A method for controlling the mixing of a feed for aquaculture and a mixing device

Abstract

The present application relates to the technical field of intelligent control, and particularly relates to a breeding feed mixing control method and a mixing device, which solves the technical problem of poor mixing quality caused by the difference in the physical properties of raw materials in the prior art. The method comprises the following steps: monitoring the component composition and the particle physical form of the mixed material in the mixing process, and the stirring parameters of the mixing device; the mixing process comprises the following steps: stirring the dry materials in the mixing device, and adding liquid by spraying; the component composition is obtained by near-infrared spectrum analysis; the particle physical form is obtained by image analysis; the influence of the stirring parameters on the mixing quality is analyzed based on the component composition, and the correction coefficient of the operating parameters of the mixing device is determined in combination with the particle physical form; and the current operating parameters of the mixing device are regulated according to the correction coefficient, so that the mixing device operates at the regulated target operating parameters.

Description

TECHNICAL FIELD

[0001] The present application relates to the technical field of intelligent control, and particularly relates to a breeding feed mixing control method and a mixing device. BACKGROUND

[0002] With the rapid development of the breeding industry, the influence of the quality and formula of feed on the breeding effect is increasingly prominent. Reasonable feed ratio can not only improve the feed conversion rate, but also effectively improve the health condition and growth rate of breeding animals, thereby improving the economic benefits of breeding. The market demand for efficient and high-quality feed is also increasing. To meet this demand, many breeding farms have begun to use professional feed mixing equipment to ensure the uniformity of feed mixing and the reasonable allocation of nutritional components. At the same time, in the process of feed mixing in the modern breeding industry, the mixing time, temperature, humidity and other key factors need to be controlled to ensure that the feed reaches the best use effect.

[0003] Although the existing feed mixing control method and device can achieve the configuration and mixing of feed to some extent, due to the significant differences in physical properties such as particle size, density and humidity of feed components, the existing device is difficult to achieve ideal mixing uniformity in actual operation, resulting in unstable feed quality. SUMMARY

[0004] In order to solve the technical problem of poor mixing quality caused by the difference in physical properties of raw materials in the prior art, the purpose of the present application is to provide a breeding feed mixing control method and a mixing device, and the technical scheme adopted is as follows:

[0005] In a first aspect, a breeding feed mixing control method is provided, comprising: monitoring the component composition and particle physical form of the mixed material in the mixing process, and the stirring parameters of the mixing device; the mixing process comprises: stirring the dry materials in the mixing device, and adding liquid by spraying; the component composition is obtained by near-infrared spectrum analysis; the particle physical form is obtained by image analysis; the influence of the stirring parameters on the mixing quality is analyzed based on the component composition, and the correction coefficient of the operating parameters of the mixing device is determined in combination with the particle physical form; the current operating parameters of the mixing device are regulated according to the correction coefficient, so that the mixing device operates at the regulated target operating parameters.

[0006] Based on the above technical solution, in the livestock feed mixing control method provided by this invention, the composition of the mixed feed is monitored in real time by near-infrared spectroscopy analysis, and the physical morphology of feed particles is obtained by image analysis technology, thereby constructing a dual evaluation system based on the uniformity of component distribution and the uniformity of particle morphology. Furthermore, by analyzing the influence of stirring parameters on the above mixing quality indicators, the correction coefficient of the operating parameters is intelligently determined, realizing dynamic parameter control, improving the uniformity and stability of feed mixing, and effectively solving the problem of poor mixing quality easily caused by differences in raw material properties in existing technologies.

[0007] In addition, it significantly reduces the need for manual intervention in the production process, achieving a high degree of automation and intelligence in the hybrid process.

[0008] In conjunction with the first aspect above, in one possible implementation, the method for analyzing the influence of stirring parameters on mixing quality based on component composition specifically includes: fitting the actual content of each component at multiple times to obtain a content change curve of the actual content of each component over time; comparing the actual content of each component with the expected content, and determining the mixing quality deviation degree to characterize the uniformity of component distribution in the mixture based on the fluctuation of the content change curve; and analyzing the influence of stirring parameters on mixing quality based on the mixing quality deviation degree and the changes in stirring parameters.

[0009] In conjunction with the first aspect above, in one possible implementation, the method for analyzing the influence of stirring parameters on mixing quality based on the mixing quality deviation and the changes in stirring parameters specifically includes: obtaining parameter change curves of the stirring parameters of the mixing equipment over time during the mixing process; comparing the parameter change curves with the content change curves to analyze the correlation between stirring parameters and mixing quality; and determining the degree of influence of stirring parameters on mixing quality based on the correlation and the mixing quality deviation.

[0010] In conjunction with the first aspect above, in one possible implementation, the method for determining the correction coefficient of the mixing equipment operating parameters based on the particle physical morphology specifically includes: determining the degree of reverse influence on the mixing quality from two dimensions, namely, the stirring parameters and the particle physical morphology, based on the degree of influence of the stirring parameters on the mixing quality and the particle physical morphology; and normalizing the degree of reverse influence to obtain the correction coefficient.

[0011] In conjunction with the first aspect above, in one possible implementation, the method for regulating the current operating parameters of the mixing equipment based on the correction coefficient specifically includes: comparing the correction coefficient with a preset threshold, and determining a regulation strategy based on the comparison result; the regulation strategy includes: increasing the stirring speed of the mixing equipment, or increasing at least one of the spraying speed and the jet pressure.

[0012] In conjunction with the first aspect mentioned above, in one possible implementation, the method for obtaining the composition of components through near-infrared spectroscopy analysis specifically includes: acquiring spectral data of the mixture using a near-infrared spectrometer inside the mixing device; and analyzing the spectral data of the mixture using a preset calibration model to obtain the actual content of multiple components in the mixture.

[0013] In conjunction with the first aspect mentioned above, in one possible implementation, the method for obtaining the physical morphology of particles through image analysis specifically includes: acquiring images of the mixed materials at multiple times in the mixing equipment; performing grayscale conversion and edge detection processing on the mixed material images to identify multiple closed regions in the mixed material images; and analyzing the uniformity of the mixed material particles during the mixing process based on the morphological characteristics of the multiple closed regions.

