Wheat germ processing scheme recommendation method and system combined with deep learning

Abstract

The invention provides a wheat germ processing scheme recommendation method and system combined with deep learning. The wheat germ processing scheme recommendation method comprises the following steps: firstly, acquiring multi-dimensional basic data (including variety characteristics, raw material quality and microstructure image data) of wheat germs and processing scene demand parameters (including processing product application scenes, processing capacity adaptation and processing resource constraint data); performing feature gene extraction processing on the data to obtain a wheat germ feature gene set and a scene demand feature gene set; inputting the sequence into a pre-trained depth scheme evolution model, and generating a processing link feature gene sequence through a gene interaction layer; constructing a plurality of processing scheme prototypes based on the processing link characteristic gene sequence; and finally, performing double-circulation adaptive optimization on the processing scheme prototype and the feature gene set to obtain a target processing recommendation scheme, and transmitting the target processing recommendation scheme to a processing execution terminal, thereby realizing scientific and accurate wheat germ processing scheme recommendation.

Description

Technical Field

[0001] This invention relates to the field of deep learning technology, and more specifically, to a method and system for recommending wheat germ processing schemes that incorporates deep learning. Background Technology

[0002] In the field of wheat germ processing, developing scientific and reasonable processing plans is crucial for improving product quality, meeting the needs of different application scenarios, and optimizing resource allocation. Traditional wheat germ processing plan development mainly relies on manual experience and simple experimental methods. Workers typically design processing procedures and parameters based on their understanding of wheat germ characteristics and past processing experience, combined with basic conditions such as processing site and equipment. However, the above methods have significant limitations.

[0003] On the one hand, human experience is subjective and limited; different individuals may have varying perceptions and judgments about the characteristics of wheat germ, leading to inconsistent quality in processing solutions. Furthermore, with the increasing variety of wheat germ varieties and the diversification of processing applications, human experience struggles to fully consider all the complex factors and accurately match the needs of different wheat germ varieties with specific application scenarios. On the other hand, simple experimental methods often only test a limited number of variables, making it difficult to comprehensively analyze and optimize the multi-dimensional characteristics of wheat germ and the various constraints during processing. This not only increases the cost and time of experiments but may also result in a final processing solution that fails to achieve optimal results, fully utilizes the value of wheat germ, and fails to meet market demand for high-quality processed products. Summary of the Invention

[0004] In view of the aforementioned problems, and in conjunction with the first aspect of the present invention, embodiments of the present invention provide a method for recommending wheat germ processing schemes using deep learning, the method comprising: The process involves obtaining multi-dimensional basic data on wheat germ and processing scenario requirements. The multi-dimensional basic data on wheat germ includes varietal characteristic data, raw material quality data, and microstructure image data. The processing scenario requirements include application scenario data of processed products, processing capacity matching data, and processing resource constraint data. Feature gene extraction was performed on multi-dimensional basic data of wheat germ and processing scenario requirement parameters to obtain a set of wheat germ feature genes and a set of scenario requirement feature genes. The set of wheat germ characteristic genes and the set of scenario requirement characteristic genes are input into a pre-trained deep solution evolution model. The gene interaction layer of the deep solution evolution model dynamically recombines the set of wheat germ characteristic genes and the set of scenario requirement characteristic genes to generate the characteristic gene sequence of the processing link. Based on the characteristic gene sequences of the processing links, multiple processing link units are combined through the scheme prototype construction layer of the deep scheme evolution model to form multiple processing scheme prototypes. A dual-loop adaptation optimization is performed on multiple processing scheme prototypes and wheat germ characteristic gene sets and scenario requirement characteristic gene sets to obtain the target processing recommendation scheme, which is then transmitted to the processing execution terminal.

[0005] In another aspect, embodiments of the present invention also provide a wheat germ processing scheme recommendation system combined with deep learning, including a processor and a machine-readable storage medium. The machine-readable storage medium is connected to the processor. The machine-readable storage medium is used to store programs, instructions or code. The processor is used to run the programs, instructions or code in the machine-readable storage medium to implement the above-described method.

[0006] Based on the above, this invention acquires multi-dimensional basic data of wheat germ and processing scenario requirements parameters, extracts feature genes from the acquired data, transforms complex data into a representative set of feature genes, and utilizes a pre-trained deep scheme evolution model to dynamically recombine the wheat germ feature gene set and the scenario requirement feature gene set through a gene interaction layer. This simulates the process of biological evolution, generating feature gene sequences for processing steps that meet specific requirements, thus achieving intelligent design and optimization of processing steps. Multiple processing scheme prototypes are constructed based on the processing step feature gene sequences, increasing the diversity of schemes and providing more possibilities for selecting the optimal scheme. By performing a double-loop adaptation optimization on multiple processing scheme prototypes and feature gene sets, various factors can be comprehensively considered to ensure that the final target processing recommendation scheme meets the processing scenario requirements while maximizing the value of wheat germ, improving processing efficiency and product quality, and reducing processing costs. Attached Figure Description

[0007] Figure 1 This is a schematic diagram of the execution flow of the wheat germ processing scheme recommendation method combined with deep learning provided in the embodiments of the present invention.

[0008] Figure 2 This is a schematic diagram of the hardware architecture of the wheat germ processing scheme recommendation system that combines deep learning, provided in an embodiment of the present invention. Detailed Implementation

[0009] The present invention will now be described in detail with reference to the accompanying drawings. Figure 1 This is a flowchart illustrating a wheat germ processing scheme recommendation method incorporating deep learning, provided in one embodiment of the present invention. The following is a detailed description of this wheat germ processing scheme recommendation method incorporating deep learning.

[0010] Step S110: Obtain multi-dimensional basic data and processing scenario requirement parameters of wheat germ. The multi-dimensional basic data of wheat germ includes variety characteristic data, raw material quality data, and microstructure image data of wheat germ. The processing scenario requirement parameters include application scenario data of processed products, processing capacity matching data, and processing resource constraint data.

[0011] In this embodiment, the implementation process of the method is illustrated by taking a batch of wheat germ that a food processing plant plans to process for the production of high-end nutritional supplements as an example. When acquiring multi-dimensional basic data on wheat germ, a combination of raw material procurement records and laboratory testing is used. Variety characteristic data is extracted from the procurement records, including the variety name of the wheat germ, soil type of the planting area, climate conditions, and planting cycle. Raw material quality data is obtained through laboratory testing, covering the content of nutrients such as protein, dietary fiber, and vitamin E, as well as moisture content, types and proportions of impurities, etc. Microscopic structural image data is obtained by using a scanning electron microscope to photograph multiple randomly selected wheat germ samples, obtaining image information such as cell arrangement, cell wall thickness, and endosperm structure at different magnifications.

[0012] The processing scenario requirements were determined through communication with the factory's production management and marketing departments. The application scenario data for the processed product clearly defines it as the production of high-end nutritional supplements, targeting health-conscious mid-to-high-end consumers. The product must meet relevant food safety standards and nutritional requirements. Processing capacity matching data was determined based on market demand, including planned daily output, continuous operating time of the production line, and processing volume per batch. Processing resource constraints data involved the factory's existing processing equipment, such as the type and capacity of cleaning equipment, specifications of drying equipment, and precision of grinding equipment. Information regarding raw material supply included the total quantity of wheat germ in this batch, supply cycle, and energy consumption limitations and environmental protection requirements during the production process.

[0013] When acquiring this data, data anonymization techniques are used to process sensitive information, such as the trade secrets of raw material suppliers and the personal information of testing personnel. Specifically, supplier names and contact information are anonymized, and the identity information of testing personnel is encrypted and stored, with access granted only to relevant management personnel with specific permissions when necessary, ensuring data security and confidentiality.

[0014] Step S120: Extract feature genes from the multi-dimensional basic data and processing scenario requirement parameters of wheat germ to obtain a set of wheat germ feature genes and a set of scenario requirement feature genes. The set of wheat germ feature genes includes variety feature genes, quality feature genes, and structural feature genes. The set of scenario requirement feature genes includes application scenario feature genes, production capacity adaptation feature genes, and resource constraint feature genes.

[0015] In this embodiment, feature gene extraction processing is performed on the acquired multi-dimensional basic data of wheat germ and processing scenario requirement parameters. First, for the multi-dimensional basic data of wheat germ, key information is extracted from variety characteristic data, raw material quality data, and microstructure image data, which are then converted into corresponding feature gene fragments and integrated into a wheat germ feature gene set. For the processing scenario requirement parameters, the core content of application scenario data, processing capacity adaptation data, and processing resource constraint data is also parsed, converted into feature gene fragments, and then formed into a scenario requirement feature gene set.

[0016] Step S121: Extract variety characteristic data from the multi-dimensional basic data of wheat germ, perform structured analysis on the variety characteristic data, separate variety origin association information, variety growth cycle association information, and variety resistance association information, and convert the variety origin association information, variety growth cycle association information, and variety resistance association information into quantifiable variety characteristic gene fragments. Each variety characteristic gene fragment corresponds to a core representation of a variety attribute.

[0017] In this embodiment, after extracting the varietal characteristic data of wheat germ, structured analysis is performed. The varietal origin association information includes soil pH, altitude, and average annual precipitation of the planting area. This information is converted into quantifiable values, such as soil pH range, altitude recorded in meters, and average annual precipitation counted in millimeters, forming varietal characteristic gene fragments corresponding to the varietal origin association information. These varietal characteristic gene fragments contain values ​​corresponding to different origin attributes.

[0018] The varietal growth cycle association information includes the number of days from sowing to harvest, the duration of each growth stage, etc. This time information is converted into a numerical sequence in days, forming the varietal characteristic gene fragments corresponding to the varietal growth cycle association information. The values ​​in the varietal characteristic gene fragments correspond to the duration of specific stages in the growth cycle.

[0019] Variety resistance association information includes resistance to common diseases and pests, drought resistance, cold resistance, etc. By consulting the breeding data and planting records of the variety, this resistance information is quantified into different levels of numerical values. For example, resistance is divided into several levels from strong to weak, and each level corresponds to a specific value, forming the variety characteristic gene fragment corresponding to the variety resistance association information. This variety characteristic gene fragment contains the quantitative values ​​of multiple resistance indicators.

