters whose status is a non-steady state, perform a traceability analysis on each unit module in the protein extraction equipment, and obtain a suspicious unit module in the protein extraction equipment; further analyze the suspicious unit module in the protein extraction equipment, and judge that the suspicious unit module in the operation state has experienced an abnormal operation. The present invention effectively improves the protein extraction efficiency, product quality and production safety of macadamia nuts.
Need to check novelty before this filing date? Find Prior Art

Description

Technical Field

[0001] The present invention relates to the technical field of protein extraction process optimization, and in particular to a method and system for optimizing protein extraction process of macadamia nuts. Background Art

[0002] Macadamia nuts are rich in protein and are a potential source of high-quality protein. The extraction process of macadamia protein is a complex process involving precise control of multiple parameters, such as temperature, pH value, stirring speed, time, etc. Traditional manual operation methods make it difficult to ensure the stability and accuracy of parameters, resulting in large fluctuations in extraction efficiency and product quality. With the rapid development of intelligent manufacturing technology, the application of intelligent control technology to macadamia protein extraction equipment to achieve real-time monitoring and automatic adjustment of parameters has become an effective way to improve extraction efficiency, stabilize product quality, and reduce labor costs. However, most of the existing intelligent control methods are limited to the control of a single parameter, lack global optimization of the entire extraction process, and are difficult to achieve coordinated regulation between parameters. Therefore, it is necessary to further optimize the macadamia protein extraction process to improve extraction efficiency and product quality. Summary of the invention

[0003] The present invention overcomes the deficiencies of the prior art and provides a method and system for optimizing the protein extraction process of macadamia nuts.

[0004] To achieve the above-mentioned purpose, the technical solution adopted by the present invention is:

[0005] The first aspect of the present invention discloses a method for optimizing the extraction process of macadamia protein, comprising the following steps:

[0006] Obtaining a preset control scheme for a protein extraction device, obtaining preset parameter values ​​of various protein extraction parameters of the protein extraction device at a number of preset working time nodes according to the preset control scheme, and obtaining a set of preset parameter values ​​of various protein extraction parameters within a preset time period;

[0007] Based on the Internet of Things, actual parameter values ​​of various protein extraction parameters in the protein extraction equipment are collected at several preset working time nodes to obtain a number set of actual parameter values ​​of various protein extraction parameters within a preset time period;

[0008] Compare and analyze the actual parameter value sets of various protein extraction parameters within a preset time period with the preset parameter value sets, and analyze the status of various protein extraction parameters in the protein extraction device; perform correlation analysis on each unit module and various protein extraction parameters based on the hierarchical analysis method, and generate a semantic network image between each unit module and various protein extraction parameters;

[0009] Obtaining protein extraction parameters in a non-steady-state state in a protein extraction device; performing a traceability analysis on each unit module in the protein extraction device based on the semantic network image and in combination with the protein extraction parameters in a non-steady-state state, and obtaining a unit module with a suspicious operating state in the protein extraction device;

[0010] Further analysis is performed on the suspicious operation status unit module in the protein extraction equipment to determine that the suspicious operation status unit module has experienced abnormal operation; if an abnormal operation condition occurs, the abnormal operation status unit module is adjusted and optimized.

[0011] Preferably, the actual parameter value sets of various protein extraction parameters within a preset time period are compared and analyzed with the preset parameter value sets to obtain the status of various protein extraction parameters in the protein extraction device, specifically:

[0012] Calculating the Pearson correlation coefficient between the actual parameter value set and the preset parameter value set of various protein extraction parameters within a preset time period;

[0013] Compare the Pearson correlation coefficients between the actual parameter value sets of various protein extraction parameters within a preset time period and the preset parameter value sets with the preset coefficient values;

[0014] If the Pearson correlation coefficient between the actual parameter value set and the preset parameter value set of a certain protein extraction parameter within a preset time period is greater than a preset coefficient value, the state of the protein extraction parameter is defined as a steady state;

[0015] If the Pearson correlation coefficient between the actual parameter value set and the preset parameter value set of a certain protein extraction parameter within a preset time period is not greater than the preset coefficient value, the state of the protein extraction parameter is defined as a non-steady-state state.

[0016] Preferably, the correlation analysis between each unit module and various protein extraction parameters is performed based on the hierarchical analysis method to generate a semantic network image between each unit module and various protein extraction parameters, specifically:

[0017] Obtaining the device manual of the protein extraction device, determining each unit module in the protein extraction device and the usage index information of each unit module according to the device manual; and performing feature extraction processing on various protein extraction parameters to obtain parameter feature information of various protein extraction parameters;

[0018] A hierarchical model is constructed, each unit module and various protein extraction parameters are defined as analysis targets, and the target layer of the hierarchical model is obtained;

[0019] The usage index information of each unit module and the parameter characteristic information of various protein extraction parameters are defined as evaluation indicators to obtain the criterion layer of the hierarchical model;

[0020] and refining the criterion layer to obtain a number of refined evaluation indicators, and generating an indicator layer of a hierarchical structure model according to the number of refined evaluation indicators;

[0021] Obtaining the importance of each detailed evaluation indicator in the indicator layer, and assigning a weight value to each detailed evaluation indicator according to the importance of each detailed evaluation indicator;

[0022] Constructing a judgment matrix according to the importance of each refined evaluation index; calculating the consistency index of the judgment matrix, and judging whether the consistency index of the judgment matrix meets the preset index requirements;

[0023] If the consistency index of the judgment matrix does not meet the preset index requirements, the weight values ​​of each detailed evaluation index are adjusted until the consistency index of the judgment matrix meets the preset index requirements;

[0024] The weight values ​​on the judgment matrix are weighted and summed to obtain the comprehensive score of the corresponding unit module; and so on, until the comprehensive score of each unit module is calculated;

[0025] The unit modules are sorted according to their comprehensive scores, and the correlation ranking between the unit modules and various protein extraction parameters is obtained to generate a correlation ranking result;

[0026] A semantic network image between each unit module and various protein extraction parameters is generated according to the correlation ranking result.