[0014] Secondly, a livestock feed mixing device is provided, comprising: a mixing container, a stirring module, a liquid input module, a data monitoring module, and an intelligent control module; the mixing container is configured to hold the mixed materials; the stirring module is configured to stir the mixed materials; the liquid input module is configured to add liquid into the mixing container by spraying; the data monitoring module includes: a near-infrared spectrometer, a camera, and a stirring parameter sensor; the near-infrared spectrometer is configured to collect spectral data of the mixed materials; the camera is configured to collect images of the mixed materials in the mixing device; the stirring parameter sensor is configured to collect the stirring parameters of the stirring module; the intelligent control module is configured to: receive the spectral data, mixed material images, and stirring parameters from the data monitoring module; obtain the composition of the mixed materials through the spectral data; obtain the particle physical morphology of the mixed materials through the mixed material images; analyze the influence of stirring parameters on the mixing quality based on the composition; and determine the correction coefficient of the operating parameters of the mixing device based on the particle physical morphology; generate control commands according to the correction coefficients to adjust the current operating parameters of the stirring module and the liquid input module; the stirring module and the liquid input module respond to the control commands and operate based on the adjusted target operating parameters.

[0015] In conjunction with the second aspect above, in one possible implementation, the liquid injection module specifically includes an inlet pipe and nozzles deployed at multiple locations on the inner wall of the mixing container; the nozzles inject liquid into the mixing container by spraying from multiple locations.

[0016] In conjunction with the second aspect above, in one possible implementation, the above-mentioned livestock feed mixing equipment further includes: an air jet module; the air jet module includes an air inlet pipe and air jet nozzles deployed at multiple locations on the inner wall of the mixing container; the air jet nozzles are used to spray gas to assist in stirring and prevent the mixed materials from sticking to the inner wall of the equipment.

[0017] Thirdly, a livestock feed mixing control device is provided, comprising: a processor and a storage medium; the storage medium includes instructions, and the processor is configured to execute the instructions to perform the actions described in the first aspect and any possible implementation thereof. This livestock feed mixing control device may be an electronic device or a chip within an electronic device.

[0018] Fourthly, a computer-readable storage medium is provided, which stores instructions that, when executed on a livestock feed mixing control device, cause the livestock feed mixing control device to perform actions as described in the first aspect and any possible implementation thereof.

[0019] Fifthly, a computer program product containing instructions is provided that, when the computer program product is run on a livestock feed mixing control device, causes the livestock feed mixing control device to perform the actions described in the first aspect and any possible implementation thereof.

[0020] The present invention has the following beneficial effects:

[0021] By using near-infrared spectroscopy to monitor the composition of mixed feed in real time and combining it with image analysis technology to obtain the physical morphology of feed particles, a dual evaluation system based on the uniformity of component distribution and particle morphology was constructed. Furthermore, by analyzing the influence of mixing parameters on the aforementioned mixing quality indicators, correction coefficients for operating parameters were intelligently determined, enabling dynamic parameter control. This improved the uniformity and stability of feed mixing and effectively solved the problem of poor mixing quality caused by differences in raw material properties in existing technologies.

[0022] In addition, it significantly reduces the need for manual intervention in the production process, achieving a high degree of automation and intelligence in the hybrid process. Attached Figure Description

[0023] To more clearly illustrate the technical solutions and advantages in the embodiments of the present invention or the prior art, the drawings used in the description of the embodiments or the prior art will be briefly introduced below. Obviously, the drawings described below are only some embodiments of the present invention. For those skilled in the art, other drawings can be obtained based on these drawings without creative effort.

[0024] Figure 1 This is a structural diagram of a livestock feed mixing device according to an embodiment of the present invention;

[0025] Figure 2 A structural diagram of another livestock feed mixing device provided in one embodiment of the present invention;

[0026] Figure 3This is a structural diagram of a liquid dispensing module provided in one embodiment of the present invention;

[0027] Figure 4 A flowchart illustrating a method for controlling feed mixing in livestock, as provided in one embodiment of the present invention;

[0028] Figure 5 A flowchart illustrating another method for controlling feed mixing in livestock, provided in one embodiment of the present invention;

[0029] Figure 6 An example graph of a protein content change curve provided in one embodiment of the present invention;

[0030] Figure 7 An example diagram of a velocity variation curve provided in one embodiment of the present invention;

[0031] Figure 8 A flowchart illustrating another method for controlling feed mixing in livestock, provided in one embodiment of the present invention;

[0032] Figure 9 This is a schematic diagram of the hardware structure of a livestock feed mixing control device according to an embodiment of the present invention. Detailed Implementation

[0033] To further illustrate the technical means and effects adopted by the present invention to achieve its intended purpose, the following, in conjunction with the accompanying drawings and preferred embodiments, details the specific implementation, structure, features, and effects of a livestock feed mixing control method and mixing equipment proposed according to the present invention. In the following description, different "one embodiment" or "another embodiment" do not necessarily refer to the same embodiment. Furthermore, specific features, structures, or characteristics in one or more embodiments can be combined in any suitable form.

[0034] Unless otherwise defined, all technical and scientific terms used herein have the same meaning as commonly understood by one of ordinary skill in the art to which this invention pertains.

[0035] The following description, in conjunction with the accompanying drawings, details a specific scheme for a method and equipment for controlling the mixing of livestock feed provided by the present invention.

[0036] Please see Figure 1 and Figure 2 The diagram shows a structural diagram of a livestock feed mixing device according to an embodiment of the present invention. The livestock feed mixing device includes: a mixing container 1, a stirring module 2, a liquid input module 3, a data monitoring module 4, and an intelligent control module 5.

[0037] The mixing container 1 is configured to hold the mixed materials. In practical applications, to better concentrate the materials in one place for mixing, the mixing container 1 is usually a conical container.

[0038] Mixing module 2 is configured to mix the materials.

[0039] In some implementations, such as Figure 2 As shown, the mixing module 2 may include a mixing roller 21 and a mixing auxiliary device 22. The mixing auxiliary device 22 is used to rotate with the mixing roller 21 and to assist the mixing roller 21 in further mixing the materials during the mixing process.

[0040] Liquid input module 3 is configured to add liquid into mixing container 1 by spraying.