[0020] Step S122: Extract raw material quality data from the multi-dimensional basic data of wheat germ, analyze the component composition information, freshness correlation information, and impurity content correlation information in the raw material quality data, and decompose the component composition information, freshness correlation information, and impurity content correlation information into quality characteristic gene fragments. Each quality characteristic gene fragment contains a specific characterization parameter of a quality indicator.

[0021] In this embodiment, when parsing the raw material quality data, the composition information includes the content of nutrients such as protein, dietary fiber, and vitamin E. The content of each component is expressed as a percentage or per 100 grams, forming a quality characteristic gene fragment corresponding to the composition information. This quality characteristic gene fragment contains quantitative parameters of multiple nutrients.

[0022] Freshness-related information involves indicators such as acid value and peroxide value. The test results of these indicators are converted into corresponding values ​​to form quality characteristic gene fragments corresponding to freshness-related information. The values ​​in these quality characteristic gene fragments reflect the freshness of wheat germ.

[0023] The impurity content association information includes the types of impurities, such as soil, weed seeds, and other grain particles, as well as the proportion of each type of impurity. After quantifying this information, a quality characteristic gene fragment corresponding to the impurity content association information is formed. This quality characteristic gene fragment contains the identifier of the impurity type and the corresponding percentage value.

[0024] Step S123: Extract microstructure image data from the multidimensional basic data of wheat germ, and obtain cell structure distribution information, tissue structure density information, and surface morphology association information from the microstructure image through image feature analysis technology. Convert the cell structure distribution information, tissue structure density information, and surface morphology association information into structural feature gene fragments. Each structural feature gene fragment corresponds to a quantitative description of a microstructure attribute.

[0025] In this embodiment, image feature analysis technology is used to process the microstructure image data. First, the image is preprocessed, including noise reduction and contrast enhancement, to extract features more clearly.

[0026] Step S1231: Input the wheat germ microstructure image data into the image feature analysis module, divide the microstructure image into regions, and separate the cell region, interstitial region, and surface coverage region.

[0027] In this embodiment, the image feature analysis module uses edge detection and region growing algorithms to divide the microstructure image into regions. The edge detection algorithm identifies the boundaries of different regions in the image, such as the boundaries between cells and the boundaries between cells and tissue gaps; the region growing algorithm groups pixels with the same or similar features based on pixel similarity to form different regions. By combining these two algorithms, the image is separated into cellular regions, tissue gap regions, and surface coverage regions, each with a clear boundary and extent.

[0028] Step S1232: Extract pixel features from the cell region, calculate the gray-level distribution information, edge contour information, and texture direction information of the pixels within the cell region, and determine the arrangement of cell structures, cell size association information, and cell integrity association information based on the gray-level distribution information, edge contour information, and texture direction information of the pixels within the cell region, thus forming cell structure distribution information.

[0029] In this embodiment, when extracting pixel features from the cell region, calculating the grayscale distribution information involves statistically analyzing the number and proportion of pixels with different grayscale values ​​within the cell region, resulting in a numerical sequence corresponding to a grayscale distribution histogram. Edge contour information is obtained by calculating the gradient changes of pixels at the cell edges, reflecting the shape and complexity of the cell contour and forming a multi-dimensional numerical set. Texture direction information is obtained by analyzing the arrangement patterns and directions of pixels within the cell region, represented by multiple directional parameters.

[0030] Based on this information, the arrangement of cell structures is determined, such as whether they are regularly arranged and their direction, and represented by corresponding codes and parameters. Cell size-related information includes the average diameter, maximum diameter, and minimum diameter of the cells. Cell integrity-related information is quantified based on whether the cells are damaged or deformed, yielding corresponding numerical values. All of this information together constitutes the cell structure distribution information.

[0031] Step S1233: Perform density analysis on the interstitial region, calculate the number of interstitial spaces per unit area, the size distribution of the interstitial spaces, and the connectivity of the interstitial spaces, and generate tissue structure density information based on the number of interstitial spaces per unit area, the size distribution of the interstitial spaces, and the connectivity of the interstitial spaces. The tissue structure density information can characterize the compactness of wheat germ tissue.

[0032] In this embodiment, when performing density analysis on the interstitial space region, the number of interstitial spaces per unit area is calculated by counting the number of interstitial spaces within a set unit area range, resulting in a numerical value. The interstitial space size distribution is obtained by measuring the area or volume of each interstitial space, classifying and statistically analyzing them according to size ranges, forming a distribution sequence. Interstitial space connectivity is obtained by analyzing whether and to what extent the interstitial spaces are interconnected, and is represented by a multi-dimensional parameter.

[0033] The tissue density information generated based on this data includes the number of gaps per unit area, gap size distribution parameters, connectivity parameters, etc. These can reflect the compactness of wheat germ tissue. The fewer and smaller the gaps and the worse the connectivity, the more compact the tissue.

[0034] Step S1234: Extract morphological features from the surface coverage area to obtain information on the surface undulation, smoothness correlation, and defect distribution. The surface undulation, smoothness correlation, and defect distribution information together constitute the surface morphological correlation information.

[0035] In this embodiment, when extracting morphological features from the surface coverage area, the surface undulation information is obtained by calculating the height changes of surface pixels, represented by multiple statistical parameters of height differences, such as maximum height difference and average height difference. Smoothness correlation information is calculated based on the rate of change of surface pixel grayscale values; a smaller rate of change indicates a smoother surface, forming a corresponding quantization parameter.

[0036] Defect distribution information includes the location, quantity, and size of defects such as cracks and spots on the surface. This information is quantified to form a multi-dimensional numerical set, where each dimension corresponds to a parameter related to a type of defect. This information collectively constitutes surface morphology correlation information.

[0037] Step S1235: Encode the cell structure distribution information, tissue structure density information, and surface morphology association information into gene fragments respectively. The cell structure distribution information corresponds to a set of structural feature gene fragments, the tissue structure density information corresponds to a set of structural feature gene fragments, and the surface morphology association information corresponds to a set of structural feature gene fragments. During the encoding process, retain the quantitative parameters and association relationships of the cell structure distribution information, tissue structure density information, and surface morphology association information.

[0038] In this embodiment, when encoding cell structure distribution information into gene fragments, quantitative parameters such as cell structure arrangement, cell size association information, and cell integrity association information are converted into specific numerical sequences according to preset encoding rules to form a set of structural feature gene fragments. Each fragment in the structural feature genome fragment corresponds to a specific attribute in the cell structure distribution information, and the association between fragments is consistent with the attribute association in the original information.

[0039] When encoding tissue density information, the number of gaps per unit area, gap size distribution parameters, connectivity parameters, etc. are converted into numerical sequences according to the encoding rules to form corresponding structural feature gene fragments, ensuring that the interrelationships between these parameters are preserved after encoding.

[0040] When encoding surface morphology-related information, the parameters in the surface undulation information, smoothness-related information, and defect distribution information are converted into numerical sequences to form corresponding structural feature gene fragments, while maintaining the correlation between the parameters.

[0041] Step S124: Associate and integrate the varietal characteristic gene fragments, quality characteristic gene fragments, and structural characteristic gene fragments, mark the interaction relationships between the varietal characteristic gene fragments, quality characteristic gene fragments, and structural characteristic gene fragments, and form a set of wheat germ characteristic genes containing a list of gene fragments and their interaction relationships.

[0042] In this embodiment, varietal characteristic gene fragments, quality characteristic gene fragments, and structural characteristic gene fragments are linked and integrated. First, the interactions between these fragments are analyzed. For example, growth cycle information in varietal characteristic gene fragments may affect the nutrient content in quality characteristic gene fragments, and cell structure information in structural characteristic gene fragments may be related to the varietal origin in varietal characteristic gene fragments.

[0043] These interactions are labeled by constructing an association matrix, where the elements represent the strength of association between different gene segments. All gene segments are then compiled into a list, which, together with the association matrix, constitutes a set of characteristic genes for wheat germ. This set of characteristic genes for wheat germ comprehensively presents the various characteristics of wheat germ and their internal relationships.

[0044] Step S125: Analyze the application scenario data, processing capacity adaptation data, and processing resource constraint data in the processing scenario demand parameters. Convert the application scenario data into application scenario feature gene fragments, the processing capacity adaptation data into capacity adaptation feature gene fragments, and the processing resource constraint data into resource constraint feature gene fragments, respectively. Associate the application scenario feature gene fragments, capacity adaptation feature gene fragments, and resource constraint feature gene fragments to form a scenario demand feature gene set. The gene fragment structure of the scenario demand feature gene set is adapted to the gene fragment structure of the wheat germ feature gene set.

[0045] In this embodiment, the processing scenario requirement parameters are analyzed, each data is converted into a corresponding feature gene fragment, and then associated to form a scenario requirement feature gene set. It is ensured that the gene fragment structure is compatible with the wheat germ feature gene set for subsequent interactive processing.

[0046] Step S1251: Analyze the application scenario data in the processing scenario requirement parameters to determine the target use type of the processed product, the target user group association information, and the product quality level association requirements. Decompose the target use type of the processed product into use feature sub-items, the target user group association information into user feature sub-items, and the product quality level association requirements into quality feature sub-items. The use feature sub-items correspond to a set of application scenario feature gene fragments, the user feature sub-items correspond to a set of application scenario feature gene fragments, and the quality feature sub-items correspond to a set of application scenario feature gene fragments.

[0047] In this embodiment, when parsing the application scenario data, the target use type of the processed product is to produce high-end nutritional supplements. It is broken down into application feature sub-items, such as the form of the supplement (powder, capsules, etc.) and the method of consumption (direct consumption, addition to other foods), etc. Each sub-item is converted into a set of application scenario feature gene fragments, which contain the specific parameters of the sub-item.

[0048] The target user group is identified as health-conscious mid-to-high-end consumers, and is broken down into user characteristic sub-items, such as age range, spending power, and health needs. Each sub-item corresponds to a set of application scenario characteristic gene fragments, and related descriptive parameters are recorded.

[0049] Product quality grade requirements are to comply with relevant food safety standards and nutritional indicators, which are broken down into quality characteristic sub-items, such as contaminant limits and minimum nutrient content. Each sub-item is converted into a set of application scenario characteristic gene fragments, including corresponding standard values ​​and testing method identifiers.