[0027] Preferably, according to the semantic network image, combined with the protein extraction parameters in a non-steady state, a traceability analysis is performed on each unit module in the protein extraction device to obtain a unit module with a suspicious operating state in the protein extraction device, specifically:

[0028] Acquiring protein extraction parameters in a protein extraction device whose state is a non-steady state, and importing the protein extraction parameters in the non-steady state into the semantic network image for pairing analysis;

[0029] Matching in the semantic network image to obtain the correlation between each unit module and the protein extraction parameter whose state is a non-steady state, and obtaining a correlation matching result;

[0030] According to the correlation matching result, the unit module corresponding to the protein extraction parameter with a non-steady-state state having a correlation greater than a preset correlation threshold is obtained to obtain the unit module with a suspicious operation state in the protein extraction device.

[0031] Preferably, further analysis is performed on the suspicious operation status unit module in the protein extraction device to determine whether the suspicious operation status unit module has experienced abnormal operation, specifically:

[0032] Obtaining an operation record log of the protein extraction device, and obtaining historical abnormal operation status events that have occurred in the protein extraction device according to the operation record log;

[0033] Based on the hierarchical clustering algorithm, the historical abnormal operation status events that occurred in the protein extraction equipment are clustered to obtain the clustering results of various abnormal operation status events of the same type;

[0034] The clustering results of various operating state anomalies of the same type are input into the feature pyramid network for feature extraction processing to obtain feature data of various types of operating state anomalies;

[0035] Constructing a knowledge base, dividing the knowledge base into a plurality of storage spaces, assigning a corresponding operation status abnormality type label to each storage space, and storing feature data of a corresponding type of operation status abnormality in the storage space assigned with the corresponding label, thereby obtaining a feature data knowledge base;

[0036] Acquire a unit module with a suspicious operation status in the protein extraction device, and acquire real-time working data of the unit module with a suspicious operation status; import the real-time working data of the unit module with a suspicious operation status into the feature data knowledge base;

[0037] Calculate the similarity between the real-time working data and the characteristic data in each storage space; if the similarity between the real-time working data and the characteristic data in one or more storage spaces is greater than a preset similarity threshold, it means that the suspicious operation status unit module has an abnormal operation condition, and the suspicious operation status unit module is marked as an abnormal operation status unit module;

[0038] And the abnormal operation status type label assigned to the storage space with similarity greater than a preset similarity threshold is obtained to obtain the abnormal operation status type label of the abnormal operation status unit module.

[0039] Preferably, if an abnormal operation occurs, the abnormal operation state unit module is adjusted and optimized, specifically:

[0040] Acquire structural drawing information of the protein extraction device, and construct a structural model diagram of the protein extraction device according to the structural drawing information;

[0041] Acquire the real-time working data of each unit module in the protein extraction device, and perform format conversion processing on the real-time working data of each unit module. After the format conversion is completed, import the real-time working data of each unit module into the structural model diagram to obtain an operation simulation system of the protein extraction device;

[0042] The big data network is searched according to the abnormal operation status type label of the abnormal operation status unit module to obtain several adjustment optimization schemes;

[0043] Based on each adjustment optimization scheme, simulated adjustment and control processing is performed on the operation simulation system of the protein extraction equipment respectively. After the adjustment and control is completed, simulated working data of the abnormal operation state unit module after simulated adjustment and control by each adjustment optimization scheme is obtained;

[0044] Acquire preset working data of the unit module with abnormal operation status, and calculate the data deviation value between the simulated working data of the unit module with abnormal operation status and the preset working data after the simulation and regulation of each adjustment optimization scheme;

[0045] The adjustment optimization scheme corresponding to the minimum data deviation value is obtained, and the adjustment optimization scheme corresponding to the minimum data deviation value is sent to the control terminal of the protein extraction equipment, so that the control terminal can perform substantial adjustment optimization processing on the abnormal operation status unit module based on the adjustment optimization scheme.

[0046] A second aspect of the present invention discloses a macadamia nut protein extraction process optimization system, which includes a memory and a processor. The memory stores a macadamia nut protein extraction process optimization method program. When the macadamia nut protein extraction process optimization method program is executed by the processor, the steps of any one of the macadamia nut protein extraction process optimization methods are implemented.

[0047] The third aspect of the present invention discloses a computer-readable storage medium, which includes a macadamia nut protein extraction process optimization method program. When the macadamia nut protein extraction process optimization method program is executed by a processor, the steps of any one of the macadamia nut protein extraction process optimization methods are implemented.

[0048] The present invention solves the technical defects existing in the background technology, and the present invention has the following beneficial effects: compare and analyze the actual parameter value number set of various protein extraction parameters within a preset time period with the preset parameter value number set, and analyze and obtain the status of various protein extraction parameters in the protein extraction device; perform correlation analysis on each unit module and various protein extraction parameters based on the hierarchical analysis method, and generate a semantic network image between each unit module and various protein extraction parameters; obtain protein extraction parameters in the protein extraction device whose status is a non-steady state; according to the semantic network image, and in combination with the protein extraction parameters whose status is a non-steady state, perform a traceability analysis on each unit module in the protein extraction device, and obtain a suspicious unit module in the protein extraction device; further analyze the suspicious unit module in the protein extraction device, and judge that the suspicious unit module in the operating state has an abnormal operating state; if an abnormal operating state occurs, adjust and optimize the abnormal unit module in the operating state. The present invention effectively improves the protein extraction efficiency, product quality and production safety of macadamia nuts through real-time monitoring, data analysis and intelligent decision-making. BRIEF DESCRIPTION OF THE DRAWINGS

[0049] In order to more clearly illustrate the embodiments of the present invention or the technical solutions in the prior art, the drawings required for use in the embodiments or the description of the prior art will be briefly introduced below. Obviously, the drawings described below are only some embodiments of the present invention. For ordinary technicians in this field, drawings of other embodiments can be obtained based on these drawings without paying creative work.