[0041] For example, in the scenario of mixing livestock feed, nutrient solution needs to be added during the mixing process. However, adding the nutrient solution directly to the mixing equipment may cause the feed to clump during the mixing process. By spraying the nutrient solution, the possibility of feed clumping and uneven distribution during the mixing process can be reduced.

[0042] The data monitoring module 4 includes: a near-infrared spectrometer 41, a camera 42, and a stirring parameter sensor 43.

[0043] Near-infrared spectrometer 41, configured to collect spectral data of mixed materials.

[0044] Camera 42 is configured to acquire images of the mixture in the mixing equipment.

[0045] The stirring parameter sensor 43 is configured to collect the stirring parameters of the stirring module 2.

[0046] The intelligent control module 5 is configured to: receive spectral data, mixed material images, and stirring parameters from the data monitoring module 4; obtain the composition of the mixed material through the spectral data; obtain the particle physical morphology of the mixed material through the mixed material images; analyze the influence of stirring parameters on mixing quality based on the composition; and determine the correction coefficient of the operating parameters of the mixing equipment in combination with the particle physical morphology; and generate control commands based on the correction coefficient to regulate the current operating parameters of the stirring module 2 and the liquid input module 3.

[0047] The stirring module 2 and the liquid feeding module 3 respond to control commands and operate based on the adjusted target operating parameters.

[0048] In some implementations, such as Figure 1 and Figure 3As shown, the liquid injection module 3 includes an inlet pipe 31 and nozzles 32 deployed at multiple locations on the inner wall of the mixing container 1. The nozzles 32 inject liquid into the mixing container 1 by spraying from multiple locations.

[0049] Spraying from multiple locations increases the contact area between the liquid and the dry material, allowing the liquid to evenly cover the dry material area within the mixing container 1, thus avoiding localized liquid concentration and clumping that can easily occur with single-location spraying.

[0050] In some implementations, such as Figure 1 and Figure 2 As shown, the livestock feed mixing equipment also includes an air jet module 6. The air jet module 6 includes an air inlet pipe 61 and air jet nozzles 62 deployed at multiple locations on the inner wall of the mixing container 1. The air jet nozzles 62 are used to spray gas to assist in mixing and prevent the mixed materials from sticking to the inner wall of the equipment.

[0051] The gas ejected by the jet module 6 can break the local circulation dead zone formed by the stirring of materials, make up for the limitations of mechanical stirring in material dispersion, further promote the uniform mixing of materials with different physical properties, and alleviate the problem of poor mixing quality caused by raw material stratification and agglomeration.

[0052] In addition, such as Figure 1 and Figure 2 As shown, in actual production, this livestock feed mixing equipment also includes: a conveyor belt 7, an observation window 8, a discharge port 9, and support legs 10.

[0053] After one mixing cycle is completed, the mixture in the mixing container 1 is discharged from the mixing equipment via the conveyor belt 7 and the discharge port 9 for storage.

[0054] Please see Figure 4 The diagram illustrates a flowchart of a method for controlling feed mixing in livestock according to an embodiment of the present invention. This method includes:

[0055] S1. Monitor the composition and particle physical morphology of the mixed materials during the mixing process, as well as the stirring parameters of the mixing equipment.

[0056] The mixing process includes: stirring the dry materials in the mixing equipment and adding liquid by spraying. Specifically, the dry materials to be mixed (such as soybean meal, corn flour, etc.) are first put into the conical mixing container. The stirring roller of the stirring module is started, and the stirring roller drives the stirring auxiliary device to rotate synchronously, mechanically stirring the dry materials to achieve initial dispersion. At the same time, the liquid addition module sprays liquid (such as vitamin solution, mineral nutrient solution) into the dry materials in a mist form through multiple nutrient solution nozzles deployed around the inner wall of the conical mixing container. The pressure of the mist spray can be controlled at 0.3-0.5 MPa to ensure that the liquid is evenly dispersed in the dry materials and avoid local agglomeration.

[0057] Specifically, the composition is obtained through near-infrared spectroscopy analysis, including: collecting spectral data of the mixture using a near-infrared spectrometer inside the mixing equipment; and analyzing the spectral data of the mixture using a preset calibration model to obtain the actual content of multiple components in the mixture.

[0058] Near-infrared spectroscopy can rapidly and non-destructively analyze the actual content of multiple components such as protein, crude fiber, and moisture in a mixture. This process can capture in real time the uneven distribution of components caused by differences in raw material properties (such as particle size and density), while also automating and increasing the efficiency of component monitoring, reducing the time cost and errors of manual sampling and analysis.

[0059] In some implementations, a near-infrared spectrometer is installed at the observation window on the side wall of a conical mixing container. The spectrometer's probe is positioned directly over the mixing area of ​​the mixture, and spectral data (spectral range 700-1100 nm) of the mixture is collected at regular intervals (e.g., every 10 seconds) during the mixing process. The collected spectral data is then input into a pre-set calibration model (trained using a partial least squares regression algorithm with standard feed samples of known component contents). The model calculates the actual content of multiple components (such as protein, crude fiber, and moisture) in the mixture.

[0060] In other implementations, in addition to near-infrared spectroscopy analysis, molecular vibrational spectral data of the mixture can be collected, and the composition can be analyzed by the intensity of characteristic peaks; or a combination of a near-infrared spectrometer and a fiber optic probe can be used, with the probe directly inserted into the mixture to collect spectral data, which is suitable for high-humidity feed mixing scenarios.

[0061] Specifically, the physical morphology of the particles is obtained through image analysis, including: acquiring images of the mixed materials at multiple times in the mixing equipment; performing grayscale and edge detection processing on the mixed material images to identify multiple closed regions in the mixed material images; and analyzing the uniformity of the mixed material particles during the mixing process based on the morphological characteristics of the multiple closed regions.

[0062] In some implementations, a high-resolution industrial camera (with a resolution of no less than 20 megapixels) is mounted on top of a conical mixing container. The camera lens is positioned at a 45° angle to the mixing area of ​​the materials. During the mixing process, the camera simultaneously acquires spectral data and captures an image of the mixture at regular intervals (e.g., every 10 seconds). The acquired images are first converted to grayscale (this simplifies data dimensions and eliminates irrelevant interference), and then edge detection algorithms are used to extract image edges, identifying multiple closed regions in the image. Each closed region corresponds to a feed particle or particle agglomeration. Finally, based on the morphological characteristics of the closed regions (such as area, perimeter, and roundness), the uniformity of the mixed material particles is calculated.