[0050] Step S1252: Analyze the processing capacity adaptation data in the processing scenario demand parameters, extract the unit time output requirement, continuous production duration requirement, and batch production scale requirement of the processing process, convert the unit time output requirement of the processing process into output feature sub-items, the continuous production duration requirement into duration feature sub-items, and the batch production scale requirement into scale feature sub-items. The output feature sub-items correspond to a set of capacity adaptation feature gene fragments, the duration feature sub-items correspond to a set of capacity adaptation feature gene fragments, and the scale feature sub-items correspond to a set of capacity adaptation feature gene fragments.

[0051] In this embodiment, the processing capacity matching data is parsed, and the output requirement per unit time is converted into output feature sub-items, such as output per hour, output per minute, etc. Each sub-item corresponds to a set of capacity matching feature gene fragments, which include a specific output value range.

[0052] The continuous production duration requirement is converted into duration feature sub-items, such as the number of continuous production hours per day and the maximum duration of each continuous production run. Each sub-item corresponds to a set of capacity adaptation feature gene fragments, and duration-related parameters are recorded.

[0053] Batch production scale requirements are converted into scale characteristic sub-items, such as the amount of raw materials input per batch and the amount of product output per batch. Each sub-item corresponds to a set of capacity adaptation characteristic gene fragments, which include specific values ​​of scale.

[0054] Step S1253: Analyze the processing resource constraint data in the processing scenario requirement parameters to obtain the type restrictions of the equipment required for processing, the stability requirements of raw material supply, and the energy consumption control requirements. Decompose the type restrictions of the equipment required for processing into equipment feature sub-items, the stability requirements of raw material supply into raw material feature sub-items, and the energy consumption control requirements into energy consumption feature sub-items. Each equipment feature sub-item corresponds to a set of resource constraint feature gene fragments, each raw material feature sub-item corresponds to a set of resource constraint feature gene fragments, and each energy consumption feature sub-item corresponds to a set of resource constraint feature gene fragments.

[0055] In this embodiment, the processing resource constraint data is parsed, and the type restrictions of the equipment required for processing are broken down into equipment feature sub-items, such as the type of cleaning equipment, the heating method of drying equipment, and the precision level of grinding equipment. Each sub-item corresponds to a set of resource constraint feature gene fragments, which include the specific parameters and model requirements of the equipment.

[0056] The stability requirement for raw material supply is broken down into raw material characteristic sub-items, such as the supply cycle of raw materials and the fluctuation range of the quantity supplied each time. Each sub-item corresponds to a set of resource constraint characteristic gene fragments, recording the relevant time and quantity parameters.

[0057] Energy consumption control requirements are broken down into energy consumption characteristic sub-items, such as the upper limit of electricity consumption per unit product and the upper limit of water consumption per unit product. Each sub-item corresponds to a set of resource constraint characteristic gene fragments, which contain the energy consumption limit value.

[0058] Step S1254: Associativity labeling is performed on application scenario feature gene fragments, capacity matching feature gene fragments, and resource constraint feature gene fragments to determine the mutual influence relationship between application scenario feature gene fragments and capacity matching feature gene fragments, the mutual influence relationship between application scenario feature gene fragments and resource constraint feature gene fragments, and the mutual influence relationship between capacity matching feature gene fragments and resource constraint feature gene fragments. For example, the association relationship between quality feature sub-items in application scenario feature gene fragments and equipment feature sub-items in resource constraint feature gene fragments.

[0059] In this embodiment, three types of characteristic gene fragments are labeled with associations. The interaction between application scenario characteristic gene fragments and production capacity adaptation characteristic gene fragments is analyzed. For example, when the product quality level requirement is high in the application scenario, it may affect the output per unit time, thus establishing an association between the two.

[0060] Determine the relationship between application scenario feature gene fragments and resource constraint feature gene fragments. For example, the form requirements of products in an application scenario may impose specific restrictions on the type of processing equipment, thus forming an association.

[0061] Clarify the interaction between capacity-matching characteristic gene fragments and resource-constraint characteristic gene fragments, such as the requirement for continuous production duration may be limited by equipment energy consumption, and establish corresponding correlations.

[0062] Step S1255: Based on the correlation labeling results, the application scenario feature gene fragments, capacity matching feature gene fragments, and resource constraint feature gene fragments are sorted and integrated according to their correlation relationships to form a scenario demand feature gene set. Each gene fragment in the scenario demand feature gene set contains corresponding demand sub-item information and associated gene fragment identifiers.

[0063] In this embodiment, based on the correlation labeling results, the application scenario feature gene fragments, capacity matching feature gene fragments, and resource constraint feature gene fragments are sorted and integrated according to their degree of correlation. Fragments with greater mutual influence are placed in adjacent positions to form an ordered set of scenario demand feature genes.

[0064] Each gene fragment contains corresponding requirement sub-item information, such as the specific parameters of the quality feature sub-item in the application scenario feature gene fragment, and also carries the identifier of the associated gene fragments, such as which capacity matching or resource constraint feature gene fragments the fragment is associated with, which facilitates quick identification and retrieval in subsequent processing.

[0065] Step S130: Input the wheat germ characteristic gene set and the scenario requirement characteristic gene set into the pre-trained deep solution evolution model. Through the gene interaction layer of the deep solution evolution model, the two types of characteristic genes, the wheat germ characteristic gene set and the scenario requirement characteristic gene set, are dynamically recombined to generate the processing link characteristic gene sequence.

[0066] In this embodiment, the wheat germ feature gene set obtained from the previous processing and the scenario requirement feature gene set are input into a pre-trained deep scheme evolution model. This deep scheme evolution model has been trained on a large amount of wheat germ processing case data and is able to understand the intrinsic relationship and interaction mechanism between the two types of gene sets. As an important component of the model, the gene interaction layer is responsible for dynamically recombinating the two types of input feature genes. By analyzing the matching degree and correlation between gene fragments, it generates feature gene sequences for processing steps that conform to the processing logic.

[0067] Step S131: Input the wheat germ feature gene set and the scenario requirement feature gene set into the gene interaction layer of the deep solution evolution model. The gene interaction layer first performs similarity matching between gene fragments in the wheat germ feature gene set and gene fragments in the scenario requirement feature gene set to determine gene fragment pairs with corresponding associations.

[0068] In this embodiment, after receiving the wheat germ characteristic gene set and the scenario requirement characteristic gene set, the gene interaction layer initiates a similarity matching program. This similarity matching program determines the correspondence between gene segments by calculating the overlap and similarity of attribute parameters, functional descriptions, and other information of each gene segment in the two gene sets. For example, quality characteristic gene segments related to protein content in the wheat germ characteristic gene set can be matched with application scenario characteristic gene segments related to protein indicators in nutritional supplements in the scenario requirement characteristic gene set. When the similarity between the two reaches a set threshold, they are identified as a pair of gene segments with a corresponding association.

[0069] Step S132: Based on the association strength value of gene fragment pairs, the gene interaction layer calls the pre-stored gene recombination rules to perform information fusion on gene fragment pairs with corresponding associations, generating fused gene fragments. The fused gene fragments contain the core information of gene fragments corresponding to the wheat germ feature gene set and the core information of gene fragments corresponding to the scenario requirement feature gene set in the gene fragment pair.

[0070] In this embodiment, after determining the gene fragment pairs, the association strength value of each gene fragment pair is calculated. Based on the different association strength values, the gene interaction layer calls the corresponding rules from the pre-stored gene recombination rule base to perform information fusion. During the fusion process, core information from two types of gene fragments can be retained, such as the core parameters of the nutritional components of wheat germ and the core requirements of nutritional indicators in the scenario requirements. This core information is integrated into a new fused gene fragment, so that the fused gene fragment can both reflect the inherent characteristics of wheat germ and meet the scenario requirements.

[0071] Step S1321: Calculate the association strength value of each gene fragment pair using the gene interaction layer. The association strength value is determined based on the attribute similarity, functional matching degree, and demand adaptability of the gene fragments corresponding to the wheat germ feature gene set and the gene fragments corresponding to the scene demand feature gene set in the gene fragment pair.

[0072] In this embodiment, when calculating the association strength value, attribute similarity compares the degree of overlap in the attribute parameter types and ranges of the two gene fragment pairs, such as the overlap between the parameter range of vitamin E content in wheat germ and the parameter range of vitamin E indicators in the scenario requirements. Functional matching analyzes the degree of association between the two in the processing, such as the matching of the cell structure characteristics of wheat germ with the requirements of the grinding step in processing. Demand fit determines whether the characteristics of wheat germ can meet the scenario requirements, such as whether the impurity content of wheat germ meets the scenario's requirements for product purity. By comprehensively calculating the values ​​of these three dimensions, the association strength value of each gene fragment pair is obtained.

[0073] Step S1322: According to the preset range of association strength value, the gene interaction layer calls the corresponding gene recombination rule. Gene fragment pairs with association strength value that meet the first preset range call the deep fusion rule, gene fragment pairs with association strength value that meet the second preset range call the medium fusion rule, and gene fragment pairs with association strength value that meet the third preset range call the basic fusion rule.

[0074] In this embodiment, three preset association strength value ranges are defined. The first preset range represents the interval with high association strength values, suitable for deep fusion rules; the second preset range represents the interval with medium association strength values, suitable for medium fusion rules; and the third preset range represents the interval with low association strength values, suitable for basic fusion rules. The gene interaction layer automatically calls the corresponding recombination rule based on the interval to which the calculated association strength value belongs. For example, when the association strength value between the gene fragment representing protein content in wheat germ and the gene fragment representing protein indicators in the scenario requirements falls within the first preset range, a deep fusion rule is invoked.

[0075] Step S1323: For gene fragment pairs that invoke deep fusion rules, the gene interaction layer performs parameter-by-parameter fusion of all attribute parameters of the gene fragment corresponding to the wheat germ feature gene set in the gene fragment pair with all attribute parameters of the gene fragment corresponding to the scene requirement feature gene set, calculates the average value of the parameters after parameter-by-parameter fusion and retains the parameter fluctuation range before parameter-by-parameter fusion, and generates a deep fusion gene fragment containing complete parameter information.