[0050] Figure 1 A first method flow chart of a method for optimizing a process for extracting protein from macadamia nuts;

[0051] Figure 2 A second method flow chart of a method for optimizing a protein extraction process for macadamia nuts;

[0052] Figure 3 This is a system block diagram of a macadamia protein extraction process optimization system. DETAILED DESCRIPTION

[0053] In order to more clearly understand the above-mentioned purpose, features and advantages of the present invention, the present invention is further described in detail below in conjunction with the accompanying drawings and specific embodiments. It should be noted that the embodiments of the present application and the features in the embodiments can be combined with each other without conflict.

[0054] In the following description, many specific details are set forth to facilitate a full understanding of the present invention. However, the present invention may also be implemented in other ways different from those described herein. Therefore, the protection scope of the present invention is not limited to the specific embodiments disclosed below.

[0055] like Figure 1 As shown, the first aspect of the present invention discloses a method for optimizing the extraction process of macadamia protein, comprising the following steps:

[0056] S102: obtaining a preset control scheme of the protein extraction device, obtaining preset parameter values ​​of various protein extraction parameters of the protein extraction device at a number of preset working time nodes according to the preset control scheme, and obtaining a number set of preset parameter values ​​of various protein extraction parameters within a preset time period;

[0057] S104: collecting actual parameter values ​​of various protein extraction parameters in the protein extraction equipment at several preset working time nodes based on the Internet of Things, and obtaining a number set of actual parameter values ​​of various protein extraction parameters within a preset time period;

[0058] S106: Compare and analyze the actual parameter value sets of various protein extraction parameters within a preset time period with the preset parameter value sets, and analyze to obtain the status of various protein extraction parameters in the protein extraction device; perform correlation analysis on each unit module and various protein extraction parameters based on the hierarchical analysis method, and generate a semantic network image between each unit module and various protein extraction parameters;

[0059] S108: obtaining protein extraction parameters of a protein extraction device in a non-steady state; performing a traceability analysis on each unit module in the protein extraction device according to the semantic network image and in combination with the protein extraction parameters of the non-steady state, and obtaining a unit module with a suspicious operation status in the protein extraction device;

[0060] S110: further analyzing the suspicious operation status unit module in the protein extraction device to determine that the suspicious operation status unit module has experienced abnormal operation; if an abnormal operation condition occurs, adjusting and optimizing the abnormal operation status unit module.

[0061] Among them, the preset control scheme of protein extraction equipment refers to a series of pre-set operating instructions and parameter settings during the protein extraction process. These instructions and parameters are designed by relevant technicians based on specific protein extraction requirements, equipment characteristics and optimized extraction process. The preset control scheme usually includes a series of detailed operating steps such as the starting conditions of the equipment operation, the operation cycle, the operating parameters of each stage (such as temperature, pressure, stirring speed, time, etc.), and the end conditions.

[0062] It should be noted that the system can monitor and compare the equipment operation status in real time and identify the non-steady-state protein extraction parameters based on the preset parameter value set and the actual parameter value set. By using the hierarchical analysis method, the system analyzes the correlation between each unit module and various protein extraction parameters, generates a semantic network image, and intuitively shows the interaction and influence between the parts. This analysis helps to understand the overall performance of the equipment and identify key parameters and their impact on the quality of the final product. Once the unstable parameters in the equipment are detected, the system will further conduct a traceability analysis to locate the suspicious unit modules that may cause abnormal operation. By deeply analyzing the operation of these modules, the system can determine whether abnormal operation occurs. Finally, for the unit modules that are confirmed to have abnormal conditions, the system will take adjustment and optimization measures to restore their normal operation. This not only improves the operation efficiency and product quality of the equipment, but also reduces the production losses caused by equipment failures, and realizes the refined management and intelligent optimization of the protein extraction process. In general, this method effectively improves the protein extraction efficiency, product quality and production safety of macadamia nuts through real-time monitoring, data analysis and intelligent decision-making.

[0063] Preferably, if Figure 2 As shown, the actual parameter value sets of various protein extraction parameters within the preset time period are compared and analyzed with the preset parameter value sets, and the status of various protein extraction parameters in the protein extraction device is obtained by analysis, specifically:

[0064] S202: Calculating the Pearson correlation coefficient between the actual parameter value set and the preset parameter value set of various protein extraction parameters within a preset time period;

[0065] S204: comparing the Pearson correlation coefficient between the actual parameter value set of various protein extraction parameters within the preset time period and the preset parameter value set and the preset coefficient value;

[0066] S206: If the Pearson correlation coefficient between the actual parameter value set and the preset parameter value set of a certain protein extraction parameter within the preset time period is greater than a preset coefficient value, the state of the protein extraction parameter is defined as a steady state;

[0067] S208: If the Pearson correlation coefficient between the actual parameter value set and the preset parameter value set of a certain protein extraction parameter within the preset time period is not greater than the preset coefficient value, the state of the protein extraction parameter is defined as a non-steady state.

[0068] It should be noted that by calculating the Pearson correlation coefficient between the actual parameter value and the preset parameter value, the linear relationship strength between the two data sets can be quantified, and the stability of each parameter in the protein extraction process can be evaluated. The Pearson correlation coefficient ranges from -1 to 1, close to 1 indicates a high positive correlation, close to -1 indicates a high negative correlation, and close to 0 indicates no obvious correlation.

