A method for ruminant feed nutrition regulation and continuous production integration

Abstract

The application discloses a kind of ruminant feed nutrition regulation and continuous production integration method, it is related to ruminant feed processing technical field.The method includes: the lattice structure characteristic analysis of execution across process to ruminant feed production chain in raw material to be processed, generates and includes the raw material nutrition characteristic vector of crude fiber crystallinity and protein molecular conformation;Based on raw material nutrition characteristic vector and preset rumen degradation rate benchmark library carry out multidimensional matching, determine the degradation rate grade and fermentation acid production potential value of each raw material component;According to degradation rate grade and fermentation acid production potential value generation includes layered feeding time sequence table of fast, medium speed and slow speed degradation component;According to layered feeding time sequence table control the steam pressure gradient in conditioning device, so that the raw material component of different degradation rate grade is sequentially experienced in conditioning device different intensity hydrothermal modification interval, generate mixture material with gradient gelatinization degree distribution.

Description

Technical Field

[0001] This invention relates to the field of ruminant feed processing technology, specifically to an integrated method for ruminant feed nutrition regulation and continuous production. Background Technology

[0002] In ruminant feed production, the degradation rates and acid production patterns of different feed ingredients in the rumen vary significantly. Precisely controlling the degradation sequence of different feed components in the rumen through processing technology is a key technical challenge for achieving efficient nutrient utilization. Existing feed conditioning and mixing methods typically apply uniform process parameters to the entire batch of raw materials. This approach ignores the essential influence of the raw material's own fiber structure and protein conformation on degradation behavior. Because the microstructural characteristics of each feed component cannot be known before processing, the steam application parameters and stirring intensity during conditioning are empirically fixed, forcing feed components with different physicochemical properties to undergo the same intensity of hydrothermal modification treatment.

[0003] The shortcomings of existing technical solutions lie in their failure to achieve cross-process analysis of the nutritional characteristics of raw materials and differentiated process responses. On the one hand, key structural information such as the crystallinity of crude fiber and the orderliness of protein secondary structure cannot be obtained in real time due to the separation of conventional detection methods from the production process, and the processing system cannot identify the differences in degradation potential among the raw material components within the same batch. On the other hand, the conditioner, as a single homogenization unit, cannot actively construct a differentiated gelatinization distribution based on the target degradation order of raw material components in the rumen. All components ultimately achieve similar degrees of gelatinization, making it impossible for rapidly degrading starch components and slowly degrading fiber components to achieve temporal matching of energy release and protein utilization in the rumen.

[0004] The problems to be addressed by this invention include: how to analyze the fibrous crystalline state and protein conformation state of raw materials before they enter deep processing steps such as crushing, without interrupting the continuity of production, to form a raw material nutritional feature vector that can characterize the rumen degradation trend; and how to use this feature vector to drive the conditioning system so that raw material components with different degradation characteristics can obtain customized hydrothermal modification intensity in a continuous production process, and the final mixture produced has a differentiated degradation activity gradient. Summary of the Invention

[0005] The purpose of this invention is to provide an integrated method for ruminant feed nutrition regulation and continuous production, so as to realize online analysis of the nutritional characteristics of raw materials during continuous production, and automatically generate differentiated stratified conditioning processes based on the analysis results, so that raw material components with different degradation characteristics can be subjected to gradient processing that matches their rumen degradation sequence.

[0006] To achieve the above objectives, the present invention provides the following technical solution: The present invention provides an integrated method for ruminant feed nutrition regulation and continuous production, the method comprising: This method performs cross-process lattice structure feature analysis on raw materials to be processed in the ruminant feed production chain, generating a raw material nutrient feature vector containing crude fiber crystallinity and protein molecular conformation. Based on the raw material nutrient feature vector, multi-dimensional matching is performed with a preset rumen degradation rate benchmark library to determine the degradation rate level and fermentation acid production potential value of each raw material component. According to the degradation rate level and fermentation acid production potential value of each raw material component, a stratified feeding sequence table containing fast-degrading components, medium-degrading components, and slow-degrading components is generated. According to the stratified feeding sequence table, the steam pressure gradient in the conditioning device is controlled so that raw material components with different degradation rate levels undergo hydrothermal modification intervals of different intensities in the conditioner, generating a mixture with a gradient gelatinization degree distribution. This method, by pre-analyzing the microscopic nutrient structure of raw materials before they enter key processes such as crushing, connects the traditional discrete processing links into a continuous production process with nutrient degradation characteristics as the main control line, realizing feedforward precise control of the rumen fermentation rhythm.

[0007] As a preferred embodiment of the present invention, the step of performing cross-process lattice structure feature analysis on the raw materials to be processed in the ruminant feed production chain to generate a raw material nutritional feature vector containing crude fiber crystallinity and protein molecule conformation specifically includes: after the raw materials to be processed have passed through a primary cleaning sieve to remove large particulate impurities and before entering the pulverizer, using a near-infrared spectral sensor array to collect multi-point diffuse reflectance spectra of the raw materials to be processed in the conveying state, obtaining an absorbance value sequence at characteristic absorption wavelengths at each point; calculating a crystallinity index characterizing the anti-degradation ability of the crude fiber component based on the ratio of the intensity of the characteristic absorption peak of cellulose to the intensity of the characteristic absorption peak of the amorphous region in the absorbance value sequence; simultaneously, calculating a secondary structure order index characterizing the ease with which the protein component is attached and decomposed by microorganisms based on the ratio of the absorption peak area of ​​amide I band to amide II band in the absorbance value sequence; and concatenating the crystallinity index and the secondary structure order index into vectors to generate the raw material nutritional feature vector. This approach simultaneously decouples the microstructural information of two key nutritional dimensions, fiber and protein, in a single spectral scan, providing a deeper rumen availability criterion for subsequent regulation than conventional component detection.

[0008] As a further preferred embodiment of the present invention, the step of determining the degradation rate level and fermentation acid production potential value of each raw material component by performing multi-dimensional matching based on the raw material nutrient feature vector and a preset rumen degradation rate benchmark library specifically includes: extracting a crystallinity index from the raw material nutrient feature vector as a first matching dimension, calculating its Euclidean distance with each benchmark crystallinity interval in the rumen degradation rate benchmark library, and assigning the degradation rate level corresponding to the interval with the closest distance to the current raw material component, thereby achieving direct classification of its rate of decomposition by microorganisms in the rumen; extracting a secondary structure order index from the raw material nutrient feature vector as a second matching dimension, retrieving multiple acid production potential templates that are related to the determined degradation rate level in the rumen degradation rate benchmark library, selecting the template corresponding to the smallest difference by calculating the absolute value of the difference between the order index and the preset reference value of each template, thereby assigning the raw material component a specific fermentation acid production potential value to reflect its tendency and ability to generate volatile fatty acids during degradation; and storing the degradation rate level and the fermentation acid production potential value in a raw material component feature record table. This process establishes a mapping bridge from "static chemical composition" to "dynamic rumen behavior" through dual analysis of microstructure and templated comparison of historical degradation data.