[0076] In this embodiment, for gene fragment pairs that invoke deep fusion rules, such as dietary fiber fragments in wheat germ.

[0077] In this embodiment, for gene fragment pairs that invoke deep fusion rules, such as the gene fragment for dietary fiber content in wheat germ and the gene fragment for dietary fiber indicators in the scenario requirements, all attribute parameters of the two are fused parameter by parameter. For example, the numerical range of dietary fiber content in wheat germ is fused with the required numerical range of dietary fiber in the scenario requirements, the average value of each corresponding parameter is calculated, and the original parameter fluctuation range is retained. The final deep fused gene fragment contains these fused average values ​​and fluctuation range information, fully presenting the parameter characteristics of both.

[0078] Step S1324: For gene fragment pairs that invoke the moderate fusion rule, the gene interaction layer selects the core attribute parameters of the gene fragment corresponding to the wheat germ feature gene set in the gene fragment pair and the core attribute parameters of the gene fragment corresponding to the scene requirement feature gene set in the gene fragment pair for fusion. It retains the original values ​​of the non-core parameters of the gene fragment corresponding to the wheat germ feature gene set in the gene fragment pair and the original values ​​of the non-core parameters of the gene fragment corresponding to the scene requirement feature gene set in the gene fragment pair, and generates a moderately fused gene fragment containing the core fusion parameters and the original non-core parameters.

[0079] In this embodiment, when invoking the moderate fusion rule, taking the wheat germ moisture content gene fragment and the moisture requirement gene fragment in the drying process of the scenario as examples, the core attribute parameters of both, such as the actual moisture content value of the wheat germ and the moisture content value after drying required by the scenario, are selected for fusion, and the fused core parameters are calculated. For non-core parameters, such as the ambient temperature during wheat germ moisture detection and the model of the drying equipment required by the scenario, their original values ​​are retained and no fusion processing is performed, thereby generating a moderately fused gene fragment containing the core fusion parameters and the original non-core parameters.

[0080] Step S1325: For gene fragment pairs that invoke the basic fusion rules, the gene interaction layer only establishes the association identifier between the gene fragments corresponding to the wheat germ feature gene set and the gene fragments corresponding to the scene requirement feature gene set in the gene fragment pair, retains the complete parameter information of the gene fragments corresponding to the wheat germ feature gene set and the complete parameter information of the gene fragments corresponding to the scene requirement feature gene set in the gene fragment pair, and generates basic fusion gene fragments with association identifiers.

[0081] In this embodiment, when the association strength value of a gene fragment pair is within a third preset range, such as the gene fragment representing the soil pH of the wheat germ planting area and the gene fragment representing the environmental protection requirements of packaging materials in the scenario requirements, the gene interaction layer only establishes an association marker between the two, indicating that they have a certain indirect association. Simultaneously, all information, including the soil pH parameter range of the wheat germ characteristic gene fragment and the environmental protection index parameters of the packaging materials of the scenario requirements gene fragment, is fully preserved to generate a basic fused gene fragment with an association marker.

[0082] Step S133: For individual gene fragments that have not formed a corresponding association, the gene interaction layer supplements the association information through feature expansion technology, so that the individual gene fragments can form new associations with other fused gene fragments and generate supplementary gene fragments.

[0083] In this embodiment, there are some individual gene fragments that do not have corresponding associations with other gene fragments, such as gene fragments of certain trace minerals in wheat germ, or gene fragments with specific requirements for the cleanliness of the processing workshop in the scenario. The gene interaction layer uses feature expansion technology to supplement relevant association information for these individual gene fragments. For example, it supplements the trace mineral gene fragment with information on the product stability that it may affect during processing, and it supplements the cleanliness requirement gene fragment with information on the association with hygiene standards of other processing links. This enables these individual gene fragments to establish new associations with the already generated fusion gene fragments, thereby generating supplementary gene fragments.

[0084] Step S134: Input the fusion gene fragment and the supplementary gene fragment into the gene sorting unit. The gene sorting unit arranges the fusion gene fragment and the supplementary gene fragment in sequence based on the conventional execution order of wheat germ processing and the functional attributes of the fusion gene fragment and the supplementary gene fragment.

[0085] In this embodiment, after the fused gene fragment and the supplementary gene fragment are generated, they are input into the gene sorting unit. The gene sorting unit first refers to the conventional execution order of wheat germ processing, such as the order of cleaning, drying, and grinding, and at the same time combines the functional attributes of each gene fragment, such as which fragments are related to the cleaning step and which are related to the drying step, to perform a preliminary sequential arrangement of all gene fragments to ensure that the arrangement result conforms to the basic processing flow logic.

[0086] Step S1341: Establish a standard execution sequence library for wheat germ processing steps, which includes the standard execution sequence for common wheat germ processing steps such as cleaning, drying, grinding, extraction, and refining.

[0087] In this embodiment, the conventional execution sequence library is established based on the common wheat germ processing flow in the industry. The cleaning stage is at the very beginning, used to remove impurities; followed by the drying stage to reduce the moisture content of the wheat germ; then the grinding stage to break it into a certain particle size; next, the extraction stage to obtain the target nutrients; and finally, the refining stage to improve product purity. Each stage has a clearly defined sequence, serving as a basic reference for gene sequencing.

[0088] Step S1342: Use the gene sequencing unit to label the functional attributes of the fusion gene fragment and the supplementary gene fragment, and determine the wheat germ processing stage type corresponding to each fusion gene fragment and the wheat germ processing stage type corresponding to each supplementary gene fragment. For example, the fusion gene fragment related to the cleaning stage is labeled as the gene fragment related to the cleaning stage, and the supplementary gene fragment related to the drying stage is labeled as the gene fragment related to the drying stage.

[0089] In this embodiment, the gene sequencing unit performs functional attribute analysis on fusion gene fragments and supplementary gene fragments. For example, a fusion gene fragment containing parameters and requirements for impurity removal is labeled as related to the cleaning process; a supplementary gene fragment involving information on temperature control and moisture evaporation is labeled as related to the drying process. Through the above labeling, the processing stage type corresponding to each gene fragment is clearly defined.

[0090] Step S1343: Based on the conventional execution sequence library, the labeled fusion gene fragments and supplementary gene fragments are initially arranged according to the conventional sequence of the corresponding wheat germ processing steps to form a preliminary sorting sequence.

[0091] In this embodiment, according to the order of cleaning, drying, grinding, extraction, and purification in the conventional execution sequence library, the gene fragments marked as cleaning are placed first, followed by the gene fragments marked as drying, and then the gene fragments corresponding to grinding, extraction, and purification are arranged in sequence to form a preliminary sorting sequence, so that the order of gene fragments is consistent with the basic processing flow.

[0092] Step S1344: For the fusion gene fragments and supplementary gene fragments of the same wheat germ processing stage type in the preliminary sorting sequence, a secondary sorting is performed based on the functional subdivision attributes of the fusion gene fragments and supplementary gene fragments of the same wheat germ processing stage type. During the secondary sorting process, when there is a functional conflict between the fusion gene fragments and supplementary gene fragments of the same wheat germ processing stage type, the priority sorting rule is determined based on the application scenario feature genes in the scenario requirement feature gene set.

[0093] In this embodiment, for gene fragments in the same processing stage within the initial sorting sequence, such as multiple gene fragments related to the cleaning stage, a secondary sorting is performed based on their functional sub-attributes. For example, gene fragments related to impurity screening are sorted first, followed by gene fragments related to air separation for impurity removal. If gene fragments in the same stage have functional conflicts, such as one fragment requiring rapid cleaning to improve efficiency while another fragment requires fine cleaning to ensure purity, the application scenario characteristic genes in the scenario requirement characteristic gene set are referenced. If the application scenario emphasizes product purity, gene fragments related to fine cleaning are prioritized.

[0094] Step S135: During the sorting process, the gene sorting unit introduces a dynamic adjustment mechanism to adjust the sorting results according to the capacity adaptation feature genes and resource constraint feature genes in the scenario requirement feature gene set, and finally forms a processing link feature gene sequence containing ordered gene fragments.

[0095] In this embodiment, the gene sequencing unit introduces a dynamic adjustment mechanism based on the initial and secondary sequencing. According to the capacity-matching characteristic genes in the set of characteristic genes for scenario requirements, such as the required output per unit time, if a higher output is required, the order of gene fragments in certain stages may be adjusted to prioritize gene fragments for efficient processing. According to resource-constrained characteristic genes, such as equipment type restrictions, if a certain type of equipment can only be used in a specific stage, the position of the relevant gene fragments can be adjusted accordingly to ensure that the sequencing result conforms to the actual production conditions, ultimately forming the characteristic gene sequence for the processing stage.

[0096] Step S140: Based on the characteristic gene sequences of the processing links, multiple processing link units are combined through the scheme prototype construction layer of the deep scheme evolution model to form multiple processing scheme prototypes.

[0097] In this embodiment, after receiving the feature gene sequence of the processing link, the prototype construction layer of the deep scheme evolution model analyzes it, clarifies the types of processing link units that need to be combined, and calls the corresponding units from the processing link unit library for adjustment and combination, supplementing the connection process, thereby forming multiple processing scheme prototypes with differences.

[0098] Step S141: Analyze the characteristic gene sequence of the processing link using the scheme prototype construction layer of the deep scheme evolution model, extract the processing link type identifier in the characteristic gene sequence of the processing link, and determine the type of processing link unit to be combined, such as cleaning unit, drying unit, and grinding unit.

[0099] In this embodiment, the prototype construction layer analyzes the characteristic gene sequences of the processing steps and identifies the processing step type identifiers. For example, if the sequence contains multiple gene fragment identifiers related to cleaning, it indicates that a cleaning unit is needed; if it contains identifiers related to drying, it indicates that a drying unit is needed, and so on, to determine all the processing step unit types that need to be combined.

[0100] Step S142: Based on the processing stage type identifier, call the processing stage unit library and extract the standard processing stage unit of the corresponding type. Each standard processing stage unit contains the basic operation process and the range of common parameters.