[0069] Preferably, the correlation analysis between each unit module and various protein extraction parameters is performed based on the hierarchical analysis method to generate a semantic network image between each unit module and various protein extraction parameters, specifically:

[0070] Obtaining the device manual of the protein extraction device, determining each unit module in the protein extraction device and the usage index information of each unit module according to the device manual; and performing feature extraction processing on various protein extraction parameters to obtain parameter feature information of various protein extraction parameters;

[0071] A hierarchical model is constructed, each unit module and various protein extraction parameters are defined as analysis targets, and the target layer of the hierarchical model is obtained;

[0072] The usage index information of each unit module and the parameter characteristic information of various protein extraction parameters are defined as evaluation indicators to obtain the criterion layer of the hierarchical model;

[0073] and refining the criterion layer to obtain a number of refined evaluation indicators, and generating an indicator layer of a hierarchical structure model according to the number of refined evaluation indicators;

[0074] Refine the criteria layer indicators, for example, refine the unit module usage indicators into "target protein type", "extraction efficiency requirements", "purity requirements", etc., and refine the protein extraction parameter characteristic information into "temperature", "pH value", "time", etc.; taking extraction efficiency as an example, an extraction rate greater than 90% is high extraction efficiency; an extraction rate between 70%-90% is medium extraction efficiency; an extraction rate less than 70% is low extraction efficiency. Temperatures below 20°C are low temperatures; temperatures between 20°C and 40°C are medium temperatures; temperatures above 40°C are high temperatures;

[0075] Obtaining the importance of each detailed evaluation indicator in the indicator layer, and assigning a weight value to each detailed evaluation indicator according to the importance of each detailed evaluation indicator;

[0076] Constructing a judgment matrix according to the importance of each refined evaluation index; calculating the consistency index of the judgment matrix, and judging whether the consistency index of the judgment matrix meets the preset index requirements;

[0077] If the consistency index of the judgment matrix does not meet the preset index requirements, the weight values ​​of each detailed evaluation index are adjusted until the consistency index of the judgment matrix meets the preset index requirements;

[0078] The weight values ​​on the judgment matrix are weighted and summed to obtain the comprehensive score of the corresponding unit module; and so on, until the comprehensive score of each unit module is calculated;

[0079] The unit modules are sorted according to their comprehensive scores, and the correlation ranking between the unit modules and various protein extraction parameters is obtained to generate a correlation ranking result;

[0080] A semantic network image between each unit module and various protein extraction parameters is generated according to the correlation ranking result.

[0081] Each unit module in the protein extraction equipment includes but is not limited to a pressure control unit module, a temperature control unit module, a pH value control unit module, an oxygen concentration control unit module, a raw material processing module, a cleaning and disinfection module, and a monitoring module.

[0082] The usage index information of each unit module refers to the specific indicators that describe the functions and performance of each unit module in the protein extraction equipment. These indicators can help understand the role, capacity and scope of application of each unit module in the entire protein extraction process. For example, for the raw material processing module, the usage index information includes processing capacity (the amount of raw materials processed per hour), processing effect (such as the degree of crushing or homogenization of the raw materials), etc.

[0083] By understanding the usage index information of each unit module, we can select the appropriate unit module according to specific needs and process requirements, and optimize and adjust it to improve the performance and efficiency of the entire protein extraction equipment. At the same time, these index information can also be used as a basis for equipment selection, design and improvement, and promote the development and application of protein extraction technology.

[0084] It should be noted that the unit modules and their use index information in the equipment are determined by obtaining the equipment manual, and the protein extraction parameters are feature extracted to obtain their parameter feature information. Then, a hierarchical model is constructed, and the unit modules and protein extraction parameters are used as analysis targets to form a target layer; the use index information and parameter feature information are used as evaluation indicators to form a criterion layer; the criterion layer is further refined to generate an indicator layer, and weights are assigned to each refined evaluation indicator to construct a judgment matrix. By calculating the consistency index of the judgment matrix, the weight value is adjusted until the preset requirements are met to ensure the rationality of the evaluation system. Finally, the comprehensive score of each unit module is calculated by weighted summation, and the scores are sorted to obtain the correlation ranking of each unit module and the protein extraction parameters. Based on this, a semantic network image is generated to intuitively display the relationship between the two. This method generates a semantic network image by systematically analyzing the relationship between the unit module and the protein extraction parameters, providing intuitive data support, so that the system can adjust the equipment parameters faster and more reasonably, and improve the system response efficiency.

[0085] Preferably, according to the semantic network image, combined with the protein extraction parameters in a non-steady state, a traceability analysis is performed on each unit module in the protein extraction device to obtain a unit module with a suspicious operating state in the protein extraction device, specifically:

[0086] Acquiring protein extraction parameters in a protein extraction device whose state is a non-steady state, and importing the protein extraction parameters in the non-steady state into the semantic network image for pairing analysis;

[0087] Matching in the semantic network image to obtain the correlation between each unit module and the protein extraction parameter whose state is a non-steady state, and obtaining a correlation matching result;

[0088] According to the correlation matching result, the unit module corresponding to the protein extraction parameter with a non-steady-state state having a correlation greater than a preset correlation threshold is obtained to obtain the unit module with a suspicious operation state in the protein extraction device.

[0089] It should be noted that the protein extraction parameters with a non-steady-state state are obtained and imported into the semantic network image for pairing analysis. Then, the correlation between each unit module and these parameters is matched in the semantic network image to obtain the correlation matching result. Finally, according to the correlation matching result, the unit module whose correlation with the non-steady-state parameter is greater than the preset threshold is obtained, and it is determined as a suspicious unit module in the operating state. This method can quickly and accurately find out the unit modules related to the non-steady-state parameters, and promptly discover possible problems in the equipment, so as to take corresponding measures to adjust and optimize, and improve the stability and reliability of the equipment. In addition, through the analysis of the semantic network image, the relationship between each unit module and the protein extraction parameters can be deeply understood, providing a reference for the design and improvement of the equipment.