[0009] Based on the above technical solution, the preferred implementation of generating a stratified feeding sequence table containing fast-degrading, medium-degrading, and slow-degrading components according to the degradation rate level and fermentation acid production potential value of each raw material component is as follows: raw material components with a degradation rate level of 1 are marked as fast-degrading components, raw material components with a degradation rate level of 2 are marked as medium-degrading components, and raw material components with a degradation rate level of 3 are marked as slow-degrading components; the raw material components within each marked group are arranged in descending order of their fermentation acid production potential value to generate fast feeding subsequences, medium feeding subsequences, and slow feeding subsequences respectively, and different feeding time intervals are set for adjacent raw material components in each subsequence; subsequently, the fast feeding subsequences, medium feeding subsequences, and slow feeding subsequences are spliced ​​together end to end according to the time evolution direction, and a preset delay time window is inserted between the sequences to generate the stratified feeding sequence table. This timing table reconstructs discrete raw material individuals into feed clusters with time and functional gradients, ensuring that differentiated processing can be completed sequentially in a subsequent single conditioning unit.

[0010] Accordingly, generating a mixture with a gradient gelatinization degree distribution specifically includes: reading the current raw material component and its degradation rate level corresponding to the current timestamp from the layered feeding time sequence table; if it is a fast-degrading component, controlling the first steam regulating valve of the conditioning device to open to a specific degree, so that the steam pressure is maintained in the fast degradation conditioning pressure range with higher energy input, and simultaneously driving the stirring shaft to run at a corresponding speed, so that this part of the material can obtain sufficient hydrothermal shear in a short time, and is discharged after reaching the first gelatinization degree target value; if it is a medium-degrading component, switching to the second... Two steam regulating valves are adjusted to their respective openings to adapt the steam pressure and stirring speed to the medium-speed hydrothermal modification range, giving the material the second gelatinization target value. For slow-degrading components, a mild slow-degradation conditioning pressure range is constructed through a third steam regulating valve, supplemented by an appropriate speed, to allow the material to complete shallow or directional gelatinization, reaching the third gelatinization target value. When the amount of modified material discharged from the outlet, accumulated according to different gelatinization degrees over time, reaches the single-batch mixing threshold, the mixer is started to mix the modified materials with gradient gelatinization degree distribution into one, generating a mixture. The resulting mixture has different starch gelatinization and protein denaturation degrees achieved according to the preset formula. After entering the rumen, the fast-degrading component ferments first to provide energy, the medium-speed component takes over to degrade and stabilize the rumen pH, and the slow component acts as a fibrous skeleton to maintain rumen peristalsis and long-term energy supply, thus synergistically optimizing the rhythm and efficiency of rumen fermentation from the feeding end.

[0011] To ensure the robustness and representativeness of the spectral data acquisition, after the raw material to be processed has passed through the primary cleaning screen to remove large particulate impurities and before entering the crusher, multi-point diffuse reflectance spectral acquisition of the raw material is performed using a near-infrared spectral sensor array. This includes: arranging multiple spectral acquisition points at equal intervals on the conveyor belt between the primary cleaning screen outlet and the crusher inlet; as the raw material passes each point on the conveyor belt, the corresponding near-infrared spectral sensor is triggered to initiate multiple scans; the average of the spectral data obtained from several scans at the same point is calculated according to wavelength to generate the average diffuse reflectance spectrum for that point; finally, the average diffuse reflectance spectra of all acquisition points are arranged in a spatial sequence to generate a multi-point diffuse reflectance spectral sequence. This process smooths out the spectral noise caused by the inhomogeneity of the raw material in both spatial and frequency dimensions, improving the accuracy of subsequent calculations of crystallinity and order index.

[0012] In calculating key nutritional indicators, as a preferred method, the calculation of the crystallinity index of the crude fiber component in the raw material to be processed based on the ratio of the intensity of the characteristic absorption peak of cellulose to the intensity of the characteristic absorption peak of the amorphous region in the absorbance value sequence includes: extracting the first absorbance value at the characteristic wavelength of the cellulose crystalline region from each sampling point in the multi-point diffuse reflectance spectral sequence and performing a global arithmetic mean to obtain the global average absorbance of the crystalline region; simultaneously extracting the second absorbance value at the characteristic wavelength of the amorphous region from each sampling point and performing a global arithmetic mean to obtain the global average absorbance of the amorphous region; and using the ratio of the global average absorbance of the crystalline region to the global average absorbance of the amorphous region as the crystallinity index. Similarly, calculating the secondary structure order index of the protein components in the raw material to be processed based on the ratio of the absorption peak areas of the amide I band and the amide II band in the absorbance value sequence includes: extracting the first integrated absorbance values ​​within the characteristic absorption range of the amide I band from the multi-site diffuse reflectance spectral sequence and arithmetically averaging them, and extracting the second integrated absorbance values ​​within the characteristic absorption range of the amide II band and arithmetically averaging them, respectively, to obtain the global integrated mean of the amide I band and the global integrated mean of the amide II band, and using the ratio of the two as the secondary structure order index. This index determination method based on global mean calculation ensures that even in the dynamic environment of industrial transport, the extracted feature vector can still stably reflect the overall nutritional conformation of the batch.

[0013] As an important supplement to this invention, the rumen degradation rate benchmark library contains multiple benchmark crystallinity intervals and their mapped degradation rate levels. Each degradation rate level is associated with multiple acid production potential templates, which clearly record the correspondence between preset orderliness reference values ​​and fermentation acid production potential values. The mapping relationship and template data on which this benchmark library relies are dynamically updated and corrected based on periodically imported historical production batches of rumen degradation test data for ruminants, ensuring that the matching decisions continuously align with actual feeding effects. Simultaneously, the feeding timestamps for the first, second, and third time intervals are dynamically determined based on the difference in fermentation acid production potential values ​​of adjacent raw material components in each feeding subsequence. The larger the difference in fermentation acid production potential values, the shorter the feeding interval, so as to form synergistic fermentation clusters through dense feeding, and rapidly establish a synchronous degradation microenvironment in the rumen microbiota.

[0014] The technical effects and advantages provided by the present invention in the above technical solution are as follows: By placing the lattice structure feature analysis step after the initial cleaning and screening process and before the grinding process, a near-infrared spectral sensor array is used to perform multi-point diffuse reflectance spectral acquisition of the raw materials to be processed on the conveyor belt. Based on the acquired absorbance value sequence, the crystallinity index of the crude fiber component and the secondary structure order index of the protein component are calculated simultaneously to generate a raw material nutritional feature vector. This vectorized characterization transforms the degradation trend of the raw materials in the rumen from an unquantifiable empirical judgment into a calculable quantitative indicator. This cross-process online analysis method allows the structural information of the raw materials to be extracted before entering the grinding and conditioning stages. It changes the situation in traditional processing where the microstructure state of the raw materials is completely "black box" to the subsequent control system. It provides a direct basis from the crystal structure and molecular conformation state of the raw materials themselves for the differentiated setting of conditioning parameters. The structural fluctuations between different batches of incoming materials are reflected in real time, ensuring the objectivity and real-time response capability of the stratified conditioning decision.

