Intelligent regulation and control method for shrimp feed production, electronic equipment and storage medium

Abstract

The invention discloses an intelligent regulation and control method for shrimp feed production, electronic equipment and a storage medium. The intelligent regulation and control method comprises the following steps: acquiring initial images of multiple angles at a shrimp feed finished product discharging end; performing preprocessing and feature extraction processing on the plurality of initial images to obtain multi-dimensional features; inputting the multi-dimensional features into a pre-trained shrimp material quality evaluation model to obtain a first quality evaluation result; according to the first quality evaluation result, determining a first processing link possibly influencing the processing quality of the shrimp material, and obtaining initial parameters of the first processing link; adjusting the initial parameter according to the first quality evaluation result and a preset target quality standard to obtain a target parameter; and carrying out image acquisition, analysis and judgment processing on the shrimp material after the target parameters to obtain a second quality evaluation result, and carrying out parameter adjustment according to the second quality evaluation result until the second quality evaluation result reaches the target quality standard. The production quality and the production efficiency of products can be improved.

Description

Technical Field

[0001] The present application relates to the technical field of feed production, and in particular to an intelligent control method for shrimp feed production, an electronic device and a storage medium. Background Art

[0002] In the field of shrimp feed production, with the vigorous development of aquaculture, the requirements for shrimp feed quality and production efficiency are increasing. The traditional shrimp feed production process mainly relies on manual experience for quality control and production link regulation, which has many disadvantages.

[0003] Manual testing of shrimp feed quality is often subjective, inefficient, and prone to misjudgment. Due to visual fatigue and individual differences in the human eye, it is difficult to maintain accurate and consistent quality assessment standards during long-term, large-scale production processes. For example, in judging the particle integrity, color uniformity, and the presence of impurities in shrimp feed, the accuracy and stability of manual testing are difficult to guarantee. This results in some shrimp feed of poor quality being mixed with qualified products and flowing into the market, affecting shrimp farming results and increasing farmers' farming risks; on the other hand, some shrimp feed whose quality could have been improved by fine-tuning production parameters is misjudged as unqualified, resulting in a waste of production resources.

[0004] At the same time, the parameters of the shrimp feed production process are manually adjusted with obvious lag. When problems with the shrimp feed quality are found, it takes a long time to check the processing links that may have problems, such as the raw material mixing ratio, granulation temperature, drying time, etc., and then adjust the relevant parameters based on experience. This trial-and-error adjustment method is not only inefficient, but also difficult to quickly and accurately determine the optimal parameter combination, making it difficult for the production process to be continuously and stably in the best state, which in turn affects the overall production efficiency of the shrimp feed and the consistency of product quality. Summary of the invention

[0005] The present application aims to solve at least one of the technical problems existing in the prior art. To this end, the present application proposes an intelligent control method for shrimp feed production, an electronic device and a storage medium, which can improve the production quality and production efficiency of the product.

[0006] In a first aspect, the present application provides a method for intelligently controlling shrimp feed production, comprising:

[0007] Acquire initial images at multiple angles at the finished product discharging end of the shrimp feed production line;

[0008] Performing preprocessing and feature extraction processing on the plurality of initial images to obtain multi-dimensional features;

[0009] Inputting the multi-dimensional features into a pre-trained shrimp feed quality assessment model to obtain a first quality assessment result; wherein the shrimp feed quality assessment model is obtained by training with a plurality of historical shrimp feed samples;

[0010] According to the first quality assessment result, determining the first processing link that may affect the shrimp feed processing quality, and obtaining initial parameters of the first processing link;

[0011] The initial parameters are adjusted again according to the first quality assessment result and the preset target quality standard to obtain the target parameters of the first processing link;

[0012] The shrimp feed after the target parameters are adjusted is subjected to image acquisition, analysis and determination processing again to obtain a second quality assessment result, and parameters are adjusted according to the second quality assessment result until the second quality assessment result reaches the target quality standard.