[0063] This process can capture problems such as agglomeration and uneven particle distribution caused by differences in raw material properties (such as particle size and adsorption) in real time. At the same time, it realizes the automation and refinement of particle morphology monitoring, reducing the error and lag of manual judgment.

[0064] In some implementations, the uniformity of the mixed material particles The calculation formula is:

[0065]

[0066] In the formula, The standard deviation of the area of ​​multiple closed regions in the image of the mixture can reflect the degree of difference in particle size.

[0067] This represents the number of multiple closed regions in the image of the mixture.

[0068] Let m be the area of ​​the m-th closed region in the image of the mixture.

[0069] Used to quantify the average absolute size of particles in an image of a mixture. The larger the average area, the larger the overall particle size (or agglomeration size).

[0070] Combining relative fluctuations with absolute size comprehensively reflects the degree of absolute difference in particle size in the j-th mixture image. Since closed regions necessarily have an area, and only when all particle areas are completely identical... However, the differences in material properties (such as particle size and density) mean that such an extreme case is almost impossible in reality. Therefore, this value must be greater than zero, and the larger the value, the more significant the absolute difference in particle size and the worse the uniformity.

[0071] This is to monitor the number of images of the mixed materials collected within a given time period.

[0072] By averaging the absolute differences among T images of the mixture, the randomness of a single frame image is eliminated, and the absolute differences in particle size over the entire monitoring period are integrated.

[0073] By taking the reciprocal, the inverse relationship between particle size variation and uniformity is transformed into a positive relationship. The smaller the value, the worse the uniformity of the particles during the mixing process; conversely, the larger the value, the better the uniformity of the particles during the mixing process, meaning that the possibility of feed clumping during the mixing process is smaller.

[0074] In other implementations, in addition to single-camera image analysis, binocular vision cameras can be used to acquire stereo images and obtain physical morphological parameters such as particle volume and spatial distribution through three-dimensional reconstruction; or high-speed cameras (frame rate ≥ 1000fps) can be used to capture dynamic stirring processes and analyze particle motion trajectories to assist in evaluating physical morphology.

[0075] In some implementations, the mixing parameters may include any one or more of the following: mixing speed (i.e., the rotational speed of the mixing roller), mixing torque, spraying speed, and jet pressure.

[0076] S2. Based on the analysis of the composition, determine the influence of stirring parameters on the mixing quality, and combine the particle physical morphology to determine the correction coefficients of the operating parameters of the mixing equipment.

[0077] In some implementations, the method for analyzing the influence of stirring parameters can be: using a machine learning model (such as random forest) for analysis, taking the stirring parameter sequence (rotation speed, duration) as input, and the mixing quality deviation, which characterizes the uniformity of component distribution, as output, and directly quantifying the influence weight of each stirring parameter on the mixing quality through a preset feature importance score. Specific implementation methods may include:

[0078] First, a high-quality training dataset is constructed, collecting the mixing parameter sequences (including mixing speed, mixing time, and mixing torque, collected at 10-second intervals), mixing quality deviation (calculated based on component content analysis from near-infrared spectroscopy and particle uniformity analysis from image analysis, with a value range of 0-1), and initial physical properties of raw materials (such as average dry particle size and initial moisture content) during the feed mixing process. The training dataset should cover at least typical raw material batches such as soybean meal, corn flour, and high-moisture bran. Multiple sets (e.g., 300 sets) of time-series data are collected for each batch. After outliers are removed using the 3σ criterion and missing values ​​are filled in using linear interpolation, the mixing parameter sequences and mixing quality deviation are strictly aligned according to the collection timestamp to form multiple sets of valid samples (the number of valid samples is equal to the product of the number of raw material batches and the number of time-series data for each batch).

[0079] Secondly, feature engineering is carried out to construct basic and interactive features from the time-series data of mixing parameters. Specifically, statistical features such as the mean rotational speed, variance of rotational speed, mean torque, coefficient of variation of torque, and cumulative mixing time within a 50-second sliding window are calculated for mixing parameters collected every 10 seconds. Simultaneously, interactive features such as the product of rotational speed and torque (representing the total mixing energy) and the product of rotational speed and spraying time (representing the mixing intensity during the liquid mixing stage) are constructed to capture the synergistic effects of multiple parameters. Further, differential features (such as the change in rotational speed between adjacent windows and the mean absolute value of torque difference) and time-series trend features (such as the slope of rotational speed trend) are extracted to characterize the dynamic changes of parameters. Subsequently, combining domain knowledge and algorithmic screening, low-variability features (such as features with variance less than 0.05) are first eliminated, then features strongly correlated with mixing quality deviation (such as features with mutual information greater than 0.3) are retained through mutual information screening, and finally, deduplication is performed through correlation (such as retaining features with higher mutual information if the correlation between two features is greater than 0.8), ultimately obtaining the core input features.

[0080] To further construct and train a random forest regression model, you can select a random forest regression model with 100 decision trees (balancing model accuracy and training efficiency), a tree depth of 8 (which can be adjusted based on the number of core input features to avoid overfitting), a node splitting criterion of mean squared error (to adapt to the continuous value output requirements of mixed quality deviation), and a feature sampling method of "sqrt" (to enhance tree diversity).

[0081] The samples were stratified into training, validation, and test sets using a 7:2:1 stratification sampling method. During training, bootstrap sampling was used to generate a bootstrap sample set, with each sample set corresponding to one decision tree. Each decision tree splits its nodes based on multiple randomly selected features with the goal of minimizing the mean squared error. Meanwhile, out-of-bag error was used to monitor the training process to avoid overfitting.

[0082] After the model is trained, its fitting effect and generalization ability are evaluated by the mean squared error and coefficient of determination (R²) of the training set, validation set, and test set. For example, for a new batch of raw materials used in the training, the mean squared error of the test set is less than 0.03 and the coefficient of determination is greater than 0.88, which verifies the model's generalization ability.