[0015] Based on the feature vector composed of the crystallinity index and the secondary structure order index, multi-dimensional matching with the rumen degradation rate benchmark library can simultaneously determine the degradation rate level and fermentation acid production potential of each raw material component, and construct a stratified feeding sequence table based on this. Within this sequence table, time-sharing feeding sequences are established according to the classification of fast-degrading components, medium-degrading components, and slow-degrading components, with delay time windows inserted between feeding sequences of adjacent categories. Based on this sequence table, the steam pressure gradient and stirring speed within the conditioning unit are controlled, allowing raw material components with different degradation rate levels to sequentially experience three differentiated hydrothermal modification zones: low-pressure fast, medium-pressure medium-speed, and high-pressure slow. Ultimately, a mixture with a gradient gelatinization degree distribution exhibiting shallow gelatinization, moderate gelatinization, and deep gelatinization is obtained. This gradient conditioning approach, tailored to the specific materials, allows for moderate control of the gelatinization degree of rapidly degrading starch components to slow down their rumen degradation rate. At the same time, it enables sufficient hydrothermal loosening of the structure of slowly degrading fiber components to enhance their degradation rate. The resulting mixture forms a mutually compensating degradation activity gradient among the components, promoting a near-synchronous release sequence of carbohydrates and proteins in the rumen. Attached Figure Description

[0016] To more clearly illustrate the technical solutions in the embodiments of this application or the prior art, the drawings used in the embodiments will be briefly introduced below. Obviously, the drawings described below are only some embodiments recorded in this invention. For those skilled in the art, other drawings can be obtained based on these drawings.

[0017] Figure 1 This is a flowchart of an integrated method for ruminant feed nutrition regulation and continuous production; Figure 2 This is a flowchart of a method for generating multi-point diffuse reflectance spectral sequences; Figure 3 This is a schematic diagram of the process for generating raw material nutrient feature vectors; Figure 4 It is a flowchart for matching raw material degradation and fermentation characteristics based on a rumen degradation rate benchmark library; Figure 5 This is a flowchart for generating the layered material feeding sequence table; Figure 6 It is a flowchart of the layered feeding, grading, conditioning and gradient gelatinization mixing control process. Detailed Implementation

[0018] To make the objectives, technical solutions, and advantages of the embodiments of the present invention clearer, the technical solutions of the embodiments of the present invention will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only some embodiments of the present invention, not all embodiments. Based on the embodiments of the present invention, all other embodiments obtained by those skilled in the art without creative effort are within the scope of protection of the present invention.

[0019] See Figure 1 This invention provides an integrated method for nutrient regulation and continuous production of ruminant feed. The method includes: performing cross-process lattice structure feature analysis on raw materials to be processed in the ruminant feed production chain to generate a raw material nutrient feature vector containing crude fiber crystallinity and protein molecule conformation; performing multi-dimensional matching based on the raw material nutrient feature vector and a preset rumen degradation rate benchmark library to determine the degradation rate level and fermentation acid production potential value of each raw material component; generating a stratified feeding sequence table containing fast-degrading components, medium-degrading components, and slow-degrading components according to the degradation rate level and fermentation acid production potential value of each raw material component; and controlling the steam pressure gradient in the conditioning device according to the stratified feeding sequence table, so that raw material components with different degradation rate levels sequentially undergo hydrothermal modification intervals of different intensities in the conditioner, generating a mixture with a gradient gelatinization degree distribution.

[0020] Example 1: In specific implementation, please refer to Figure 2 In the integrated method of ruminant feed nutrition regulation and continuous production, the raw materials to be processed are conveyed to the feed inlet of the crusher via a conveyor belt after primary cleaning and screening. On the conveyor belt between the discharge outlet of the primary cleaning screen and the feed inlet of the crusher, multiple near-infrared spectral sensors are evenly spaced along the longitudinal direction of the conveyor belt, forming a near-infrared spectral sensor array. The location of each near-infrared spectral sensor constitutes a spectral acquisition point.

[0021] In practice, the conveyor belt runs at a preset constant speed, and the raw material to be processed forms a material layer of a certain thickness on the conveyor belt. When the raw material moves with the conveyor belt to the first spectral acquisition point, the position sensor located directly above the conveyor belt detects the arrival of the material and sends a trigger signal to the corresponding near-infrared spectral sensor, which then starts scanning. For the same batch of raw material, the near-infrared spectral sensor continuously acquires diffuse reflectance spectral data three times at fixed time intervals within the time it takes for the material to pass through its effective scanning area. Each acquisition obtains absorbance values ​​at multiple discrete wavelength points within the range of a first preset wavelength to a second preset wavelength. The three acquisition times are recorded as the first time point, the second time point, and the third time point, respectively.

[0022] In some embodiments, the near-infrared spectral sensor integrates a light source and a detector. Near-infrared light emitted by the light source illuminates the surface of the raw material to be processed, and the detector receives the light signal diffusely reflected from within the raw material and converts it into spectral data. During each acquisition, the near-infrared spectral sensor operates at a wavelength... Obtain an absorbance value at the specified location.

[0023] For this spectral acquisition point, the diffuse reflectance spectral data acquired at the first time point will be displayed at wavelength... The absorbance value at that point is recorded as The diffuse reflectance spectral data collected at the second time point are then analyzed at wavelengths... The absorbance value at that point is recorded as The diffuse reflectance spectral data collected at the third time point were then analyzed at wavelengths... The absorbance value at that point is recorded as ,in This refers to the serial number of the spectral acquisition point. The value is taken from 1 to the total number of points, based on the number of points set at equal intervals. The wavelength of the spectral acquisition point is calculated using the following formula. Average absorbance at :

[0024] in, Indicates the sequence number is The spectral acquisition point is at wavelength Average absorbance at that location Indicates the sequence number is When performing the first diffuse reflectance spectral acquisition at the spectral acquisition point, at the wavelength The absorbance value at that point Indicates the sequence number is When performing a second diffuse reflectance spectral acquisition at the spectral acquisition point, at the wavelength The absorbance value at that point Indicates the sequence number is When performing the third diffuse reflectance spectral acquisition at the spectral acquisition point, at the wavelength The absorbance value at that point. Wavelength Iterate through every discrete wavelength point within the range from the first preset wavelength to the second preset wavelength.