[0090] Preferably, further analysis is performed on the suspicious operation status unit module in the protein extraction device to determine whether the suspicious operation status unit module has experienced abnormal operation, specifically:

[0091] Obtaining an operation record log of the protein extraction device, and obtaining historical abnormal operation status events that have occurred in the protein extraction device according to the operation record log;

[0092] Based on the hierarchical clustering algorithm, the historical abnormal operation status events that occurred in the protein extraction equipment are clustered to obtain the clustering results of various abnormal operation status events of the same type;

[0093] More specifically, first, each abnormal event is considered as an independent cluster, then the distance between all clusters is calculated, and the two clusters with the closest distance are merged into a new cluster. This process is repeated until all clusters are merged into one large cluster, or the preset number of clusters is reached. This process generates a tree structure called a dendrogram, where each branch represents a cluster and the distance between branches represents the similarity between clusters. Finally, by setting a threshold, the dendrogram is cut to obtain different types of abnormal event clustering results;

[0094] The clustering results of various operating state anomalies of the same type are input into the feature pyramid network for feature extraction processing to obtain feature data of various types of operating state anomalies;

[0095] The clustering results of various operating status anomalies of the same type are input into the feature pyramid network for feature extraction processing. The principle is to use the pyramid structure to extract features at different scales, and finally fuse these features together to obtain a more comprehensive and abstract feature representation. Specifically, the feature pyramid network will perform multi-layer convolution operations on the input clustering result data, and each layer of convolution will extract features of different scales. Then, through upsampling and fusion operations, features of different scales are integrated together to form a richer feature expression. Finally, through the fully connected layer and classifier, these feature data can be mapped to different anomaly types, thereby realizing the feature extraction and identification of various types of operating status anomalies;

[0096] Constructing a knowledge base, dividing the knowledge base into a plurality of storage spaces, assigning a corresponding operation status abnormality type label to each storage space, and storing feature data of a corresponding type of operation status abnormality in the storage space assigned with the corresponding label, thereby obtaining a feature data knowledge base;

[0097] Obtaining a unit module with a suspicious operating status in the protein extraction device, and obtaining real-time working data of the unit module with a suspicious operating status; importing the real-time working data of the unit module with a suspicious operating status into the feature data knowledge base; wherein the working data includes voltage, current, efficiency, signal transmission quality, etc.;

[0098] Calculate the similarity between the real-time working data and the characteristic data in each storage space; if the similarity between the real-time working data and the characteristic data in one or more storage spaces is greater than a preset similarity threshold, it means that the suspicious operation status unit module has an abnormal operation condition, and the suspicious operation status unit module is marked as an abnormal operation status unit module;

[0099] And the abnormal operation status type label assigned to the storage space with similarity greater than a preset similarity threshold is obtained to obtain the abnormal operation status type label of the abnormal operation status unit module.

[0100] It should be noted that by obtaining the operation log of the equipment, abnormal operation status events that occurred in the past are identified. Then, these abnormal events are classified using a hierarchical clustering algorithm, and abnormal events with similar characteristics are classified into one category to form clustering results of abnormal events in each category. Next, these clustering results are input into the feature pyramid network for feature extraction, with the purpose of extracting representative features from historical abnormal events to form a feature data set. The feature data is then stored in a knowledge base, which is divided into multiple storage spaces, each of which corresponds to a specific type of abnormal operation status and stores feature data matching the type of abnormality. When the system detects a suspicious operation status unit module, the real-time working data of the module is collected and compared with the feature data in the knowledge base. By calculating the similarity between the real-time working data and various feature data in the knowledge base, the system can determine whether the module has an operation status similar to that of the historical abnormal events. If the similarity between the real-time working data and the feature data in one or more storage spaces exceeds the preset threshold, it means that the module may be experiencing the same abnormal operation status as the historical abnormal event, and the system will mark the module as an abnormal operation status unit module. Finally, the system assigns the correct abnormal type label to the abnormal operation status unit module by identifying the abnormal operation status type label corresponding to the storage space that is most similar to the real-time working data. This learning and prediction method based on historical data can identify the upcoming abnormal operation status in advance, provide a scientific basis for equipment maintenance and fault prevention, and improve equipment reliability and production efficiency.

[0101] Preferably, if an abnormal operation occurs, the abnormal operation state unit module is adjusted and optimized, specifically:

[0102] Acquire structural drawing information of the protein extraction device, and construct a structural model diagram of the protein extraction device according to the structural drawing information;

[0103] Acquire the real-time working data of each unit module in the protein extraction device, and perform format conversion processing on the real-time working data of each unit module. After the format conversion is completed, import the real-time working data of each unit module into the structural model diagram to obtain an operation simulation system of the protein extraction device;

[0104] The big data network is searched according to the abnormal operation status type label of the abnormal operation status unit module to obtain several adjustment optimization schemes;

[0105] Based on each adjustment optimization scheme, simulated adjustment and control processing is performed on the operation simulation system of the protein extraction equipment respectively. After the adjustment and control is completed, simulated working data of the abnormal operation state unit module after simulated adjustment and control by each adjustment optimization scheme is obtained;

[0106] Acquire preset working data of the unit module with abnormal operation status, and calculate the data deviation value between the simulated working data of the unit module with abnormal operation status and the preset working data after the simulation and regulation of each adjustment optimization scheme;

[0107] The adjustment optimization scheme corresponding to the minimum data deviation value is obtained, and the adjustment optimization scheme corresponding to the minimum data deviation value is sent to the control terminal of the protein extraction equipment, so that the control terminal can perform substantial adjustment optimization processing on the abnormal operation status unit module based on the adjustment optimization scheme.

[0108] The protein extraction parameters include pressure, temperature, pH value, oxygen concentration, solvent concentration and stirring speed. The working data include current, voltage, power, pressure, flow rate, speed, conductivity and valve status.