[0101] In this embodiment, the processing unit library stores various standard processing units. Based on the determined processing unit type, such as a cleaning unit or a drying unit, the corresponding standard unit is extracted from the library. Each standard cleaning unit includes basic operational procedures such as screening, air separation, and magnetic separation, as well as conventional parameter ranges such as screening mesh size and air separation speed; the standard drying unit includes the basic hot air drying process and conventional parameter ranges such as temperature and time.

[0102] Step S143: Map the gene fragment information in the characteristic gene sequence of the processing step to the corresponding standard processing step unit, adjust the basic operation process and conventional parameter range of the corresponding standard processing step unit, so that the attributes of the adjusted processing step unit are consistent with the gene fragment information in the characteristic gene sequence of the processing step, and generate a customized processing step unit.

[0103] In this embodiment, gene fragment information in the characteristic gene sequences of the processing steps is mapped to corresponding standard processing step units. For example, information about impurity types and removal requirements in gene fragments related to the cleaning step is mapped to a standard cleaning unit, adding cleaning operation steps for specific impurities and adjusting the parameter range of the screening mesh; temperature requirements and moisture control information in gene fragments related to the drying step are mapped to a standard drying unit, modifying the parameter range of the drying temperature and the holding time in the operation process, generating customized processing step units that conform to the gene fragment information.

[0104] For example, in step S1431: the prototype construction layer of the deep scheme evolution model establishes a mapping table between gene fragment information in the feature gene sequence of the processing link and the attributes of the processing link unit. The mapping table is used to limit the processing link unit attribute category corresponding to each gene fragment category.

[0105] In this embodiment, the mapping table clarifies the correspondence between different gene fragment categories and processing unit attribute categories. For example, gene fragment categories related to impurities correspond to the impurity removal attribute category of the cleaning unit; gene fragment categories related to moisture correspond to the moisture control attribute category of the drying unit; gene fragment categories related to particle size correspond to the particle size adjustment attribute category of the grinding unit, etc. This table ensures that gene fragment information can be accurately mapped to the corresponding attributes of the processing units.

[0106] Step S1432: Based on the mapping relationship table, assign the information of each gene fragment in the characteristic gene sequence of the processing link to the attribute field of the corresponding processing link unit.

[0107] In this embodiment, based on the mapping table, the information of each gene fragment in the characteristic gene sequence of the processing step is assigned to the attribute field of the corresponding processing step unit. For example, the gene fragment information "impurities with a diameter greater than a certain value need to be removed" is assigned to the "impurity size screening" attribute field of the cleaning unit; and the gene fragment information "the moisture content after drying needs to be lower than a certain value" is assigned to the "final moisture content" attribute field of the drying unit.

[0108] Step S1433: Adjust the basic operation process of the processing unit, and add or delete operation steps in the basic operation process according to the functional requirements of the gene fragment information in the characteristic gene sequence of the processing unit.

[0109] In this embodiment, the basic operation flow of the processing unit is adjusted according to the functional requirements of the gene fragment information assigned to the attribute field. For example, if the gene fragment information of the cleaning unit requires the removal of magnetic impurities, but the basic operation flow of the standard cleaning unit does not include a magnetic separation step, then a magnetic separation step is added to the basic operation flow; if the gene fragment information of the drying unit indicates that secondary drying is not required, but the standard flow includes a secondary drying step, then that step is deleted.

[0110] Step S1434: Adjust the conventional parameter range of the processing unit, use the quantized parameter in the gene fragment information of the characteristic gene sequence of the processing unit as the benchmark value after the conventional parameter range is adjusted, and determine the upper and lower limits of the conventional parameter range after adjustment based on the fluctuation information in the gene fragment information of the characteristic gene sequence of the processing unit.

[0111] In this embodiment, when adjusting the conventional parameter range of the processing unit, the quantized parameter in the gene fragment information is used as the benchmark value. For example, if the quantized parameter in the gene fragment information of the grinding unit is a certain particle size value, this value is used as the benchmark value for the grinding particle size; the fluctuation information in the gene fragment information allows for a certain range of upward and downward fluctuation, and the upper and lower limits of the adjusted grinding particle size parameter range are determined based on this fluctuation information, so that the parameter range can cover a reasonable fluctuation range.

[0112] Step S1435: After the adjustment is completed, the correlation between the adjusted operation process and the adjusted parameter range of the processing unit is verified. After the verification is passed, a customized processing unit is generated.

[0113] In this embodiment, the correlation verification checks whether the adjusted operation flow and parameter range match each other and are consistent. For example, in the cleaning unit, if a magnetic separation step is added, it is necessary to verify whether there is a corresponding magnetic separation intensity parameter range; in the drying unit, if the drying temperature range is adjusted, it is necessary to verify whether the temperature range is compatible with the drying time parameter range to ensure that the expected drying effect can be achieved within the set time and temperature range. After the verification is passed, a customized processing unit is generated.

[0114] Step S144: According to the sorting order of the characteristic gene sequences of the processing links, the customized processing link units are connected in series and combined to supplement the connection process between adjacent customized processing link units, forming a preliminary scheme framework.

[0115] In this embodiment, the corresponding customized processing units are sequentially connected in series according to the order of cleaning, drying, grinding, extraction, and refining in the characteristic gene sequence of the processing steps. Connecting processes are added between adjacent units. For example, the connecting process between the cleaning unit and the drying unit includes the conveying method and speed control of transporting the cleaned wheat germ to the drying equipment; the connecting process between the drying unit and the grinding unit includes temperature detection and moisture content sampling of the dried wheat germ to ensure its condition meets the feeding requirements of the grinding unit, while also setting anti-caking measures during the conveying process.

[0116] The connection process between the grinding unit and the extraction unit includes particle size screening of the grinding product, filtering through a sieve with a specific aperture to remove large particles that do not meet the extraction requirements, and recording the amount of material after screening as the feed basis for the extraction unit.

[0117] The connection process between the extraction unit and the purification unit includes preliminary purity testing of the extract, qualitative analysis of the impurity content in the extract using rapid test strips, and adjustment of the pressure and flow rate of the delivery pipeline to ensure that the extract enters the purification unit stably.

[0118] By supplementing these connecting processes, the various customized processing units are made into a continuous and smooth whole, forming a preliminary solution framework. This preliminary solution framework clarifies the input-output relationship of each link and the operational requirements of the intermediate transitions.

[0119] Step S145: Adjust the sequence of customized processing units and the operation parameters of customized processing units in the preliminary scheme framework to a limited extent to generate multiple scheme frameworks with differences. After supplementing each scheme framework with complete operation instructions, a processing scheme prototype is formed.

[0120] In this embodiment, when adjusting the initial scheme framework, the order of some non-critical steps is fine-tuned first, while keeping the main processing steps unchanged. For example, some refining pretreatment steps after the extraction step are moved to the extraction step, forming an adjusted scheme framework; or a short cooling step is added after the drying step before entering the grinding step, forming another adjusted scheme framework.

[0121] For the operating parameters of customized processing units, limited fluctuations are made within the original parameter range. For example, the screening frequency of the cleaning unit is adjusted up or down by a set ratio, the temperature of the drying unit is adjusted slightly within the allowable range, and the rotation speed of the grinding unit is fine-tuned according to the material characteristics.

[0122] These adjustments resulted in multiple differentiated solution frameworks. For each framework, complete operational instructions were added, including specific operational steps for each stage, operator qualification requirements, equipment maintenance guidelines, and contingency plans for abnormal situations. For example, in a framework including a cooling stage, the startup procedure for the cooling equipment, methods for controlling cooling time, and standards for testing cooling effectiveness were detailed. After these additions, each framework formed an independent processing solution prototype.

[0123] Step S150: Perform double-loop adaptation optimization on multiple processing scheme prototypes and two types of feature gene sets: wheat germ feature gene set and scenario requirement feature gene set, to obtain the target processing recommendation scheme, and transmit the target processing recommendation scheme to the processing execution terminal.

[0124] In this embodiment, multiple processing scheme prototypes are subjected to dual-loop adaptation optimization. The first optimization loop focuses on the compatibility of the processing scheme prototype with the wheat germ characteristic gene set, while the second optimization loop focuses on the compatibility with the scenario requirement characteristic gene set. Through repeated loop adjustments, the adaptability of the scheme is improved, and finally the target processing recommendation scheme is determined and transmitted to the processing execution terminal.

[0125] Step S151: In the first round of optimization, each processing scheme prototype is matched with the wheat germ characteristic gene set. Processing steps in each processing scheme prototype that do not match the gene fragments in the wheat germ characteristic gene set are extracted. The operation parameters and operation process of the processing steps in each processing scheme prototype that do not match the gene fragments in the wheat germ characteristic gene set are adjusted to improve the matching degree between each processing scheme prototype and the wheat germ characteristic gene set.

[0126] In this embodiment, during the first round of optimization, each processing scheme prototype is matched with the wheat germ characteristic gene set. Each processing step in the prototype is compared with a gene fragment in the wheat germ characteristic gene set to identify mismatches. For example, the temperature parameters of the drying step in a certain prototype do not match the heat resistance information of the variety characteristic gene fragment in the wheat germ characteristic gene set; the variety has low heat resistance, while the current drying temperature is too high.

[0127] For these mismatched processing steps, the operating parameters and procedures were adjusted. For the drying step, the drying temperature was lowered while the drying time was extended to ensure effective drying without damaging the nutritional components of the wheat germ. Furthermore, the airflow velocity of the drying equipment was adjusted to match the microstructural characteristics of the wheat germ, avoiding material loss due to excessive airflow. These adjustments improved the compatibility between each processing prototype and the characteristic gene set of wheat germ.

[0128] Step S152: After completing the first round of optimization, enter the second round of optimization. Perform an adaptation analysis on the optimized processing scheme prototype obtained after the first round of optimization and the set of scene requirement feature genes. Identify the processing links in the optimized processing scheme prototype obtained after the first round of optimization that do not meet the requirements of the set of scene requirement feature genes. Adjust the order and operation rules of the processing links that do not meet the requirements of the set of scene requirement feature genes based on the gene fragments in the set of scene requirement feature genes.