[0025] In practical implementation, the serial number obtained after the above calculation will be... The curve formed by the average absorbance of a given spectral acquisition point at all discrete wavelengths is used as the average diffuse reflectance spectrum corresponding to that acquisition point. As the material continues to move with the conveyor belt to the next spectral acquisition point, the near-infrared spectral sensor at that next acquisition point is triggered in the same manner and acquires three diffuse reflectance spectral data points, calculating the average diffuse reflectance spectrum for that acquisition point. This acquisition and calculation process is sequentially performed on all equally spaced spectral acquisition points to obtain the average diffuse reflectance spectrum corresponding to each acquisition point.

[0026] After all spectral acquisition points have been collected and calculated, the average diffuse reflectance spectra of each acquisition point are arranged sequentially along the material conveying direction on the conveyor belt, generating a multi-point diffuse reflectance spectrum sequence. This multi-point diffuse reflectance spectrum sequence contains the average absorbance information of each acquisition point at all discrete wavelengths. Optionally, the spacing value set above is determined based on the width of the conveyor belt and the coverage area of ​​the near-infrared spectral sensor array to ensure representative sampling of the cross-section of the raw material to be processed.

[0027] Example 2: In specific implementation, please refer to Figure 3 Based on the generated multi-site diffuse reflectance spectral sequence, cross-process lattice structure feature analysis is performed to generate a raw material nutritional feature vector containing crude fiber crystallinity and protein molecule conformation. In the multi-site diffuse reflectance spectral sequence, each spectral acquisition point corresponds to an average diffuse reflectance spectrum, and each average diffuse reflectance spectrum consists of the average absorbance values ​​of multiple discrete wavelength points within a first preset wavelength to a second preset wavelength range. From the average diffuse reflectance spectrum of each spectral acquisition point, the absorbance value at the characteristic wavelength of the cellulose crystalline region is extracted as the first absorbance value. The specific value of the characteristic wavelength of the cellulose crystalline region is pre-calibrated based on the near-infrared absorption response characteristics corresponding to the ordered arrangement of hydrogen bonds in the cellulose molecular chain. The arithmetic mean of the first absorbance values ​​of all spectral acquisition points is calculated to obtain the global average absorbance of the crystalline region.

[0028] In practice, the absorbance value at the characteristic wavelength of the amorphous region is extracted from the average diffuse reflectance spectrum of each spectral acquisition point as the second absorbance value. The specific value of the characteristic wavelength of the amorphous region is pre-calibrated based on the near-infrared absorption response characteristics corresponding to the disordered regions in the cellulose molecular chain. The arithmetic mean of the second absorbance values ​​of all spectral acquisition points is obtained to obtain the global average absorbance of the amorphous region. In practice, the ratio of the global average absorbance of the crystalline region to the global average absorbance of the amorphous region is calculated, and the obtained ratio is directly used as the crystallinity index of the crude fiber component in the raw material to be processed.

[0029] In practice, spectral data within the characteristic absorption range of the amide I band are extracted from the average diffuse reflectance spectrum of each spectral acquisition point and integrated to obtain the first integrated absorbance value. The characteristic absorption range of the amide I band is pre-calibrated based on the spectral absorption interval corresponding to the carbonyl stretching vibration in the protein molecule. The arithmetic mean of the first integrated absorbance values ​​of all spectral acquisition points is then calculated to obtain the global integrated mean of the amide I band.

[0030] In practice, spectral data within the characteristic absorption range of the amide II band are extracted from the average diffuse reflectance spectrum of each spectral acquisition point and integrated to obtain the second integrated absorbance value. The characteristic absorption range of the amide II band is pre-calibrated based on the spectral absorption interval corresponding to the coupling of nitrogen-hydrogen bending vibrations and carbon-nitrogen stretching vibrations in the protein molecule. The arithmetic mean of the second integrated absorbance values ​​of all spectral acquisition points is then obtained to obtain the global integrated mean of the amide II band.

[0031] In specific implementation, the ratio of the global average integral value of amide I band to the global average integral value of amide II band is calculated, and the obtained ratio is directly used as the secondary structure order index of the protein component in the raw material to be processed. In some embodiments, the secondary structure order index can be calculated according to the following formula:

[0032] in, It represents the index of the secondary structure order of protein components in the raw material to be processed. This represents the global integral mean of the amide I band, obtained by arithmetically averaging the first integrated absorbance values ​​of all spectral acquisition points within the characteristic absorption range of the amide I band. The global integral mean of the amide II band is obtained by arithmetically averaging the second integral absorbance values ​​of all spectral acquisition points within the characteristic absorption range of the amide II band.

[0033] In practice, the calculated crystallinity index and secondary structure order index are concatenated as vectors to generate a raw material nutritional feature vector. The vector concatenation method involves using the crystallinity index as the first dimension element and the secondary structure order index as the second dimension element to form a two-dimensional feature vector.

[0034] Example 3: In specific implementation, please refer to Figure 4 A rumen degradation rate benchmark library is pre-generated and stored. This library contains multiple benchmark crystallinity ranges and corresponding degradation rate levels. Each degradation rate level is associated with multiple acid production potential templates. Each acid production potential template contains a preset ordering reference value and its corresponding fermentation acid production potential value. The data in the rumen degradation rate benchmark library is updated periodically based on historical production batches of rumen degradation test data from rumen animals.

[0035] In practice, raw materials from historical production batches undergo rumen degradation experiments. The experiment involves placing raw material samples in nylon bags inside the rumen of ruminants equipped with rumen fistulas. The nylon bags are removed at multiple time points, and the residual sample content is measured. The dry matter degradation rate at each time point is calculated. The dry matter degradation rate data is fitted using an exponential curve model, and degradation rate levels are determined based on the rate constant of the fitted curve. Simultaneously, the crystallinity and protein secondary structure order values ​​of the raw material samples are measured to establish a correlation between crystallinity values ​​and degradation rate levels. Crystallinity values ​​are divided into multiple benchmark crystallinity intervals, each corresponding to a degradation rate level. For raw material samples within each degradation rate level, the in vitro fermentation gas production or volatile fatty acid concentration is further measured. The protein secondary structure order value is correlated with the fermentation acid production potential value to generate multiple acid production potential templates. All of the above data is entered into a rumen degradation rate benchmark library for regular updates.

[0036] In specific implementation, after generating the raw material nutrient feature vector through the aforementioned steps, the crystallinity index is extracted from the raw material nutrient feature vector as the first matching dimension. The crystallinity index is then compared one by one with each benchmark crystallinity interval in the rumen degradation rate benchmark library using Euclidean distance. The calculation method is to take the absolute value of the difference between the crystallinity index and the center value of the current benchmark crystallinity interval; this absolute value is the Euclidean distance. This process is repeated for all benchmark crystallinity intervals in the rumen degradation rate benchmark library, calculating the Euclidean distance for each interval. All calculated Euclidean distances are compared, and the benchmark crystallinity interval with the smallest Euclidean distance value is selected. The degradation rate level corresponding to the selected benchmark crystallinity interval in the rumen degradation rate benchmark library is then used as the degradation rate level of the current raw material component. In some embodiments, the center value of the benchmark crystallinity interval is defined as the arithmetic mean of the upper and lower limits of the benchmark crystallinity interval.