[0109] It should be noted that, first, by obtaining the structural drawing information of the equipment, a detailed structural model diagram is constructed to provide a basic framework for the subsequent operation simulation. Secondly, the real-time working data of each unit module is collected and processed, and integrated into the structural model diagram to form an operation simulation system to simulate the operation status of the equipment. Then, for the identified abnormal operation status unit module, a number of matching adjustment optimization schemes are found through big data network retrieval. These schemes are intended to eliminate the abnormal state by adjusting the operating parameters or equipment layout of the equipment. Subsequently, each adjustment optimization scheme is simulated and tested using the operation simulation system, and the simulated working data is obtained, compared with the preset working data, and the data deviation value is calculated to evaluate the effect of each scheme. Finally, the scheme with the smallest data deviation value is selected, which is considered to be the optimal adjustment optimization scheme, and this scheme is sent to the control terminal of the equipment. The control terminal adjusts and optimizes the actual equipment according to this scheme, in order to improve the operation status of the abnormal unit module. This method combines virtual simulation with intelligent decision-making to achieve efficient and accurate diagnosis and optimization of abnormal operation status of protein extraction equipment, significantly improves the operation efficiency and product quality of the equipment, and reduces the complexity and uncertainty of manual intervention. It not only saves the cost and time of actual debugging, but also enhances the maintainability and reliability of the equipment, providing strong technical support for improving the efficiency and stability of the entire production process.

[0110] In addition, the method further comprises the following steps:

[0111] Obtaining component information of a target batch of macadamia nut protein liquid, retrieving several contaminating bacterial species in a big data network based on the component information; and obtaining optimal survival environment parameters for various contaminating bacterial species;

[0112] Obtaining full process flow data for macadamia protein extraction, and obtaining preset production environment parameters for each sub-production area in the macadamia protein extraction production line based on the full process flow data;

[0113] Calculating the Pearson correlation coefficient between the preset production environment parameters of each sub-production area and the optimal survival environment parameters of various contaminating bacteria; defining the sub-production area corresponding to the Pearson correlation coefficient greater than the preset coefficient threshold as a high-risk pollution source;

[0114] At a preset time point, a real-time bacterial density distribution map of the colonies in each high-risk pollution source is obtained based on an optical microscope, and the cosine similarity between the real-time bacterial density distribution map and the standard bacterial density distribution map is calculated based on a cosine similarity algorithm;

[0115] If the cosine similarity is not greater than the preset similarity, feature extraction processing is performed on the real-time bacteria density distribution graph and the standard bacteria density distribution graph based on the ORB algorithm to obtain a first linear bacteria density distribution curve graph and a second linear bacteria density distribution curve graph;

[0116] The first linear bacteria density distribution curve graph and the second linear bacteria density distribution curve graph are registered based on an iterative closest point matching algorithm;

[0117] After the registration is completed, the non-overlapping curve area between the first linear bacteria density distribution curve graph and the second linear bacteria density distribution curve graph is extracted and defined as the abnormal bacteria density area. The abnormal bacteria density area is sent to the preset control terminal, so that the preset control terminal controls the disinfection equipment to perform disinfection on the abnormal bacteria density area to improve the quality of macadamia nut protein extraction.

[0118] It should be noted that the composition of macadamia protein liquid in the current batch production is analyzed by big data network to identify the contaminating bacteria and their optimal survival environment parameters that may have an adverse effect on the extraction process. Secondly, based on the full process data of macadamia protein extraction, the preset production environment parameters of each sub-production area are obtained. Then, the high-risk pollution source is identified by calculating the Pearson correlation coefficient between the preset production environment parameters of each sub-production area and the optimal survival environment parameters of the contaminating bacteria. Next, the bacterial density distribution in the high-risk pollution source is monitored in real time using an optical microscope, and the similarity between the real-time data and the standard data is evaluated by the cosine similarity algorithm. When the similarity is lower than the preset threshold, it indicates that the bacterial density is abnormal. At this time, the ORB algorithm is used to extract the features of the real-time and standard data, generate a linear bacterial density distribution curve, and realize the registration of the two images by the iterative nearest point matching algorithm. After registration, by comparing the curves, the abnormal bacterial density area is identified, and its information is sent to the preset control terminal, and the terminal instructs the disinfection equipment to process these areas. Through real-time monitoring and precise analysis, microbial contamination that may occur during the extraction process can be effectively prevented and controlled, thereby improving the purity and quality of macadamia protein. The whole process from data collection, analysis to execution has been automated, greatly improving production efficiency. Through mathematical models such as the Pearson correlation coefficient and cosine similarity, high-risk pollution sources and areas with abnormal bacterial density can be accurately located, avoiding blind operations and saving resources. Based on the results of data analysis, the system can independently decide whether disinfection measures need to be taken, reducing the pressure of manual monitoring and improving the accuracy and timeliness of decision-making. Overall, this method realizes the intelligent and refined management of the macadamia nut protein extraction process, significantly improves product quality and production efficiency, and improves quality control in the macadamia nut protein extraction process.

[0119] In addition, the method further comprises the following steps:

[0120] Obtaining target order demand information of the production workshop, and determining the factory quality requirement information of the current batch of macadamia protein liquid according to the target order demand information; and obtaining the factory delivery time of the current batch of macadamia protein liquid, and determining the preset storage time of the current batch of macadamia protein liquid according to the factory delivery time;

[0121] Obtain various rated parameters of a storage device for storing a current batch of macadamia protein liquid, and retrieve various optimal storage parameters for storing the current batch of macadamia protein liquid in a big data network based on the various rated parameters of the storage device, the preset storage time, and the factory quality requirement information to obtain an optimal storage parameter set; the storage parameters include temperature, light intensity, humidity, and oxygen concentration;

[0122] Obtaining a control log of a storage device, retrieving various historical storage parameter sets through the control log, and calculating mutual information values ​​between the optimal storage parameters and various historical storage parameter sets;

[0123] If there is a situation where at least one mutual information value is greater than the preset mutual information value, the maximum mutual information value is extracted, and the control scheme of the storage equipment of the historical storage parameter set corresponding to the maximum mutual information value is obtained in the control log, and the obtained control scheme is used as the optimal control scheme, and the storage equipment is controlled based on the optimal control scheme to store and process the current batch of macadamia nut protein liquid;

[0124] If each mutual information value is not greater than the preset mutual information value, the rated parameter range of each control module in the storage equipment is obtained, and various real-time storage parameters of the storage equipment are obtained, and the difference between various real-time storage parameters and various optimal storage parameters is calculated to obtain the storage parameter difference of various storage parameters;

[0125] Based on the rated parameter range of each control module in the storage equipment and the storage parameter difference of various storage parameters, the real-time storage parameters of the storage equipment are regulated and processed until the storage parameter difference of various storage parameters is less than the preset difference, then an optimal control scheme is generated, and based on the optimal control scheme, the storage equipment is controlled to store and process the current batch of macadamia protein liquid.