[0129] In this embodiment, after completing the first round of optimization, the second round of optimization begins. The optimized processing scheme prototype is then matched with the set of scenario requirement feature genes to identify processing steps that do not meet the scenario requirements. For example, if the unit-time output of a certain optimized processing scheme prototype is lower than the requirement of the capacity matching feature gene fragment in the set of scenario requirement feature genes, it cannot meet the daily planned output.

[0130] Based on gene fragments in the gene set representing the specific needs of the application scenario, the sequence and operating rules of processing steps that did not meet the requirements were adjusted. For the aforementioned insufficient output, analysis revealed that the low efficiency of the grinding step was the cause. Therefore, the connection sequence between the grinding and drying steps was adjusted to reduce intermediate waiting time. Simultaneously, the operating rules of the grinding equipment were modified to improve the continuity of grinding. Furthermore, based on the purity requirements of the gene fragments representing the application scenario, the filtration operating rules in the extraction step were strengthened, and the number of filtrations was increased to ensure that the product quality meets the standards for high-end nutritional supplements.

[0131] Step S153: Between the first and second optimization cycles, the parameter information of the processing steps adjusted in the first optimization cycle is fed back to the parsing process of the scene requirement feature gene set to correct the correlation between gene fragments in the scene requirement feature gene set.

[0132] In this embodiment, between the first and second optimization cycles, parameter information of the processing steps adjusted in the first optimization cycle is collected, such as the adjusted drying temperature and grinding speed. This parameter information is then fed back into the analysis process of the scenario requirement feature gene set to re-examine the correlation between gene fragments in the scenario requirement feature gene set.

[0133] For example, lowering the drying temperature in the first optimization cycle might affect the timing of subsequent steps, thus impacting production capacity. Therefore, based on the adjusted drying time, the correlation between the production capacity adaptation feature gene fragment and the equipment running time in the resource constraint feature gene fragment of the scenario demand feature gene set is corrected to ensure that the scenario demand feature gene set can more accurately reflect the constraints and demands in the actual processing process.

[0134] Step S154: Repeat the first and second rounds of optimization loops. After each round of optimization loops, calculate the comprehensive fit value between each processing scheme prototype and the two sets of characteristic gene sets: the wheat germ characteristic gene set and the scenario requirement characteristic gene set. The comprehensive fit value is determined as follows: First, normalize the fit degree between each processing scheme prototype and the wheat germ characteristic gene set, and the fit degree between each processing scheme prototype and the scenario requirement characteristic gene set to the same evaluation scale. Then, perform a weighted calculation based on the normalized results.

[0135] In this embodiment, the first and second rounds of optimization loops are repeated. After each loop, the comprehensive fit value of each processing scheme prototype is calculated. During the calculation, the fit degree with the wheat germ characteristic gene set and the fit degree with the scenario requirement characteristic gene set are first normalized to make them fall on the same evaluation scale, for example, by converting them both to values ​​between 0 and 1.

[0136] Then, weights are assigned based on the importance of the two factors. For example, in the context of producing high-end nutritional supplements, the weight of the fit of the scenario requirements can be slightly higher than the weight of the fit of wheat germ features. The two normalized fits are multiplied by their corresponding weights, and the results are then concatenated to form a comprehensive fit value. This comprehensive fit value contains information in two dimensions, reflecting the fit with the two sets of feature genes respectively.

[0137] Step S155: When the overall adaptation value reaches the preset adaptation standard and the change in the overall adaptation value after two consecutive executions of the first and second optimization cycles is less than the preset threshold, stop executing the optimization cycle and select the processing scheme prototype with the highest overall adaptation value as the target processing recommendation scheme.

[0138] In this embodiment, the preset adaptation standard is that the values ​​of both dimensions in the comprehensive adaptation value are not lower than 0.85, and the preset threshold is 0.03. During the repeated loop, the change in the comprehensive adaptation value is continuously monitored. When the comprehensive adaptation value of a certain processing scheme prototype reaches the preset adaptation standard, and the change in the comprehensive adaptation value after two consecutive loops is less than 0.03, the optimization loop stops.

[0139] The processing scheme with the highest overall fit value was selected from all prototype processing schemes as the target processing recommendation. This target processing recommendation not only meets the characteristics of wheat germ itself, but also effectively matches the needs of the processing scenario, such as achieving target yield, meeting product quality standards, and making reasonable use of equipment resources.

[0140] Step S156: Transmit the target processing recommendation scheme to the processing execution terminal.

[0141] In this embodiment, the determined target processing recommendation scheme is converted to a format that meets the data reception requirements of the processing execution terminal. The converted scheme includes detailed parameters, operation procedures, equipment numbers, time nodes, and other information for each processing step. The scheme data is sent to the processing execution terminal via an encrypted transmission protocol, and a data verification mechanism is used during transmission to ensure data integrity and accuracy. After receiving the data, the processing execution terminal decrypts and parses the scheme, providing operators with visual operation guidance to guide the actual wheat germ processing process.

[0142] The method may also include the step of pre-training a deep scheme evolution model.

[0143] In this embodiment, the deep processing evolution model needs to be pre-trained before application. The pre-training process uses historical processing data, including multi-dimensional basic data of wheat germ processed in the past, corresponding processing scenario requirement parameters, and data on the final processing scheme and its implementation effect.

[0144] Step S211: Collect historical processing data, clean and preprocess the historical processing data, remove abnormal and duplicate data, and supplement missing data.

[0145] In this embodiment, historical data on wheat germ processing over the past three years is collected and stored in the company's database. The data is cleaned, and statistical analysis is used to identify abnormal data, such as processing parameters that significantly exceed reasonable ranges and production records that do not match actual conditions. These abnormal data are then removed.

[0146] For duplicate data, retain only one complete record. For missing data, make reasonable inferences based on data from the same batch or under similar conditions to supplement it. For example, if some quality test data for a batch of wheat germ is missing, it can be supplemented by referring to data from other batches of the same variety and under the same planting conditions. The preprocessed historical data forms a structured dataset, which is convenient for model training.

[0147] Step S212: Divide the preprocessed historical data into a training set, a validation set, and a test set. The training set is used to learn the model parameters, the validation set is used to adjust the model's hyperparameters, and the test set is used to evaluate the model's generalization ability.

[0148] In this embodiment, the preprocessed historical data is divided into a training set, a validation set, and a test set in a 7:2:1 ratio. The training set contains a large amount of historical data from different varieties and processing scenarios, which is used by the model to learn the correlation between features and the generation rules of processing schemes.

[0149] The validation set is used to evaluate the model's performance during training. Based on the validation results, the model's hyperparameters, such as the learning rate, number of network layers, and number of hidden units, are adjusted to optimize the model's fit. The test set is used after the model has been trained to evaluate its performance on unseen data, verifying the model's generalization ability and ensuring that the model can be applied to new processing scenarios.

[0150] Step S213: Construct the network structure of the deep scheme evolution model, including a gene interaction layer and a scheme prototype construction layer. The gene interaction layer adopts a bidirectional long short-term memory network structure, and the scheme prototype construction layer adopts a convolutional neural network structure.

[0151] In this embodiment, a network structure for a deep scheme evolution model is constructed. The gene interaction layer adopts a bidirectional long short-term memory network structure, which can process sequence data and capture the long-term dependencies and bidirectional association information of gene fragments in the wheat germ characteristic gene set and the scenario requirement characteristic gene set.

[0152] The bidirectional Long Short-Term Memory (LSTM) network comprises hidden layers in both forward and backward directions. Each hidden layer consists of multiple memory units. A gating mechanism controls the input, forgetting, and output of information, effectively addressing the problem of long sequence dependencies. The prototype construction layer employs a convolutional neural network (CNN) structure. This CNN structure extracts local features from the gene sequences of processing stages through convolutional kernels and performs feature dimensionality reduction and abstraction through pooling operations, enabling it to efficiently learn the combination patterns of processing stage units from the sequence.

[0153] Step S214: Set the hyperparameters for model training, including learning rate, training batch size, number of training iterations, regularization coefficient, etc.

[0154] In this embodiment, hyperparameters for model training are set. The learning rate is initially set to a small value to allow the model to learn stably. As training progresses, a learning rate decay strategy is adopted to gradually reduce the learning rate and improve the model's convergence accuracy.

[0155] The training batch size is determined based on the computer hardware performance and dataset size to ensure efficient use of computing resources in each training iteration. The number of training iterations is set to a relatively large value, and an early stopping mechanism is implemented: training stops when the loss function value on the validation set no longer decreases for several consecutive iterations, preventing overfitting. Regularization coefficients are used to control model complexity and prevent overfitting; this is achieved by adding a regularization term to the loss function.

[0156] Step S215: Input the training set into the deep scheme evolution model, use the backpropagation algorithm to train the model, and adjust the hyperparameters through the validation set until the loss function value of the model on the validation set reaches the preset threshold.

[0157] In this embodiment, the training set is input into the deep scheme evolution model for training. In each training batch, the model processes the input gene set to generate prediction results of processing scheme prototypes. The prediction results are compared with the actual processing schemes in historical data, and a loss function value is calculated. This loss function reflects the difference between the predicted scheme and the actual scheme.

[0158] A backpropagation algorithm is employed, adjusting the weight parameters of each layer of the model based on the loss function value. The error is propagated back from the output layer to the input layer, continuously optimizing the parameters. During training, the model performance is periodically evaluated using a validation set, and hyperparameters are adjusted based on the validation results. For example, if the validation loss increases, the learning rate is appropriately decreased or the regularization coefficient is increased. Training continues until the model's loss function value on the validation set reaches a preset threshold, at which point the model's prediction performance is considered satisfactory.

[0159] Step S216: Use the test set to evaluate the trained deep solution evolution model, and calculate the model's accuracy, recall, F1 score and other evaluation metrics. If the evaluation metrics meet the preset standards, the model training is complete; otherwise, adjust the model structure and hyperparameters and retrain.

[0160] In this embodiment, a test set is used to evaluate the trained model. The wheat germ feature gene set and the scenario requirement feature gene set from the test set are input into the model to obtain prototype processing schemes generated by the model. These prototype schemes are then compared with the actual processing schemes in the test set, and evaluation metrics such as accuracy, recall, and F1 score are calculated.