[0037] In practice, a secondary structure orderliness index is extracted from the raw material nutrient feature vector as the second matching dimension. In the rumen degradation rate benchmark library, based on the determined degradation rate level of the current raw material component, all acid-producing potential templates associated with that degradation rate level are retrieved. Each acid-producing potential template contains a preset orderliness reference value and a fermentation acid-producing potential value. The extracted secondary structure orderliness index is calculated one by one with the preset orderliness reference value in each retrieved acid-producing potential template by taking the absolute value of the difference between the secondary structure orderliness index and the preset orderliness reference value. All acid-producing potential templates associated with that degradation rate level are traversed, and the absolute value of the difference corresponding to each acid-producing potential template is calculated. All calculated absolute values ​​of difference are compared, and the acid-producing potential template with the smallest absolute value of difference is selected. The fermentation acid-producing potential value in the selected acid-producing potential template is taken as the fermentation acid-producing potential value of the current raw material component.

[0038] In practical implementation, the minimum value of the absolute value of the difference can be calculated using the following formula:

[0039] in, The index representing the secondary structure orderliness of the current raw material components is related to the first... The absolute value of the difference between the preset orderliness reference values ​​in each acid production potential template. This represents the index of the secondary structure order of the current raw material components. Indicates the first The preset orderliness reference value recorded in each acid production potential template This is an integer index ranging from 1 to the total number of acid-producing potential templates stored in relation to the current degradation rate level. (The index is used for...) Starting from 1, iterate up to the total number of acid-producing potential templates to calculate the corresponding templates for all acid-producing potential templates. Value, select to make Take the smallest value The template corresponding to the value of acid production potential.

[0040] In practice, the determined degradation rate level and fermentation acid production potential value are associated and stored in the raw material component characteristic record table. The raw material component characteristic record table includes a raw material component name field, a degradation rate level field, and a fermentation acid production potential value field. Each raw material component corresponds to one record, which stores both the degradation rate level and the fermentation acid production potential value.

[0041] Example 4: In specific implementation, please refer to Figure 5 Based on the degradation rate level and fermentation acid production potential value of each raw material component stored in the raw material component characteristic record table, a stratified feeding sequence table is generated. The degradation rate level is divided into three categories: raw material components with a degradation rate level of 1 are marked as fast-degrading components, raw material components with a degradation rate level of 2 are marked as medium-degrading components, and raw material components with a degradation rate level of 3 are marked as slow-degrading components.

[0042] In practical implementation, for each raw material component marked as a rapidly degradable component, the fermentation acid production potential value of each rapidly degradable component is extracted. All rapidly degradable components are arranged in descending order of their fermentation acid production potential values ​​to form an ordered sequence, which is the rapidly feeding subsequence. Each element in the rapidly feeding subsequence corresponds to a rapidly degradable component, and the order in which the element is placed in the sequence determines the feeding order of that rapidly degradable component. Based on the arrangement order of each component in the rapidly feeding subsequence, a first time interval feeding timestamp is set for each rapidly degradable component in the sequence. The timestamp of the first rapidly degradable component in the rapidly feeding subsequence is set as the preset rapid feeding start time. For each rapidly degradable component in the second and subsequent positions in the rapidly feeding subsequence, the first time interval feeding timestamp of that rapidly degradable component is determined as follows: the first time interval feeding timestamp of the rapid degradable component preceding it in the rapidly feeding subsequence is obtained, and a first time interval duration dynamically determined based on the difference in fermentation acid production potential values ​​between two adjacent rapidly degradable components is added. The difference in fermentation acid production potential is the absolute value of the difference between the fermentation acid production potential of the first rapidly degrading component and the fermentation acid production potential of the second rapidly degrading component. The principle for determining the duration of the first time interval is that the larger the difference in fermentation acid production potential, the shorter the corresponding first time interval.

[0043] In practical implementation, the dynamic calculation of the duration of the first time interval can be expressed by the following formula:

[0044] in, This indicates the sequence number in the rapid feeding subsequence. The rapidly degradable components ranked in the first place and those ranked in the second place The duration of the first time interval between rapidly degrading components at a given location. The integer index starting from 2 up to the total number of rapidly degradable components in the rapid feeding subsequence; This indicates the sequence number in the rapid feeding subsequence. The fermentation acid production potential value of the rapidly degradable component at the site. This indicates the sequence number in the rapid feeding subsequence. The fermentation acid production potential value of the rapidly degradable components at the site; This represents the conversion factor for the feeding interval of rapidly degradable components. The value is calculated in advance based on the maximum allowable feeding rate of the conditioning unit when processing rapidly degrading components and the influence factor of the difference in fermentation acid production potential value on the feeding rate. This value is written into the control system parameter area when the conditioning unit controller is initialized and remains constant within a production batch.

[0045] In practice, for each raw material component marked as a medium-rate degradation component, a medium-rate feeding subsequence is generated in the same manner. The fermentation acid production potential values ​​of all medium-rate degradation components are extracted and arranged in descending order of these values ​​to generate the medium-rate feeding subsequence. A second time interval feeding timestamp is assigned to each medium-rate degradation component in the medium-rate feeding subsequence. The timestamp of the medium-rate degradation component ranked first in the medium-rate feeding subsequence is set as the preset medium-rate feeding start time. For the second and subsequent medium-rate degradation components, the second time interval feeding timestamp is equal to the second time interval feeding timestamp of the previous medium-rate degradation component plus the second time interval duration. The second time interval duration is dynamically determined based on the difference in fermentation acid production potential values ​​between two adjacent medium-rate degradation components; the larger the difference in fermentation acid production potential values, the shorter the second time interval duration. The calculation method for the second time interval duration has the same functional form as the calculation method for the first time interval duration, but uses an independently set medium-rate degradation component feeding interval conversion coefficient.

[0046] In practice, for each raw material component marked as a slow-degrading component, a slow-feeding subsequence is generated. All slow-degrading components are arranged from highest to lowest according to their fermentation acid production potential, generating the slow-feeding subsequence. A third time interval feeding timestamp is assigned to each slow-degrading component in the slow-feeding subsequence. The timestamp of the slow-degrading component ranked first in the slow-feeding subsequence is set as the preset slow-feeding start time. For the second and subsequent slow-degrading components, the third time interval feeding timestamp is equal to the third time interval feeding timestamp of the previous slow-degrading component plus the third time interval duration. The third time interval duration is dynamically determined based on the difference in fermentation acid production potential between two adjacent slow-degrading components; the larger the difference in fermentation acid production potential, the shorter the third time interval duration. The calculation method for the third time interval duration is the same as before, using an independently set slow-degrading component feeding interval conversion factor.