[0083] Finally, based on the feature importance scoring mechanism of random forest, the contribution of each feature to the reduction of mean squared error during the split of each decision tree is summarized and normalized to obtain the influence weight of each mixing parameter. For example, in the batch of chick feed, the importance of features such as rotation speed variance, mean torque, and the product of rotation speed and torque accounts for 35%, 25%, and 20%, respectively, thereby quantifying the degree of influence of each mixing parameter on the mixing quality.

[0084] In some implementations, the correction coefficient can be calculated by weighting the influence of stirring parameters on the mixing quality and the uniformity of the mixed materials. The priority difference between stirring effect and particle state is reflected by preset weights, and then the correction coefficient (with a value range of 0-1) is obtained by normalization.

[0085] S3. Adjust the current operating parameters of the mixing equipment according to the correction coefficient so that the mixing equipment operates at the adjusted target operating parameters.

[0086] In some implementations, the correction coefficient can be compared with a preset threshold, and the control strategy can be determined based on the comparison result. Control strategies include: increasing the mixing speed of the mixing equipment, or increasing at least one of the spraying speed and jet pressure.

[0087] Specifically, preset thresholds are set in advance based on the physical properties of the raw materials (such as particle size and moisture content) to avoid the problem of insufficient adaptability of fixed thresholds to raw materials with different physical properties. For example, for soybean meal and corn flour mixtures, the preset threshold is set to 0.5; for high-moisture bran, the preset threshold is set to 0.6. The higher the threshold, the stricter the requirements for mixing quality.

[0088] The correction coefficient is compared with the preset threshold in real time to determine the direction of regulation.

[0089] If the correction coefficient is greater than or equal to the preset threshold, it is determined that the current mixing quality deviation is large, such as uneven component distribution or obvious particle agglomeration, and the stirring intensity should be increased first. The intelligent control module sends a control command to the stirring module to increase the current speed of the stirring roller (e.g., 80 revolutions per minute, RPM) to the target speed (e.g., 120 RPM), and can also extend the stirring time (e.g., from 5 minutes to 8 minutes) to alleviate the problem of stratification or particle agglomeration caused by the difference in raw material density by enhancing the mechanical dispersion effect.

[0090] In some implementations, when the correction coefficient is much greater than the preset threshold (e.g., >0.8), multi-parameter linkage control can be used to send instructions to the stirring module, liquid injection module, and jet module at the same time, such as increasing the stirring speed by 20 RPM, increasing the spraying speed by 5 mL / s, and increasing the jet pressure by 0.05 MPa. Through the coordinated action of multiple modules, the mixing effect is enhanced and the mixing time to meet the standard is shortened.

[0091] If the correction coefficient is less than or equal to the preset threshold, the current mixing quality is considered to be basically up to standard, and the liquid mixing and anti-adhesion effects can be further optimized. The intelligent control module sends a command to the liquid dispensing module to increase the current spraying speed of the nozzle (e.g., 10 mL / s) to the target spraying speed (e.g., 15 mL / s), thereby enhancing the atomization and dispersion of the nutrient solution and preventing localized liquid accumulation and clumping; at the same time, it sends a command to the jet module to adjust the current air pressure of the jet nozzle (e.g., 0.2 MPa) to the target air pressure (e.g., 0.3 MPa), thereby strengthening the air film protection effect and preventing materials from adhering to the inner wall of the mixing container.

[0092] In some implementations, the target runtime parameters The calculation formula can be:

[0093]

[0094] In the formula, These are the current operating parameters of the mixing equipment. The smaller the correction coefficient, the smaller the adjustment increment of the target operating parameter.

[0095] In other implementations, a proportional-integral-differential (PID) control algorithm can be used. The difference between the correction coefficient and the preset threshold is taken as input, and the PID calculation automatically outputs a precise parameter adjustment amount. For example, when the difference is 0.1, the stirring speed is increased by 15 RPM and the spraying speed is increased by 3 mL / s, so as to achieve continuous and smooth parameter control and reduce the fluctuation of step control.

[0096] Furthermore, after the control command is sent, the intelligent control module continuously receives feedback from the data monitoring module on the composition, particle physical morphology, and stirring parameters, and verifies in real time whether the mixing quality has improved. If the mixing quality deviation does not decrease by 5% within 30 seconds, the target operating parameters are fine-tuned again (such as increasing the stirring speed by 10 RPM) until the mixing equipment is running stably with the target operating parameters, ensuring that the mixed materials reach the preset uniformity standard.

[0097] Based on the above technical solution, near-infrared spectroscopy is used to monitor the composition of mixed feed in real time, and image analysis technology is combined to obtain the physical morphology of feed particles, thus constructing a dual evaluation system based on the uniformity of component distribution and particle morphology. Furthermore, by analyzing the influence of stirring parameters on the above mixing quality indicators, the correction coefficients of operating parameters are intelligently determined, enabling dynamic parameter control. This improves the uniformity and stability of feed mixing and effectively solves the problem of poor mixing quality caused by differences in raw material properties in existing technologies.

[0098] In addition, it significantly reduces the need for manual intervention in the production process, achieving a high degree of automation and intelligence in the hybrid process.

[0099] In one possible implementation, combining Figure 4 ,like Figure 5 As shown, the method for analyzing the influence of stirring parameters on mixing quality based on component composition analysis in S2 above can be specifically implemented through the following S21 to S23, which are explained in detail below:

[0100] S21. Fit the actual content of each component at multiple times to obtain the content change curve of the actual content of each component over time.

[0101] In some implementations, the actual content data of multiple target components (such as protein, crude fiber, moisture, calcium, etc., the specific components being determined according to the feed formulation) in the mixed feed are retrieved at time intervals (e.g., once every 10 seconds, for a total of 30 sets, covering the entire stage from initial mixing of dry feed to stable mixing after liquid spraying). Each set of data includes the collection time and the actual content value of the corresponding component.

[0102] Each set of component data is preprocessed to remove outliers. For example, sudden changes in content values ​​caused by brief interference from the spectrometer, such as a protein content deviation of more than 5% from adjacent time points, are identified as outliers and replaced with the average of the two adjacent sets of data to ensure data validity.