[0161] Accuracy reflects the proportion of correct solutions generated by the model among all generated solutions, recall reflects the proportion of actually correct solutions generated by the model, and the F1 score is a combined indicator of accuracy and recall. If these evaluation indicators meet the preset standards, such as accuracy and recall both being no less than 0.8, and the F1 score being no less than 0.8, then the model training is complete. If the standards are not met, the reasons are analyzed, the model structure is adjusted, such as increasing the number of network layers or changing the convolutional kernel size, and hyperparameters are adjusted, and retraining is performed until the model performance meets the standards.

[0162] Figure 2 The following is a schematic diagram of the hardware structure of a wheat germ processing scheme recommendation system 100 based on deep learning, provided in an embodiment of the present invention, for implementing the above-described wheat germ processing scheme recommendation method based on deep learning. Figure 2 As shown, the wheat germ processing scheme recommendation system 100, which incorporates deep learning, may include a processor 110, a machine-readable storage medium 120, a bus 130, and a communication unit 140.

[0163] In the specific implementation process, one or more processors 110 execute computer-executable instructions stored in machine-readable storage medium 120, so that processor 110 can execute the wheat germ processing scheme recommendation method combined with deep learning as described in the above method embodiment. Processor 110, machine-readable storage medium 120 and communication unit 140 are connected through bus 130. Processor 110 can be used to control the sending and receiving actions of communication unit 140.

[0164] The specific implementation process of processor 110 can be found in the various method embodiments executed by the wheat germ processing scheme recommendation system 100 combined with deep learning, as described above. The implementation principle and technical effect are similar, and will not be repeated here.

[0165] Furthermore, this embodiment of the invention also provides a readable storage medium containing computer-executable instructions. When the processor executes the computer-executable instructions, the above-mentioned method for recommending wheat germ processing schemes in conjunction with deep learning is implemented.

[0166] It should be noted that, in order to simplify the description of this invention and thus aid in the understanding of one or more embodiments, the foregoing description of the embodiments of this invention sometimes combines multiple features into a single embodiment, drawing, or description thereof. Similarly, it should be noted that, in order to simplify the description of this invention and thus aid in the understanding of one or more embodiments, the foregoing description of the embodiments of this invention sometimes combines multiple features into a single embodiment, drawing, or description thereof.

Claims

1. A method for recommending wheat germ processing schemes using deep learning, characterized in that, The method includes: The process involves obtaining multi-dimensional basic data on wheat germ and processing scenario requirements. The multi-dimensional basic data on wheat germ includes varietal characteristic data, raw material quality data, and microstructure image data. The processing scenario requirements include application scenario data of processed products, processing capacity matching data, and processing resource constraint data. Feature gene extraction was performed on multi-dimensional basic data of wheat germ and processing scenario requirement parameters to obtain a set of wheat germ feature genes and a set of scenario requirement feature genes. The set of wheat germ characteristic genes and the set of scenario requirement characteristic genes are input into a pre-trained deep solution evolution model. The gene interaction layer of the deep solution evolution model dynamically recombines the set of wheat germ characteristic genes and the set of scenario requirement characteristic genes to generate the characteristic gene sequence of the processing link. Based on the characteristic gene sequences of the processing links, multiple processing link units are combined through the scheme prototype construction layer of the deep scheme evolution model to form multiple processing scheme prototypes. A dual-loop adaptation optimization is performed on multiple processing scheme prototypes and wheat germ characteristic gene sets and scenario requirement characteristic gene sets to obtain the target processing recommendation scheme, which is then transmitted to the processing execution terminal.

2. The method for recommending wheat germ processing schemes using deep learning as described in claim 1, characterized in that, The process involves extracting feature genes from multi-dimensional basic data of wheat germ and processing scenario requirement parameters to obtain a set of wheat germ feature genes and a set of scenario requirement feature genes, including: Variety characteristic data were extracted from multi-dimensional basic data of wheat germ. The variety characteristic data were structured and analyzed to separate variety origin association information, variety growth cycle association information, and variety resistance association information. The variety origin association information, variety growth cycle association information, and variety resistance association information were converted into quantifiable variety characteristic gene fragments. Each variety characteristic gene fragment corresponds to the core representation of a variety attribute. Raw material quality data is extracted from multi-dimensional basic data of wheat germ. The composition information, freshness correlation information, and impurity content correlation information in the raw material quality data are analyzed. The composition information, freshness correlation information, and impurity content correlation information are decomposed into quality characteristic gene fragments. Each quality characteristic gene fragment contains a specific characterization parameter of a quality indicator. Microstructure image data is extracted from multidimensional basic data of wheat germ. Cell structure distribution information, tissue structure density information, and surface morphology correlation information in the microstructure image are obtained through image feature analysis technology. Cell structure distribution information, tissue structure density information, and surface morphology correlation information are converted into structural feature gene fragments. Each structural feature gene fragment corresponds to a quantitative description of a microstructure attribute. The varietal characteristic gene fragments, quality characteristic gene fragments, and structural characteristic gene fragments are associated and integrated, and the interaction relationships between the varietal characteristic gene fragments, quality characteristic gene fragments, and structural characteristic gene fragments are marked to form a set of wheat germ characteristic genes containing a list of gene fragments and their interaction relationships. The application scenario data, processing capacity adaptation data, and processing resource constraint data in the processing scenario demand parameters are analyzed. The application scenario data is converted into application scenario feature gene fragments, the processing capacity adaptation data is converted into capacity adaptation feature gene fragments, and the processing resource constraint data is converted into resource constraint feature gene fragments. The application scenario feature gene fragments, capacity adaptation feature gene fragments, and resource constraint feature gene fragments are associated to form a scenario demand feature gene set. The gene fragment structure of the scenario demand feature gene set is adapted to the gene fragment structure of the wheat germ feature gene set.

3. The method for recommending wheat germ processing schemes using deep learning according to claim 2, characterized in that, The process involves extracting microstructural image data from multidimensional basic data of wheat germ, and using image feature analysis technology to obtain cell structure distribution information, tissue structure density information, and surface morphology correlation information from the microstructural images. This cell structure distribution information, tissue structure density information, and surface morphology correlation information are then converted into structural feature gene fragments, including: The image data of wheat germ microstructure is input into the image feature analysis module to divide the microstructure image into regions, separating the cellular region, interstitial region, and surface coverage region. Pixel features are extracted from the cell region, and the gray-level distribution information, edge contour information, and texture direction information of the pixels within the cell region are calculated. Based on the gray-level distribution information, edge contour information, and texture direction information of the pixels within the cell region, the arrangement of cell structures, cell size correlation information, and cell integrity correlation information are determined to form cell structure distribution information. Density analysis is performed on the interstitial region to calculate the number of interstitial spaces, the size distribution of interstitial spaces, and the connectivity of interstitial spaces per unit area. Based on the number of interstitial spaces, the size distribution of interstitial spaces, and the connectivity of interstitial spaces per unit area, tissue structure density information is generated. The tissue structure density information can characterize the compactness of wheat germ tissue. Morphological features are extracted from the surface coverage area to obtain information on surface undulation, smoothness correlation, and defect distribution. These three information together constitute surface morphological correlation information. Cell structure distribution information, tissue structure density information, and surface morphology association information are encoded into gene fragments respectively. Cell structure distribution information corresponds to a set of structural feature gene fragments, tissue structure density information corresponds to a set of structural feature gene fragments, and surface morphology association information corresponds to a set of structural feature gene fragments. During the encoding process, the quantitative parameters and association relationships of cell structure distribution information, tissue structure density information, and surface morphology association information are preserved.

4. The method for recommending wheat germ processing schemes using deep learning according to claim 2, characterized in that, The analysis of application scenario data, processing capacity adaptation data, and processing resource constraint data in the processing scenario demand parameters involves converting the application scenario data into application scenario feature gene fragments, the processing capacity adaptation data into capacity adaptation feature gene fragments, and the processing resource constraint data into resource constraint feature gene fragments. These application scenario feature gene fragments, capacity adaptation feature gene fragments, and resource constraint feature gene fragments are then associated to form a scenario demand feature gene set, including: The application scenario data in the processing scenario requirement parameters are analyzed to determine the target use type of the processed product, the target user group association information, and the product quality level association requirements. The target use type of the processed product is decomposed into use feature sub-items, the target user group association information is decomposed into user feature sub-items, and the product quality level association requirements are decomposed into quality feature sub-items. The use feature sub-items correspond to a set of application scenario feature gene fragments, the user feature sub-items correspond to a set of application scenario feature gene fragments, and the quality feature sub-items correspond to a set of application scenario feature gene fragments. The processing capacity adaptation data in the processing scenario demand parameters is analyzed, and the unit time output requirement, continuous production duration requirement, and batch production scale requirement of the processing process are extracted. The unit time output requirement of the processing process is converted into output feature sub-items, the continuous production duration requirement is converted into duration feature sub-items, and the batch production scale requirement is converted into scale feature sub-items. The output feature sub-items correspond to a set of capacity adaptation feature gene fragments, the duration feature sub-items correspond to a set of capacity adaptation feature gene fragments, and the scale feature sub-items correspond to a set of capacity adaptation feature gene fragments. The processing resource constraint data in the processing scenario requirement parameters are analyzed to obtain the type restrictions of the equipment required for processing, the stability requirements of raw material supply, and the energy consumption control requirements. The type restrictions of the equipment required for processing are broken down into equipment feature sub-items, the stability requirements of raw material supply are broken down into raw material feature sub-items, and the energy consumption control requirements are broken down into energy consumption feature sub-items. Each equipment feature sub-item corresponds to a set of resource constraint feature gene fragments, each raw material feature sub-item corresponds to a set of resource constraint feature gene fragments, and each energy consumption feature sub-item corresponds to a set of resource constraint feature gene fragments. Associative markers are applied to application scenario characteristic gene fragments, capacity matching characteristic gene fragments, and resource constraint characteristic gene fragments to determine the mutual influence relationships between application scenario characteristic gene fragments and capacity matching characteristic gene fragments, between application scenario characteristic gene fragments and resource constraint characteristic gene fragments, and between capacity matching characteristic gene fragments and resource constraint characteristic gene fragments. For example, the association relationship between quality characteristic sub-items in application scenario characteristic gene fragments and equipment characteristic sub-items in resource constraint characteristic gene fragments. Based on the association labeling results, the application scenario feature gene fragments, capacity matching feature gene fragments, and resource constraint feature gene fragments are sorted and integrated according to their association relationships to form a scenario demand feature gene set. Each gene fragment in the scenario demand feature gene set contains corresponding demand sub-item information and associated gene fragment identifiers.