[0047] In practice, after setting the timestamps for each of the rapid, medium-speed, and slow feeding subsequences, the three feeding subsequences are spliced ​​together in chronological order. The splicing method is as follows: all timestamps of the rapid feeding subsequence are placed in the earliest time segment; a first delay time window is inserted between the last timestamp of the rapid feeding subsequence and the first timestamp of the medium-speed feeding subsequence. The duration of this first delay time window is pre-set based on the steam pressure stabilization time required for the conditioning unit to switch from the rapid degradation conditioning pressure range to the medium-speed degradation conditioning pressure range. This stabilization time is obtained through a step response test of the conditioning unit's steam pipeline and stored in the control system; a second delay time window is inserted between the last timestamp of the medium-speed feeding subsequence and the first timestamp of the slow feeding subsequence. The duration of this second delay time window is also pre-set based on the steam pressure stabilization time required for the conditioning unit to switch from the medium-speed degradation conditioning pressure range to the slow degradation conditioning pressure range. The complete time sequence obtained after splicing is the stratified feeding sequence table. The stratified feeding sequence table includes the component identifier, degradation rate level, fermentation acid production potential value, and feeding trigger timestamp for each raw material component.

[0048] Example 5: In specific implementation, please refer to Figure 6 Following the time sequence specified in the stratified feeding schedule, the control system monitors the current time in real time and reads the current raw material component and its degradation rate level corresponding to the feeding trigger timestamp matching the current time from the stratified feeding schedule. Each record in the stratified feeding schedule contains a component identifier, degradation rate level, fermentation acid production potential value, and feeding trigger timestamp. The control system processes each record one by one according to the order of the feeding trigger timestamps.

[0049] In practical implementation, when the control system reads that the degradation rate level of the current raw material component is Level 1, it sends a control command to the first steam regulating valve of the conditioning device, controlling the valve core of the first steam regulating valve to open from the closed state to the first opening value. The first opening value is calculated based on the target pressure value of the rapid degradation conditioning pressure range and the main steam pressure value of the steam supply pipeline of the conditioning device, using the flow characteristic curve of the steam regulating valve. After the first steam regulating valve opens to the first opening value, steam enters the conditioning chamber through the first steam regulating valve, causing the steam pressure in the conditioning chamber to rise from the current pressure value and be maintained within the rapid degradation conditioning pressure range. The lower limit of the rapid degradation conditioning pressure range is the preset minimum rapid degradation conditioning pressure, and the upper limit of the rapid degradation conditioning pressure range is the preset maximum rapid degradation conditioning pressure. Simultaneously, the control system sends a speed command to the stirring shaft drive motor in the conditioning chamber, controlling the stirring shaft to run at a first speed. The value of the first speed is calculated in advance based on the first gelatinization target value required for the rapidly degrading component to reach in the rapid hydrothermal modification range, combined with the effective volume of the conditioning chamber and the geometric parameters of the stirring blades. Under the combined action of the rapid degradation conditioning pressure zone and the first rotation speed, the current raw material component undergoes a rapid hydrothermal modification zone within the conditioner cavity. The duration of this rapid hydrothermal modification zone is determined by the interval between the feeding timestamp of the current raw material component and the feeding timestamp of the next record in the stratified feeding sequence table. After undergoing the rapid hydrothermal modification zone, the current raw material component is discharged from the conditioner outlet, and the discharged material has a degree of gelatinization close to the first target value. The specific value of the first target value of gelatinization is preset based on the target degradation rate of the rapidly degrading component in the rumen of a ruminant and the correspondence between the degree of starch gelatinization and the degradation rate.

[0050] In practice, when the control system detects that the degradation rate level of the current raw material component is level two, it sends a control command to the second steam regulating valve of the conditioning unit, controlling the second steam regulating valve to open to the second opening value. The second opening value is calculated based on the target pressure value of the medium-speed degradation conditioning pressure range and the main steam pressure value of the steam supply pipeline of the conditioning unit, using the flow characteristic curve of the second steam regulating valve. After the second steam regulating valve opens to the second opening value, the steam pressure in the conditioner is maintained within the medium-speed degradation conditioning pressure range. The lower limit of the medium-speed degradation conditioning pressure range is the preset minimum conditioning pressure for medium-speed degradation, and the upper limit is the preset maximum conditioning pressure for medium-speed degradation. The pressure value of the medium-speed degradation conditioning pressure range is lower than the pressure value of the rapid degradation conditioning pressure range. Simultaneously, the control system controls the stirring shaft in the conditioner to operate at a second rotation speed. The value of the second rotation speed is pre-calculated based on the target value of the second degree of gelatinization required for the medium-speed degradation component in the medium-speed hydrothermal modification range, combined with the effective volume of the conditioner cavity and the geometric parameters of the stirring blades. The second rotational speed is lower than the first rotational speed. Under the combined action of the medium-speed degradation conditioning pressure range and the second rotational speed, the current raw material component undergoes a medium-speed hydrothermal modification range within the conditioner cavity. The duration of this medium-speed hydrothermal modification range is determined by the interval between the feeding timestamp of the current raw material component and the feeding timestamp of the next record in the stratified feeding sequence table. After passing through the medium-speed hydrothermal modification range, the current raw material component is discharged from the conditioner outlet, and the discharged material has a degree of gelatinization close to the second gelatinization target value. The specific value of the second gelatinization target value is preset based on the target degradation rate of the medium-speed degradation component in the rumen of a ruminant and the correspondence between starch gelatinization and degradation rate. The second gelatinization target value is lower than the first gelatinization target value.

[0051] In practice, when the control system detects that the degradation rate level of the current raw material component is level three, it sends a control command to the third steam regulating valve of the conditioning unit, controlling the third steam regulating valve to open to the third opening value. The third opening value is calculated based on the target pressure value of the slow degradation conditioning pressure range and the main steam pressure value of the steam supply pipeline of the conditioning unit, using the flow characteristic curve of the third steam regulating valve. After the third steam regulating valve opens to the third opening value, the steam pressure in the conditioner is maintained within the slow degradation conditioning pressure range. The lower limit of the slow degradation conditioning pressure range is the preset minimum conditioning pressure for slow degradation, and the upper limit is the preset maximum conditioning pressure for slow degradation. The pressure value of the slow degradation conditioning pressure range is lower than the pressure value of the medium-speed degradation conditioning pressure range. Simultaneously, the control system controls the stirring shaft in the conditioner to operate at the third rotation speed. The value of the third rotation speed is pre-calculated based on the target value of the third degree of gelatinization required for the slow degradation component in the slow hydrothermal modification range, combined with the effective volume of the conditioner cavity and the geometric parameters of the stirring blades. The third rotational speed is lower than the second rotational speed. Under the combined effect of the slow degradation conditioning pressure zone and the third rotational speed, the current raw material component undergoes a slow hydrothermal modification zone within the conditioner cavity. The duration of this slow hydrothermal modification zone is determined by the interval between the feeding timestamp of the current raw material component and the feeding timestamp of the next record in the stratified feeding sequence table. After passing through the slow hydrothermal modification zone, the current raw material component is discharged from the conditioner outlet, and the discharged material has a degree of gelatinization close to the third gelatinization target value. The specific value of the third gelatinization target value is preset based on the target degradation rate of the slow-degrading component in the rumen of a ruminant and the correspondence between starch gelatinization and degradation rate. The third gelatinization target value is lower than the second gelatinization target value.