[0103] The preprocessed component data are sorted chronologically to obtain the original data sequence. A fitting algorithm is then used to fit the original data sequence of each component individually, resulting in a unique fitting curve for each component. This curve can intuitively reflect the dynamic changes in the uniformity of the corresponding component's distribution during the mixing process. For example, the smoother the curve fluctuations, the more uniform the component distribution tends to be.

[0104] For example, Figure 6 The curve showing the change in the actual content (in %) of a protein over time (in min) is shown.

[0105] S22. Compare the actual content of each component with the expected content, and combine the fluctuation of the content change curve to determine the mixing quality deviation used to characterize the uniformity of component distribution in the mixture.

[0106] In some implementations, the expected deviation of the actual content of each component from the expected content can be calculated first, and then the mixing quality deviation of the entire mixing process can be analyzed based on the expected deviation of all components.

[0107] In some implementations, components Expected deviation The calculation formula can be:

[0108]

[0109] In the formula, For ingredients The variance of the content change curve is used to quantify the degree of fluctuation in the content of a single component over time. The larger the value, the more likely it is to contain a large amount of components. The more drastic the content fluctuations during the mixing process, the worse the mixing uniformity.

[0110] For ingredients The expected content.

[0111] For ingredients The average actual content during the monitoring period.

[0112] The combined effect of fluctuation and deviation from expected content was quantified. A larger product indicates greater fluctuation in actual content and that the monitored actual content is lower than the expected content, i.e., the composition during the mixing process is more volatile. Expected deviation Larger.

[0113] Furthermore, the mixed quality deviation The calculation formula can be:

[0114]

[0115] In the formula, This represents the quantity of all components in the mixture.

[0116] By averaging the deviations of all components, the influence of differences in component quantity on the results is eliminated, yielding a deviation that reflects the overall mixing quality. The larger the value, the greater the overall expected deviation of all components during the monitoring period, indicating poor mixing quality during that monitoring period, i.e., a large deviation in mixing quality.

[0117] In other implementations, the ratio of the range to the mean can be used instead of the standard deviation to calculate the fluctuation coefficient, avoiding the problem of the standard deviation being disturbed by extreme values. This is suitable for scenarios where occasional particle agglomeration during the mixing process leads to sudden changes in content.

[0118] S23. Based on the mixing quality deviation and the changes in mixing parameters, analyze the influence of mixing parameters on mixing quality.

[0119] During the mixing process, stirring parameters (such as the stirring speed of the stirring roller) will affect the mixing quality to a certain extent. Since liquid needs to be added during the mixing process, improper stirring parameters may lead to uneven material distribution and clumping.

[0120] In some implementations, the method of S23 may specifically include:

[0121] Obtain the parameter change curves of the mixing equipment's stirring parameters over time during the mixing process. Figure 7 Taking stirring speed as an example, a speed change curve of stirring speed (unit RPM) as a function of monitoring time (unit min) is shown; by comparing the parameter change curve with the content change curve, the correlation between stirring parameters and mixing quality is analyzed.

[0122] Specifically, taking stirring speed as an example, the correlation between stirring speed and mixing quality... The calculation formula can be:

[0123]

[0124] In the formula, This represents the quantity of all components in the mixture.

[0125] Indicates the stirring speed during the monitoring period. Parameter variation curves and components The mean square error between the content change curves is used to quantify the degree of difference between the two curves. The greater the difference, the more significant the impact of the change in stirring speed on the composition. The lower the contribution to mixing uniformity.

[0126] Then, based on the correlation and the degree of mixing quality deviation, the degree of influence of the stirring parameters on the mixing quality is determined.

[0127] Specifically, taking stirring speed as an example, the degree of influence of stirring speed on mixing quality. The calculation formula can be:

[0128]

[0129] In the formula, To monitor the mixed quality deviation during the monitoring period.

[0130] This is used to quantify the mixing quality deviation and the average contribution of stirring speed. The product is greater than or equal to zero, and the larger the product, the worse the mixing quality and the lower the contribution of stirring speed to the mixing.

[0131] Taking a negative sign is used to convert the negative impact of mixing into the improvement of mixing quality by stirring speed, and this is done by calculating an exponential function. The range of values ​​is defined as (0, 1).

[0132] The higher the value, the better the mixing speed improves the mixing quality.

[0133] In other implementations, mutual information can be used to analyze the correlation between the mixing quality deviation and the stirring parameters. Mutual information can capture nonlinear correlations (such as a sharp decrease in the rate of deviation reduction after the rotation speed exceeds a certain threshold), and is suitable for scenarios where there is a nonlinear relationship between stirring parameters and mixing quality (such as mixing high-viscosity materials).

[0134] Based on the above technical solutions, dynamic monitoring of component uniformity during the mixing process, quantitative evaluation of mixing quality, and precise analysis of stirring effect are realized. This effectively solves the problem of poor mixing quality caused by differences in raw material properties, and ultimately significantly improves the uniformity and nutritional balance of feed mixing.

[0135] In one possible implementation, combining Figure 4 ,like Figure 8 As shown, the method for determining the correction coefficients of the mixing equipment operating parameters in S2 above, based on the particle physical morphology, can be specifically implemented through the following S24 to S25, which are explained in detail below:

[0136] S24. Based on the degree of influence of stirring parameters on mixing quality and particle physical morphology, determine the degree of reverse influence on mixing quality from two dimensions: stirring parameters and particle physical morphology.

[0137] In some implementation methods, the degree of reverse impact The calculation formula can be:

[0138]

[0139] In the formula, The value ranges from (0, 1) to represent the degree of influence of stirring speed on the mixing quality.

[0140] The uniformity of the mixture particles is denoted as (0, 1).

[0141] Through calculation and The reciprocal of the factor converts the degree of influence and uniformity into inverse indicators, meaning that the smaller the reciprocal value, the better the corresponding indicator performs.

[0142] By calculating the Euclidean norm, the overall difference in mixing effect and particle morphology is combined to obtain the degree of overall difference, thus eliminating the limitations of a single index in evaluating mixing quality.

[0143] S25. Normalize the degree of reverse influence to obtain the correction coefficient.