5. The method for recommending wheat germ processing schemes using deep learning according to claim 1, characterized in that, The process involves inputting the wheat germ characteristic gene set and the scenario requirement characteristic gene set into a pre-trained deep solution evolution model. Through the gene interaction layer of the deep solution evolution model, the wheat germ characteristic gene set and the scenario requirement characteristic gene set are dynamically recombinated to generate processing stage characteristic gene sequences, including: The set of wheat germ characteristic genes and the set of scenario requirement characteristic genes are input into the gene interaction layer of the deep solution evolution model. The gene interaction layer first performs similarity matching on gene fragments in the wheat germ characteristic gene set and gene fragments in the scenario requirement characteristic gene set to determine gene fragment pairs with corresponding associations. Based on the association strength value of gene fragment pairs, the gene interaction layer calls the pre-stored gene recombination rules to perform information fusion on gene fragment pairs with corresponding associations, generating fused gene fragments. The fused gene fragments contain the core information of gene fragments corresponding to the wheat germ feature gene set and the core information of gene fragments corresponding to the scenario requirement feature gene set in the gene fragment pair. For individual gene fragments that have not formed corresponding associations, the gene interaction layer supplements the association information through feature expansion technology, enabling individual gene fragments to form new associations with other fused gene fragments and generate supplementary gene fragments; The fusion gene fragment and the supplementary gene fragment are input into the gene sorting unit. The gene sorting unit arranges the fusion gene fragment and the supplementary gene fragment in sequence based on the conventional execution order of wheat germ processing and the functional attributes of the fusion gene fragment and the supplementary gene fragment. During the sorting process, the gene sorting unit introduces a dynamic adjustment mechanism to adjust the sorting results based on the capacity adaptation feature genes and resource constraint feature genes in the feature gene set of scenario requirements, ultimately forming a processing link feature gene sequence containing ordered gene fragments.

6. The method for recommending wheat germ processing schemes using deep learning according to claim 5, characterized in that, Based on the association strength value of gene fragment pairs, the gene interaction layer calls pre-stored gene recombination rules to perform information fusion on gene fragment pairs with corresponding associations, generating fused gene fragments, including: The association strength value of each gene fragment pair is calculated using the gene interaction layer. The association strength value is determined based on the attribute similarity, functional matching degree, and demand adaptability of the gene fragment corresponding to the wheat germ feature gene set and the gene fragment corresponding to the scene demand feature gene set in the gene fragment pair. Based on the preset range of association strength values, the gene interaction layer calls the corresponding gene recombination rules. Gene fragment pairs whose association strength values ​​meet the first preset range call the deep fusion rule, gene fragment pairs whose association strength values ​​meet the second preset range call the medium fusion rule, and gene fragment pairs whose association strength values ​​meet the third preset range call the basic fusion rule. For gene fragment pairs that invoke deep fusion rules, the gene interaction layer performs parameter-by-parameter fusion of all attribute parameters of the gene fragment corresponding to the wheat germ feature gene set in the gene fragment pair with all attribute parameters of the gene fragment corresponding to the scene requirement feature gene set, calculates the average value of the parameters after parameter-by-parameter fusion and retains the parameter fluctuation range before parameter-by-parameter fusion, and generates a deep fusion gene fragment containing complete parameter information. For gene fragment pairs that invoke the moderate fusion rule, the gene interaction layer selects the core attribute parameters of the gene fragment corresponding to the wheat germ feature gene set and the core attribute parameters of the gene fragment corresponding to the scene requirement feature gene set in the gene fragment pair and fuses them. It retains the original values ​​of the non-core parameters of the gene fragment corresponding to the wheat germ feature gene set and the original values ​​of the non-core parameters of the gene fragment corresponding to the scene requirement feature gene set in the gene fragment pair, and generates a moderately fused gene fragment containing the core fusion parameters and the original non-core parameters. For gene fragment pairs that invoke the basic fusion rules, the gene interaction layer only establishes the association identifier between the gene fragment corresponding to the wheat germ feature gene set and the gene fragment corresponding to the scene requirement feature gene set in the gene fragment pair, retains the complete parameter information of the gene fragment corresponding to the wheat germ feature gene set and the complete parameter information of the gene fragment corresponding to the scene requirement feature gene set in the gene fragment pair, and generates a basic fusion gene fragment with the association identifier.

7. The wheat germ processing scheme recommendation method combining deep learning according to claim 5, characterized in that, The process involves inputting the fusion gene fragment and the supplementary gene fragment into a gene sequencing unit. The gene sequencing unit, based on the conventional execution sequence of wheat germ processing and the functional attributes of the fusion gene fragment and the supplementary gene fragment, sequentially arranges the fusion gene fragment and the supplementary gene fragment, including: Establish a standard execution sequence library for wheat germ processing, which includes the standard execution sequences for common wheat germ processing steps such as cleaning, drying, grinding, extraction, and refining. The gene sequencing unit is used to label the functional attributes of the fusion gene fragment and the supplementary gene fragment, and to determine the wheat germ processing stage type corresponding to each fusion gene fragment and the wheat germ processing stage type corresponding to each supplementary gene fragment. For example, the fusion gene fragment related to the cleaning stage is labeled as the gene fragment related to the cleaning stage, and the supplementary gene fragment related to the drying stage is labeled as the gene fragment related to the drying stage. Based on the aforementioned conventional execution sequence library, the labeled fusion gene fragments and supplementary gene fragments are initially arranged according to the conventional order of the corresponding wheat germ processing steps to form a preliminary sorting sequence; For fusion gene fragments and supplementary gene fragments of the same wheat germ processing stage type in the preliminary sorting sequence, a secondary sorting is performed based on the functional subdivision attributes of the fusion gene fragments and supplementary gene fragments of the same wheat germ processing stage type. During the secondary sorting process, when there is a functional conflict between the fusion gene fragments and supplementary gene fragments of the same wheat germ processing stage type, the priority sorting rule is determined based on the application scenario feature genes in the scenario requirement feature gene set.

8. The method for recommending wheat germ processing schemes using deep learning according to claim 1, characterized in that, The process, based on the characteristic gene sequences of processing steps, combines processing step units through a deep scheme evolution model's scheme prototype construction layer to form multiple processing scheme prototypes, including: The feature gene sequences of processing links are analyzed by constructing the prototype layer of the deep scheme evolution model. The processing link type identifiers in the feature gene sequences of processing links are extracted to determine the types of processing link units that need to be combined, such as cleaning unit, drying unit, and grinding unit. Based on the processing step type identifier, call the processing step unit library and extract the standard processing step unit of the corresponding type. Each standard processing step unit contains the basic operation process and the range of common parameters. The gene fragment information in the characteristic gene sequence of the processing link is mapped to the corresponding standard processing link unit. The basic operation process and conventional parameter range of the corresponding standard processing link unit are adjusted so that the attributes of the adjusted processing link unit are consistent with the gene fragment information in the characteristic gene sequence of the processing link, thereby generating a customized processing link unit. Based on the sorting order of characteristic gene sequences in the processing steps, the customized processing steps are connected in series and combined to supplement the connection process between adjacent customized processing steps, forming a preliminary scheme framework. The sequence of customized processing units and the operation parameters of customized processing units in the initial scheme framework are adjusted to a limited extent to generate multiple scheme frameworks with differences. Each scheme framework with differences is supplemented with complete operation instructions to form a processing scheme prototype.

9. The method for recommending wheat germ processing schemes using deep learning according to claim 1, characterized in that, The process involves a double-loop adaptation and optimization of multiple processing scheme prototypes with wheat germ characteristic gene sets and scenario requirement characteristic gene sets to obtain a target processing recommendation scheme, including: In the first round of optimization, each processing scheme prototype is matched with the wheat germ characteristic gene set. Processing steps that do not match the gene fragments in the wheat germ characteristic gene set are extracted from each processing scheme prototype. The operation parameters and operation process of the processing steps that do not match the gene fragments in the wheat germ characteristic gene set are adjusted to improve the matching degree between each processing scheme prototype and the wheat germ characteristic gene set. After completing the first round of optimization, the second round of optimization is entered. The optimized processing scheme prototype obtained from the first round of optimization is matched with the set of scene requirement feature genes. The processing links in the optimized processing scheme prototype obtained from the first round of optimization that do not meet the requirements of the set of scene requirement feature genes are identified. The order and operation rules of the processing links that do not meet the requirements of the set of scene requirement feature genes are adjusted based on the gene fragments in the set of scene requirement feature genes. Between the first and second optimization cycles, the parameter information of the processing steps adjusted in the first optimization cycle is fed back to the parsing process of the scene requirement feature gene set to correct the correlation between gene fragments in the scene requirement feature gene set. The first and second optimization cycles are repeated. After each cycle, the comprehensive fit value of each processing scheme prototype with the wheat germ characteristic gene set and the scenario requirement characteristic gene set is calculated. The comprehensive fit value is determined as follows: First, the fit degree of each processing scheme prototype with the wheat germ characteristic gene set and with the scenario requirement characteristic gene set is normalized to the same evaluation scale. Then, a weighted calculation is performed based on the normalized results. When the overall fit value reaches the preset fit standard and the change in the overall fit value after two consecutive executions of the first and second optimization cycles is less than the preset threshold, the optimization cycle is stopped, and the processing scheme prototype with the highest overall fit value is selected as the target processing recommendation scheme.

10. A wheat germ processing scheme recommendation system combining deep learning, characterized in that, The wheat germ processing scheme recommendation system combining deep learning includes a processor and a memory, the memory and the processor are connected, the memory is used to store programs, instructions or code, and the processor is used to run the programs, instructions or code in the memory to implement the wheat germ processing scheme recommendation method combining deep learning as described in any one of claims 1-9.

Images