[0052] In practice, each time a batch of modified material is discharged from the conditioner outlet, a cumulative quantity detection device located below the conditioner outlet detects the weight of that batch of modified material and sends the weight signal to the control system. The control system calculates the cumulative quantity of all modified material discharged in the current production cycle in real time and compares the calculated cumulative quantity with a preset single-batch mixing quantity threshold. The value of the single-batch mixing quantity threshold is determined based on the rated mixing capacity of the mixer and is set as a preset proportional coefficient of the rated mixing capacity of the mixer. This preset proportional coefficient is determined based on the loading rate at which the mixer achieves optimal mixing uniformity under actual operating conditions.

[0053] In practice, the following formula can be used to determine whether the cumulative amount has reached the threshold for single-batch mixing:

[0054] in, Indicates the time point The cumulative amount of modified material discharged from the conditioner outlet; Indicates the time point The total number of batches of modified material discharged from the conditioner outlet; Indicates the first discharge from the conditioner outlet The weight of the batch of modified material. From 1 to The integer sequence number represents the raw material component discharged after hydrothermal modification treatment in each batch of modified materials.

[0055] In practice, when the control system detects that the accumulated amount of modified material has reached the single-batch mixing threshold, the control system sends a start command to the stirring blade drive motor of the mixer. The stirring blades begin to rotate, mechanically mixing all the modified material accumulated in the mixer. These modified materials include materials that have undergone rapid hydrothermal modification and have a gelatinization degree close to the first gelatinization degree target value, materials that have undergone medium-speed hydrothermal modification and have a gelatinization degree close to the second gelatinization degree target value, and materials that have undergone slow hydrothermal modification and have a gelatinization degree close to the third gelatinization degree target value. The stirring blades operate according to the preset mixing speed and mixing time, ensuring that the three types of modified materials with different gelatinization degree distributions are fully and uniformly mixed in the mixer, generating a mixture with a gradient gelatinization degree distribution. In the mixture, the material particles derived from the fast-degrading, medium-degrading, and slow-degrading components retain the gelatinization characteristics obtained after undergoing their respective hydrothermal modification ranges, forming a gradient distribution structure from low gelatinization to high gelatinization. This gradient gelatinization distribution structure enables the mixture to achieve the segmented release of nutrients with different degradation rates along the time axis in the rumen of ruminants.

[0056] The above description is merely a specific embodiment of this application, but the scope of protection of this application is not limited thereto. Any changes or substitutions that can be easily conceived by those skilled in the art within the scope of the technology disclosed in this application should be included within the scope of protection of this application.

Claims

1. A method for integrated nutritional regulation and continuous production of ruminant feed, characterized in that, The method includes: Perform cross-process lattice structure feature analysis on raw materials to be processed in the ruminant feed production chain to generate raw material nutritional feature vectors that include crude fiber crystallinity and protein molecule conformation. Based on the raw material nutrient feature vector and the preset rumen degradation rate benchmark library, multi-dimensional matching is performed to determine the degradation rate level and fermentation acid production potential value of each raw material component. Based on the degradation rate level and fermentation acid production potential of each raw material component, a stratified feeding sequence table containing fast-degrading components, medium-degrading components, and slow-degrading components is generated. According to the aforementioned stratified feeding sequence table, the steam pressure gradient in the conditioning device is controlled so that raw material components with different degradation rate levels sequentially undergo hydrothermal modification intervals of different intensities in the conditioner, generating a mixture with a gradient gelatinization degree distribution.

2. The integrated method for nutrient regulation and continuous production of ruminant feed according to claim 1, characterized in that, The process of performing cross-process lattice structure feature analysis on raw materials to be processed in the ruminant feed production chain generates a raw material nutritional feature vector containing crude fiber crystallinity and protein molecule conformation, including: After the raw material to be processed is sieved to remove large particulate impurities and before it enters the pulverizer, the diffuse reflectance spectrum of the raw material to be processed is collected at multiple points by a near-infrared spectral sensor array to obtain the absorbance value sequence at each point at the characteristic absorption wavelength. The crystallinity index of the crude fiber component in the raw material to be processed is calculated based on the ratio of the intensity of the characteristic absorption peak of cellulose to the intensity of the characteristic absorption peak of the amorphous region in the absorbance value sequence. The secondary structure order index of the protein component in the raw material to be processed is calculated based on the ratio of the absorption peak areas of amide I band to amide II band in the absorbance value sequence. The crystallinity index and the secondary structure order index are concatenated as vectors to generate the raw material nutritional feature vector.

3. The integrated method for nutrient regulation and continuous production of ruminant feed according to claim 2, characterized in that, The process of determining the degradation rate level and fermentation acid production potential of each raw material component by performing multi-dimensional matching based on the raw material nutrient feature vector and a preset rumen degradation rate benchmark library includes: The crystallinity index is extracted from the nutritional feature vector of the raw material as the first matching dimension. The Euclidean distance between the crystallinity index and each benchmark crystallinity interval in the rumen degradation rate benchmark library is calculated. The degradation rate level corresponding to the benchmark crystallinity interval with the smallest Euclidean distance is selected as the degradation rate level of the current raw material component. The secondary structure order index is extracted from the nutrient feature vector of the raw material as the second matching dimension. Multiple acid production potential templates that are related to the degradation rate level are retrieved from the rumen degradation rate benchmark library. The absolute value of the difference between the secondary structure order index and the preset order reference value in each acid production potential template is calculated. The fermentation acid production potential value corresponding to the acid production potential template with the smallest absolute value of the difference is selected as the fermentation acid production potential value of the current raw material component. The degradation rate level and the fermentation acid production potential value are associated and stored in the raw material component characteristic record table.