[0144] Specifically, due to the degree of reverse influence The value is greater than or equal to By normalizing and mapping to the [0, 1] interval, the correction coefficients are obtained. . The larger the value, the worse the overall performance in terms of mixing improvement and particle uniformity, and the greater the correction required to the mixing equipment parameters.

[0145] Based on the above technical solution, the degree of inverse influence of stirring parameters on mixing quality and particle physical morphology is determined from two dimensions: the degree of influence of stirring parameters on mixing quality and the degree of inverse influence of particle physical morphology. This achieves the integrated quantification of multi-dimensional mixing effects. Then, through normalization processing, correction coefficients that can be directly used for parameter control of mixing equipment are obtained, effectively solving the problem of poor mixing quality caused by differences in raw material properties, and providing a quantitative basis for the precise optimization of equipment operating parameters.

[0146] It should be noted that the order of the above embodiments of the present invention is merely for descriptive purposes and does not represent the superiority or inferiority of the embodiments. The processes depicted in the accompanying drawings do not necessarily require a specific or sequential order to achieve the desired result. In some embodiments, multitasking and parallel processing are also possible or may be advantageous.

[0147] The various embodiments in this specification are described in a progressive manner. The same or similar parts between the various embodiments can be referred to each other. Each embodiment focuses on describing the differences from other embodiments.

[0148] In this embodiment of the invention, the livestock feed mixing control device can be divided into functional units according to the above method example. For example, each function can be divided into its own functional unit, or two or more functions can be integrated into one processing unit. The integrated unit can be implemented in hardware or software. It should be noted that the unit division in this embodiment is illustrative and only represents one logical functional division; in actual implementation, there may be other division methods.

[0149] This invention also provides a schematic diagram of the hardware structure of a livestock feed mixing control device, see below. Figure 9 The livestock feed mixing control device 900 includes a processor 901, and optionally, a memory 902 connected to the processor 901.

[0150] In the first possible implementation, see Figure 9 The livestock feed mixing control device 900 also includes a transceiver 903. The processor 901, memory 902, and transceiver 903 are connected via a bus. The transceiver 903 is used to communicate with other devices or communication networks. Optionally, the transceiver 903 may include a transmitter and a receiver. The device in the transceiver 903 that implements the receiving function can be considered as a receiver, which is used to perform the receiving steps in the embodiments of the present invention. The device in the transceiver 903 that implements the transmitting function can be considered as a transmitter, which is used to perform the transmitting steps in the embodiments of the present invention.

[0151] Based on the first possible implementation method Figure 9 The schematic diagram shown can be used to illustrate the structure of the livestock feed mixing control device involved in the above embodiments.

[0152] in, Figure 9 The diagram can also illustrate the system chip in the livestock feed mixing control device. In this case, the actions performed by the aforementioned livestock feed mixing control device can be implemented by this system chip; the specific actions performed are described above and will not be repeated here.

[0153] In implementation, each step of the method provided in this embodiment can be completed by integrated logic circuits in the processor or by instructions in software form. The steps of the method disclosed in this embodiment can be directly manifested as being executed by a hardware processor, or being executed by a combination of hardware and software modules in the processor.

[0154] The processor in this invention may include, but is not limited to, at least one of the following: a central processing unit (CPU), a microprocessor, a digital signal processor (DSP), a microcontroller unit (MCU), or an artificial intelligence processor, etc., which are various computing devices that run software. Each computing device may include one or more cores for executing software instructions to perform calculations or processing. The processor may be a standalone semiconductor chip or integrated with other circuits into a single semiconductor chip. For example, it may be integrated with other circuits (such as encoding / decoding circuits, hardware acceleration circuits, or various bus and interface circuits) to form a System-on-a-Chip (SoC), or it may be integrated as a built-in processor within an ASIC. The ASIC with the integrated processor may be packaged separately or together with other circuits. In addition to the cores for executing software instructions to perform calculations or processing, the processor may further include necessary hardware accelerators, such as field-programmable gate arrays (FPGAs), programmable logic devices (PLDs), or logic circuits that implement dedicated logic operations.

[0155] The memory in the embodiments of the present invention may include at least one of the following types: read-only memory (ROM) or other types of static storage devices capable of storing static information and instructions; random access memory (RAM) or other types of dynamic storage devices capable of storing information and instructions; or electrically erasable programmable read-only memory (EEPROM). In some scenarios, the memory may also be a compact disc read-only memory (CD-ROM) or other optical disc storage, optical disc storage (including compressed optical discs, laser discs, optical discs, digital universal optical discs, Blu-ray discs, etc.), magnetic disk storage media or other magnetic storage devices, or any other medium capable of carrying or storing desired program code in the form of instructions or data structures and accessible by a computer, but is not limited thereto.

[0156] This invention also provides a computer-readable storage medium including instructions that, when run on a computer, cause the computer to perform any of the methods described above.

[0157] This invention also provides a computer program product containing instructions that, when run on a computer, cause the computer to perform any of the methods described above.

[0158] This invention also provides a chip, which includes a processor and an interface circuit. The interface circuit is coupled to the processor. The processor is used to run computer programs or instructions to implement the above-described method. The interface circuit is used to communicate with other modules outside the chip.

[0159] In the above embodiments, implementation can be achieved, in whole or in part, through software, hardware, firmware, or any combination thereof. When implemented using software programs, implementation can be, in whole or in part, in the form of a computer program product. This computer program product includes one or more computer instructions. When the computer program instructions are loaded and executed on a computer, all or part of the processes or functions described in the embodiments of the present invention are generated. The computer can be a general-purpose computer, a special-purpose computer, a computer network, or other programmable device. The computer instructions can be stored in a computer-readable storage medium or transmitted from one computer-readable storage medium to another. For example, computer instructions can be transmitted from one website, computer, server, or data center to another website, computer, server, or data center via wired (e.g., coaxial cable, fiber optic, digital subscriber line (DSL)) or wireless (e.g., infrared, wireless, microwave, etc.) means. The computer-readable storage medium can be any available medium accessible to a computer or a data storage device containing one or more servers, data centers, etc., that can be integrated with the medium. The available media can be magnetic media (e.g., floppy disks, hard disks, magnetic tapes), optical media (e.g., DVDs), or semiconductor media (e.g., solid-state disks (SSDs)).