[0126] It should be noted that, first, by obtaining the target order demand information and the factory delivery time, the factory quality requirements and preset storage time of the product are determined. Then, according to the rated parameters of the storage equipment, the preset storage time and the factory quality requirements, the optimal storage parameter set is retrieved and determined in the big data network, including key indicators such as temperature, light intensity, humidity and oxygen concentration. The system further analyzes the control log and compares the mutual information value between the historical storage parameter set and the optimal storage parameter to identify the most effective storage strategy. If at least one mutual information value is found to exceed the preset threshold, the system will select the historical storage parameter set corresponding to the maximum mutual information value as the optimal control scheme to guide the operation of the storage equipment. On the contrary, if all mutual information values ​​do not reach the preset threshold, the system will dynamically adjust the real-time storage parameters according to the rated parameter range of the control module and the deviation of the current storage parameters until all parameter differences are less than the preset threshold, and automatically generate the optimal control scheme. Through big data analysis and historical data comparison, the system can customize a suitable storage strategy to ensure the quality and stability of the product during storage. It also automatically identifies the parameter deviation during the storage process and makes real-time adjustments through intelligent algorithms, which improves the automation level and response speed of warehouse management. Through precise parameter control, unnecessary energy consumption is reduced and the efficiency of storage space utilization is improved. In summary, through intelligent storage management, the storage process of macadamia protein liquid is optimized and controlled, storage efficiency and product quality are improved, and costs and risks are reduced.

[0127] like Figure 3 As shown, the second aspect of the present invention discloses a macadamia nut protein extraction process optimization system, the macadamia nut protein extraction process optimization system comprises a memory 30 and a processor 60, the memory 30 stores a macadamia nut protein extraction process optimization method program, when the macadamia nut protein extraction process optimization method program is executed by the processor 60, the steps of any one of the macadamia nut protein extraction process optimization methods are implemented.

[0128] The third aspect of the present invention discloses a computer-readable storage medium, which includes a macadamia nut protein extraction process optimization method program. When the macadamia nut protein extraction process optimization method program is executed by a processor, the steps of any one of the macadamia nut protein extraction process optimization methods are implemented.

[0129] The above are only specific embodiments of the present invention, but the protection scope of the present invention is not limited thereto. Any technician familiar with the technical field can easily think of changes or substitutions within the technical scope disclosed by the present invention, which should be covered by the protection scope of the present invention.

Claims

1. A method for optimizing the extraction process of macadamia protein, characterized in that: The following steps are involved: Obtaining a preset control scheme for a protein extraction device, obtaining preset parameter values ​​of various protein extraction parameters of the protein extraction device at a number of preset working time nodes according to the preset control scheme, and obtaining a set of preset parameter values ​​of various protein extraction parameters within a preset time period; Based on the Internet of Things, actual parameter values ​​of various protein extraction parameters in the protein extraction equipment are collected at several preset working time nodes to obtain a number set of actual parameter values ​​of various protein extraction parameters within a preset time period; Compare and analyze the actual parameter value sets of various protein extraction parameters within a preset time period with the preset parameter value sets, and analyze the status of various protein extraction parameters in the protein extraction device; perform correlation analysis on each unit module and various protein extraction parameters based on the hierarchical analysis method, and generate a semantic network image between each unit module and various protein extraction parameters; Obtaining protein extraction parameters in a non-steady-state state in a protein extraction device; performing a traceability analysis on each unit module in the protein extraction device based on the semantic network image and in combination with the protein extraction parameters in a non-steady-state state, and obtaining a unit module with a suspicious operating state in the protein extraction device; Further analysis is performed on the suspicious operation status unit module in the protein extraction equipment to determine that the suspicious operation status unit module has experienced abnormal operation; if an abnormal operation condition occurs, the abnormal operation status unit module is adjusted and optimized.

2. The method for optimizing the protein extraction process of macadamia nuts according to claim 1, characterized in that: The actual parameter value sets of various protein extraction parameters within the preset time period are compared and analyzed with the preset parameter value sets, and the status of various protein extraction parameters in the protein extraction equipment is obtained by analysis, specifically: Calculating the Pearson correlation coefficient between the actual parameter value set and the preset parameter value set of various protein extraction parameters within a preset time period; Compare the Pearson correlation coefficients between the actual parameter value sets of various protein extraction parameters within a preset time period and the preset parameter value sets with the preset coefficient values; If the Pearson correlation coefficient between the actual parameter value set and the preset parameter value set of a certain protein extraction parameter within a preset time period is greater than a preset coefficient value, the state of the protein extraction parameter is defined as a steady state; If the Pearson correlation coefficient between the actual parameter value set and the preset parameter value set of a certain protein extraction parameter within a preset time period is not greater than the preset coefficient value, the state of the protein extraction parameter is defined as a non-steady-state state.