4. The integrated method for nutrient regulation and continuous production of ruminant feed according to claim 3, characterized in that, The process involves generating a stratified feeding sequence table, comprising fast-degrading, medium-degrading, and slow-degrading components, based on the degradation rate levels and fermentation acid production potential values ​​of each raw material component. Raw material components with a degradation rate level of 1 are labeled as fast-degrading components, raw material components with a degradation rate level of 2 are labeled as medium-degrading components, and raw material components with a degradation rate level of 3 are labeled as slow-degrading components. For each raw material component marked as a rapidly degradable component, they are arranged in descending order of their fermentation acid production potential value to generate a rapidly feeding subsequence, and a first time interval feeding timestamp is set for each rapidly degradable component in the rapidly feeding subsequence. For each raw material component marked as a medium-rate degradation component, they are arranged in descending order of their fermentation acid production potential value to generate a medium-rate feeding subsequence, and a second time interval feeding timestamp is set for each medium-rate degradation component in the medium-rate feeding subsequence. For each raw material component marked as a slow-degrading component, they are arranged in descending order of their fermentation acid production potential value to generate a slow-feeding subsequence, and a third time interval feeding timestamp is set for each slow-degrading component in the slow-feeding subsequence. The fast feeding subsequence, medium-speed feeding subsequence, and slow feeding subsequence are spliced ​​together in chronological order. A first delay time window is inserted between the fast feeding subsequence and the medium-speed feeding subsequence, and a second delay time window is inserted between the medium-speed feeding subsequence and the slow feeding subsequence, to generate the layered feeding sequence table.

5. The integrated method for nutrient regulation and continuous production of ruminant feed according to claim 4, characterized in that, The generation of the mixture with a gradient gelatinization degree distribution includes: Read the current raw material component and its degradation rate level corresponding to the current timestamp from the layered feeding time sequence table; If the degradation rate level of the current raw material component is level one, the first steam regulating valve in the conditioning device is opened to the first opening value to maintain the steam pressure in the conditioning unit within the rapid degradation conditioning pressure range. At the same time, the stirring shaft in the conditioning unit is controlled to run at the first speed so that the current raw material component is discharged after passing through the rapid hydrothermal modification range in the conditioning unit. The rapid hydrothermal modification range corresponds to the first gelatinization target value. If the degradation rate level of the current raw material component is level two, the second steam regulating valve in the conditioning device is opened to the second opening value to maintain the steam pressure in the conditioning unit within the medium-speed degradation conditioning pressure range. At the same time, the stirring shaft in the conditioning unit is controlled to run at the second rotation speed so that the current raw material component is discharged after passing through the medium-speed hydrothermal modification range in the conditioning unit. The medium-speed hydrothermal modification range corresponds to the second gelatinization target value. If the degradation rate level of the current raw material component is level three, then the third steam regulating valve in the conditioning device is opened to the third opening value, so that the steam pressure in the conditioner is maintained in the slow degradation conditioning pressure range. At the same time, the stirring shaft in the conditioner is controlled to run at the third speed, so that the current raw material component is discharged after passing through the slow hydrothermal modification range in the conditioner. The slow hydrothermal modification range corresponds to the third gelatinization target value. When the cumulative amount of modified material discharged from the conditioner outlet is detected to reach the single batch mixing threshold, the mixing blades of the mixer are started to mix the discharged modified materials with different gelatinization degree distributions to generate the mixture.

6. The integrated method for nutrient regulation and continuous production of ruminant feed according to claim 5, characterized in that, The process of acquiring multi-point diffuse reflectance spectra of the raw materials to be processed using a near-infrared spectral sensor array after the primary cleaning screen removes large particulate impurities and before the raw materials enter the pulverizer includes: Multiple spectral acquisition points are set at equal intervals on the conveyor belt between the discharge port of the primary screen and the feed port of the crusher; When the raw material to be processed moves to each spectral acquisition point along the conveyor belt, the near-infrared spectral sensor at the corresponding point is triggered to start scanning. Each sensor acquires at least three diffuse reflectance spectral data. The average value of the three diffuse reflectance spectral data acquired at the same acquisition point is taken according to the wavelength and used as the average diffuse reflectance spectrum of that acquisition point. The average diffuse reflectance spectra of all the sampling points are arranged in order of location to generate a multi-point diffuse reflectance spectrum sequence.

7. The integrated method for nutrient regulation and continuous production of ruminant feed according to claim 6, characterized in that, The step of calculating the crystallinity index of the crude fiber component in the raw material to be processed based on the ratio of the intensity of the characteristic absorption peak of cellulose to the intensity of the characteristic absorption peak of the amorphous region in the absorbance value sequence includes: The first absorbance value at the characteristic wavelength of the cellulose crystallization region of each sampling point is extracted from the multi-point diffuse reflectance spectral sequence, and the arithmetic mean of the first absorbance values ​​of each sampling point is calculated to obtain the global average absorbance value of the crystallization region. The second absorbance value at the characteristic wavelength of the amorphous region of each sampling point is extracted from the multi-point diffuse reflectance spectral sequence, and the second absorbance value of each sampling point is arithmetically averaged to obtain the global amorphous region absorbance mean. Calculate the ratio of the average absorbance of the global crystalline region to the average absorbance of the global amorphous region, and use the ratio as the crystallinity index of the crude fiber component in the raw material to be processed.

8. The integrated method for nutrient regulation and continuous production of ruminant feed according to claim 7, characterized in that, The step of calculating the secondary structure order index of the protein component in the raw material to be processed based on the ratio of the absorption peak areas of amide I band to amide II band in the absorbance value sequence includes: The first integrated absorbance value of each acquisition point within the characteristic absorption range of the amide I band is extracted from the multi-point diffuse reflectance spectral sequence, and the first integrated absorbance value of each acquisition point is arithmetically averaged to obtain the global amide I band integrated mean. The second integrated absorbance value of each acquisition point within the characteristic absorption range of the amide II band is extracted from the multi-point diffuse reflectance spectral sequence, and the second integrated absorbance value of each acquisition point is arithmetically averaged to obtain the global amide II band integrated mean. The ratio of the global amide I band integral mean to the global amide II band integral mean is calculated, and the ratio is used as the secondary structure order index of the protein component in the raw material to be processed.

9. The integrated method for nutrient regulation and continuous production of ruminant feed according to claim 3, characterized in that, The rumen degradation rate benchmark library contains multiple benchmark crystallinity ranges and degradation rate levels that correspond one-to-one with each benchmark crystallinity range. Each degradation rate level is associated with multiple acid production potential templates. Each acid production potential template contains a preset ordering reference value and a corresponding fermentation acid production potential value. The rumen degradation rate benchmark library is updated regularly based on historical production batches of rumen degradation test data of ruminants.

10. The integrated method for nutrient regulation and continuous production of ruminant feed according to claim 4, characterized in that, The first time interval feeding timestamp, the second time interval feeding timestamp, and the third time interval feeding timestamp are dynamically determined based on the difference in fermentation acid production potential values ​​of adjacent raw material components in each feeding subsequence. The larger the difference in fermentation acid production potential values, the shorter the corresponding feeding time interval.

Images