[0160] Although the invention has been described herein in conjunction with various embodiments, those skilled in the art, by reviewing the accompanying drawings and the disclosure, will understand and implement other variations of the disclosed embodiments in carrying out the claimed invention. In this invention, the word "comprising" does not exclude other components or steps, and "a" or "an" does not exclude a plurality. A single processor or other unit can implement several of the functions listed in this invention.

[0161] Although the invention has been described in conjunction with specific features and embodiments, it is obvious that various modifications and combinations can be made therein without departing from the spirit and scope of the invention. Accordingly, this specification and drawings are merely illustrative of the invention and are to be considered as covering any and all modifications, variations, combinations, or equivalents within the scope of the invention. Clearly, those skilled in the art can make various alterations and modifications to the invention without departing from its spirit and scope. Thus, if such modifications and modifications of the invention fall within the scope of the invention and its equivalents, the invention is also intended to include such modifications and modifications.

Claims

1. A method for controlling the mixing of livestock feed, characterized in that, include: Monitor the composition and particle uniformity of the mixture during the mixing process, as well as the stirring parameters of the mixing equipment; The mixing process includes: stirring the dry materials in the mixing equipment and adding liquid by spraying; the composition is obtained by near-infrared spectroscopy analysis; the particle uniformity of the mixed materials is obtained by image analysis, including: acquiring images of the mixed materials at multiple times in the mixing equipment, performing grayscale and edge detection processing on the mixed material images, identifying multiple closed regions in the mixed material images, and analyzing the particle uniformity of the mixed materials during the mixing process based on the morphological characteristics of the multiple closed regions; Based on the analysis of the component composition, the influence of stirring parameters on the mixing quality is analyzed, and combined with the particle uniformity of the mixed materials, the correction coefficients for the operating parameters of the mixing equipment are determined. The current operating parameters of the mixing equipment are adjusted according to the correction coefficient, so that the mixing equipment operates at the adjusted target operating parameters.

2. The method for controlling feed mixing in livestock farming according to claim 1, characterized in that, The analysis of the influence of stirring parameters on mixing quality based on the component composition includes: The actual content of each component at multiple time points is fitted to obtain the content change curve of each component over time; By comparing the actual content of each component with the expected content, and taking into account the fluctuation of the content change curve, the mixing quality deviation degree used to characterize the uniformity of component distribution in the mixture is determined. Based on the mixing quality deviation and the changes in the mixing parameters, the influence of the mixing parameters on the mixing quality is analyzed.

3. The method for controlling feed mixing in livestock farming according to claim 2, characterized in that, The step of analyzing the influence of the stirring parameters on the mixing quality based on the mixing quality deviation and the changes in the stirring parameters includes: Obtain the parameter change curves of the mixing equipment's stirring parameters over time during the mixing process; By comparing the parameter change curves with the content change curves, the correlation between the stirring parameters and the mixing mass is analyzed. Based on the correlation and the mixing quality deviation, the degree of influence of the stirring parameters on the mixing quality is determined.

4. The method for controlling feed mixing in livestock farming according to claim 3, characterized in that, The correction coefficient for determining the operating parameters of the mixing equipment based on the particle uniformity of the mixed materials includes: Based on the degree of influence of the stirring parameters on the mixing quality and the uniformity of the mixed materials particles, determine the degree of reverse influence on the mixing quality from two dimensions: the stirring parameters and the uniformity of the mixed materials particles. The degree of the reverse influence is normalized to obtain the correction coefficient.

5. The method for controlling feed mixing in livestock farming according to claim 4, characterized in that, The step of adjusting the current operating parameters of the mixing equipment according to the correction coefficient includes: The correction coefficient is compared with a preset threshold, and the control strategy is determined based on the comparison result; the control strategy includes: increasing the stirring speed of the mixing equipment, or increasing at least one of the spraying speed and the jet pressure.

6. The method for controlling feed mixing in livestock farming according to any one of claims 1 to 5, characterized in that, The composition of the components was obtained by near-infrared spectroscopy analysis, including: The spectral data of the mixture are collected by a near-infrared spectrometer inside the mixing device; By using a preset calibration model, the spectral data of the mixture are analyzed to obtain the actual content of multiple components in the mixture.

7. A livestock feed mixing device, characterized in that, include: Mixing container, stirring module, liquid dispensing module, data monitoring module, intelligent control module; The mixing container is configured to hold the mixed materials; The stirring module is configured to stir the mixture. The liquid input module is configured to add liquid into the mixing container by spraying. The data monitoring module includes: a near-infrared spectrometer, a camera, and a stirring parameter sensor; the near-infrared spectrometer is configured to collect spectral data of the mixture; the camera is configured to collect images of the mixture in the mixing equipment; and the stirring parameter sensor is configured to collect stirring parameters of the stirring module. The intelligent control module is configured to: receive the spectral data, the image of the mixture, and the stirring parameters from the data monitoring module; obtain the composition of the mixture through the spectral data; and obtain the particle uniformity of the mixture through the image of the mixture. Obtaining the particle uniformity includes: acquiring images of the mixture at multiple times within the mixing equipment; performing grayscale conversion and edge detection processing on the images; identifying multiple closed regions in the images; analyzing the particle uniformity of the mixture during the mixing process based on the morphological characteristics of the multiple closed regions; analyzing the influence of the stirring parameters on the mixing quality based on the composition; and determining a correction coefficient for the operating parameters of the mixing equipment based on the particle uniformity of the mixture; and generating control commands based on the correction coefficient to regulate the current operating parameters of the stirring module and the liquid input module. The stirring module and the liquid injection module respond to the control command and operate based on the adjusted target operating parameters.

8. The livestock feed mixing equipment according to claim 7, characterized in that, The liquid delivery module includes an inlet pipe and nozzles deployed at multiple locations on the inner wall of the mixing container; The nozzles add liquid to the mixing container by spraying from multiple locations.

9. The livestock feed mixing equipment according to claim 7, characterized in that, Also includes: Jet module; The jet module includes an air inlet pipe and jet nozzles deployed at multiple locations on the inner wall of the mixing container; the jet nozzles are used to spray gas to assist in stirring and prevent the mixture from sticking to the inner wall of the equipment.

Images