[0013] According to the shrimp feed production intelligent control method of the first aspect of the present application, there are at least the following beneficial effects: at the finished product discharge end of the shrimp feed production line, the initial image is obtained by using image acquisition devices at multiple angles, which can fully reflect the appearance information of the finished shrimp feed, such as particle shape, color, surface texture, etc. The multiple initial images collected are preprocessed, and feature extraction is performed to extract multi-dimensional features. The extracted multi-dimensional features are input into a pre-trained shrimp feed quality assessment model, which is obtained based on a large number of historical shrimp feed samples. The model can accurately output the first quality assessment result according to the input features. According to the first quality assessment result, combined with the production process knowledge and the internal logic of the model, the first processing link that may affect the processing quality of the shrimp feed is determined, and the initial parameters are adjusted again according to the first quality assessment result and the preset target quality standard to obtain the target parameters of the first processing link. The shrimp feed after adjusting the target parameters is re-imaged, analyzed and judged to obtain the second quality assessment result. Repeat this process, that is, adjust the parameters again according to the second quality assessment result, until the second quality assessment result reaches the target quality standard. After each parameter adjustment, the adjustment effect can be quickly evaluated through image acquisition and analysis to form a closed-loop intelligent control system. The use of image acquisition and analysis technology and a well-trained quality assessment model avoids the subjectivity and instability of human eye detection. Through multi-dimensional feature extraction and the learning ability of the model, the quality of shrimp feed can be judged more accurately. Once the quality assessment results deviate, the processing links that may have problems can be quickly determined based on the model and production process knowledge, and the relevant parameters can be quickly adjusted using intelligent algorithms, solving the problem of insufficient accuracy and stability of manual detection, reducing misjudgments, ensuring that only qualified shrimp feed enters the market, ensuring the effect of shrimp farming, reducing the risks of farmers, and improving the production quality of products. At the same time, compared with manual trial-and-error investigation and adjustment, it greatly shortens the time and improves production efficiency.

[0014] According to some embodiments of the first aspect of the present application, the multi-dimensional features include geometric features, color features, and texture features;

[0015] The preprocessing and feature extraction processing of the plurality of initial images to obtain multi-dimensional features includes:

[0016] Removing high-frequency noise from the initial image and suppressing low-frequency interference from the initial image to obtain a first image;

[0017] Gray-scale the first image to obtain a target image;

[0018] Extracting edge information of the target image and obtaining geometric features according to the edge information;

[0019] Performing color conversion and statistical processing on the initial image to obtain color features;

[0020] Texture features are obtained according to the target images at multiple different angles.

[0021] According to some embodiments of the first aspect of the present application, the historical shrimp feed samples include sample images and actual defect type labels corresponding to each of the sample images;

[0022] The shrimp feed quality assessment model is obtained by the following steps:

[0023] According to the sample image, a geometric sample feature, a color sample feature and a texture sample feature are obtained;

[0024] The geometric sample features, the color sample features and the texture sample features are respectively input into an initial blanking quality assessment model for classification processing to obtain a predicted defect type label;

[0025] Obtaining a loss value according to the actual defect type label and the predicted defect type label;

[0026] According to the loss value, the parameters of the shrimp feed quality assessment model are adjusted until the loss value reaches a preset loss threshold.

[0027] According to some embodiments of the first aspect of the present application, inputting the multi-dimensional features into a pre-trained shrimp feed quality assessment model to obtain a first quality assessment result includes:

[0028] Inputting the geometric features, the color features, and the texture features into a pre-trained quality assessment model;

[0029] Extracting the target perimeter, target area and target circularity according to the geometric features;

[0030] Obtaining a geometric defect result according to the target perimeter and a preset perimeter threshold range, the target area and a preset area threshold range, the target circularity and a preset circularity threshold range;

[0031] According to the color characteristics, a first color intensity, a second color intensity and a third color intensity are obtained;

[0032] Obtaining a color defect result according to the first color intensity and a preset first color range, the second color intensity and a preset second color range, and the third color intensity and a preset third color range;

[0033] Determining a target smoothness of the shrimp feed surface according to the texture characteristics;

[0034] Obtaining a texture defect result according to the target smoothness and a preset smoothness threshold range;

[0035] A first quality assessment result is generated according to the geometric defect result, the color defect result and the texture defect result.