3. The method for optimizing the extraction process of macadamia protein according to claim 1, characterized in that: Based on the hierarchical analysis method, the correlation between each unit module and various protein extraction parameters was analyzed to generate a semantic network image between each unit module and various protein extraction parameters, specifically: Obtaining the device manual of the protein extraction device, determining each unit module in the protein extraction device and the usage index information of each unit module according to the device manual; and performing feature extraction processing on various protein extraction parameters to obtain parameter feature information of various protein extraction parameters; A hierarchical model is constructed, each unit module and various protein extraction parameters are defined as analysis targets, and the target layer of the hierarchical model is obtained; The usage index information of each unit module and the parameter characteristic information of various protein extraction parameters are defined as evaluation indicators to obtain the criterion layer of the hierarchical model; and refining the criterion layer to obtain a number of refined evaluation indicators, and generating an indicator layer of a hierarchical structure model according to the number of refined evaluation indicators; Obtaining the importance of each detailed evaluation indicator in the indicator layer, and assigning a weight value to each detailed evaluation indicator according to the importance of each detailed evaluation indicator; Construct a judgment matrix based on the importance of each detailed evaluation indicator; Calculating the consistency index of the judgment matrix, and judging whether the consistency index of the judgment matrix meets the preset index requirement; If the consistency index of the judgment matrix does not meet the preset index requirements, the weight values ​​of each detailed evaluation index are adjusted until the consistency index of the judgment matrix meets the preset index requirements; Performing weighted summation on the weight values ​​on the judgment matrix to obtain a comprehensive score of the corresponding unit module; And so on, until the comprehensive score of each unit module is calculated; The unit modules are sorted according to their comprehensive scores, and the correlation ranking between the unit modules and various protein extraction parameters is obtained to generate a correlation ranking result; A semantic network image between each unit module and various protein extraction parameters is generated according to the correlation ranking result.

4. The method for optimizing the protein extraction process of macadamia nuts according to claim 1, characterized in that: According to the semantic network image, combined with the protein extraction parameters in a non-steady state, the source analysis of each unit module in the protein extraction equipment is performed to obtain the suspicious unit modules in the protein extraction equipment, specifically: Acquiring protein extraction parameters in a protein extraction device whose state is a non-steady state, and importing the protein extraction parameters in the non-steady state into the semantic network image for pairing analysis; Matching in the semantic network image to obtain the correlation between each unit module and the protein extraction parameter whose state is a non-steady state, and obtaining a correlation matching result; According to the correlation matching result, the unit module corresponding to the protein extraction parameter with a non-steady-state state having a correlation greater than a preset correlation threshold is obtained to obtain the unit module with a suspicious operation state in the protein extraction device.

5. The method for optimizing the protein extraction process of macadamia nuts according to claim 1, characterized in that: Further analysis is performed on the suspicious operation status unit module in the protein extraction equipment to determine that the suspicious operation status unit module has experienced abnormal operation, specifically: Obtaining an operation record log of the protein extraction device, and obtaining historical abnormal operation status events that have occurred in the protein extraction device according to the operation record log; Based on the hierarchical clustering algorithm, the historical abnormal operation status events that occurred in the protein extraction equipment are clustered to obtain the clustering results of various abnormal operation status events of the same type; The clustering results of various operating state anomalies of the same type are input into the feature pyramid network for feature extraction processing to obtain feature data of various types of operating state anomalies; Constructing a knowledge base, dividing the knowledge base into a plurality of storage spaces, assigning a corresponding operation status abnormality type label to each storage space, and storing feature data of a corresponding type of operation status abnormality in the storage space assigned with the corresponding label, thereby obtaining a feature data knowledge base; Acquire a unit module with a suspicious operation status in the protein extraction device, and acquire real-time working data of the unit module with a suspicious operation status; import the real-time working data of the unit module with a suspicious operation status into the feature data knowledge base; Calculating the similarity between the real-time working data and the characteristic data in each storage space; If the similarity between the real-time working data and the characteristic data in one or more storage spaces is greater than a preset similarity threshold, it means that the suspicious operation status unit module has experienced an abnormal operation, and the suspicious operation status unit module is marked as an abnormal operation status unit module; And the abnormal operation status type label assigned to the storage space with similarity greater than a preset similarity threshold is obtained to obtain the abnormal operation status type label of the abnormal operation status unit module.

6. The method for optimizing the extraction process of macadamia protein according to claim 5, characterized in that: If an abnormal operation occurs, the abnormal unit module will be adjusted and optimized, specifically: Acquire structural drawing information of the protein extraction device, and construct a structural model diagram of the protein extraction device according to the structural drawing information; Acquire the real-time working data of each unit module in the protein extraction device, and perform format conversion processing on the real-time working data of each unit module. After the format conversion is completed, import the real-time working data of each unit module into the structural model diagram to obtain an operation simulation system of the protein extraction device; The big data network is searched according to the abnormal operation status type label of the abnormal operation status unit module to obtain several adjustment optimization schemes; Based on each adjustment optimization scheme, simulated adjustment and control processing is performed on the operation simulation system of the protein extraction equipment respectively. After the adjustment and control is completed, simulated working data of the abnormal operation state unit module after simulated adjustment and control by each adjustment optimization scheme is obtained; Acquire preset working data of the unit module with abnormal operation status, and calculate the data deviation value between the simulated working data of the unit module with abnormal operation status and the preset working data after the simulation and regulation of each adjustment optimization scheme; The adjustment optimization scheme corresponding to the minimum data deviation value is obtained, and the adjustment optimization scheme corresponding to the minimum data deviation value is sent to the control terminal of the protein extraction equipment, so that the control terminal can perform substantial adjustment optimization processing on the abnormal operation status unit module based on the adjustment optimization scheme.

7. The method for optimizing protein extraction from macadamia nuts according to claim 1, characterized in that: The protein extraction parameters include pressure, temperature, pH value, oxygen concentration, solvent concentration and stirring speed.

8. The method for optimizing protein extraction from macadamia nuts according to claim 6, characterized in that: The working data includes current, voltage, power, pressure, flow, speed, conductivity and valve status.

9. A macadamia protein extraction process optimization system, characterized in that: The macadamia protein extraction process optimization system includes a memory and a processor, wherein a macadamia protein extraction process optimization method program is stored in the memory. When the macadamia protein extraction process optimization method program is executed by the processor, the steps of the macadamia protein extraction process optimization method as described in any one of claims 1 to 8 are implemented.

10. A computer-readable storage medium, characterized in that: The computer-readable storage medium includes a macadamia nut protein extraction process optimization method program. When the macadamia nut protein extraction process optimization method program is executed by a processor, the steps of the macadamia nut protein extraction process optimization method as described in any one of claims 1 to 8 are implemented.