[0036] According to some embodiments of the first aspect of the present application, determining the first processing link that may affect the shrimp feed processing quality according to the first quality assessment result, and obtaining initial parameters of the first processing link, includes:

[0037] When there is a geometric defect result in the first quality assessment result, determining that the first processing link that may affect the shrimp feed processing quality is a crushing link or a granulation link;

[0038] respectively obtaining the initial parameters of the pulverizing step and the initial parameters of the granulating step;

[0039] Correspondingly, the initial parameters are adjusted again according to the first quality assessment result and the preset target quality standard to obtain the target parameters of the first processing link, including:

[0040] Generating a first adjustment strategy according to the geometric defect result and a preset target quality standard;

[0041] According to the first adjustment strategy, the crushing initial parameters and the granulation initial parameters are adjusted to obtain crushing target parameters of the crushing link and granulation target parameters of the granulation link.

[0042] According to some embodiments of the first aspect of the present application, determining the first processing link that may affect the shrimp feed processing quality according to the first quality assessment result, and obtaining initial parameters of the first processing link, includes:

[0043] When there is a color defect result in the first quality assessment result, determining that the first processing link that may affect the shrimp feed processing quality is the ingredient link, the conditioning link or the drying link;

[0044] Respectively obtaining initial parameters of batching in the batching process, initial parameters of conditioning in the conditioning process, and initial parameters of drying in the drying process;

[0045] Correspondingly, the initial parameters are adjusted again according to the first quality assessment result and the preset target quality standard to obtain the target parameters of the first processing link, including:

[0046] generating a second adjustment strategy according to the color defect result and a preset target quality standard;

[0047] According to the second adjustment strategy, the batching initial parameters, the conditioning initial parameters and the drying initial parameters are adjusted to obtain the batching target parameters of the batching link, the conditioning target parameters of the conditioning link and the drying target parameters of the drying link.

[0048] According to some embodiments of the first aspect of the present application, determining the first processing link that may affect the shrimp feed processing quality according to the first quality assessment result, and obtaining initial parameters of the first processing link, includes:

[0049] When there is a texture defect result in the first quality assessment result, determining that the first processing link that may affect the shrimp feed processing quality is a crushing link, a filtering link or a conditioning link;

[0050] Respectively obtaining initial crushing parameters of the crushing step, initial filtering parameters of the filtering step, and initial conditioning parameters of the conditioning step;

[0051] Correspondingly, the initial parameters are adjusted again according to the first quality assessment result and the preset target quality standard to obtain the target parameters of the first processing link, including:

[0052] generating a third adjustment strategy according to the texture defect result and a preset target quality standard;

[0053] According to the third adjustment strategy, the crushing initial parameters, the filtering initial parameters and the tempering initial parameters are adjusted to obtain the crushing target parameters of the crushing link, the filtering target parameters of the filtering link and the tempering target parameters of the tempering link.

[0054] According to some embodiments of the first aspect of the present application, the present invention further includes:

[0055] When adjusting parameters, get the current timestamp;

[0056] Generate change information according to the current timestamp, the adjusted first processing link, the initial parameter value before adjustment, the target parameter value after adjustment, and the corresponding first quality assessment result;

[0057] The change information is saved in a preset log text.

[0058] In a second aspect, the present application further provides an electronic device, including:

[0059] at least one memory;

[0060] at least one processor;

[0061] at least one program;

[0062] The program is stored in the memory, and the processor executes at least one of the programs to implement the intelligent control method for shrimp feed production as described in any embodiment of the first aspect.

[0063] In a third aspect, the present application further provides a computer-readable storage medium, wherein the computer-readable storage medium stores a computer-executable signal, and the computer-executable signal is used to execute the intelligent control method for shrimp feed production as described in any embodiment of the first aspect.

[0064] Additional aspects and advantages of the present application will be given in part in the description below, and in part will become apparent from the description below, or will be learned through the practice of the present application. BRIEF DESCRIPTION OF THE DRAWINGS

[0065] Additional aspects and advantages of the present application will become apparent and easily understood from the description of the embodiments in conjunction with the following drawings, in which:

[0066] Figure 1 A flow chart of the intelligent control method for shrimp feed production provided in this application;

[0067] Figure 2 For this application Figure 1 Flow chart of step S120;

[0068] Figure 3 This is a flowchart of the present application regarding the training of the shrimp feed quality assessment model in step S130;

[0069] Figure 4 For this application Figure 1 Flow chart of step S130;

[0070] Figure 5 For this application Figure 1 Flowchart of processing of geometric defect results in steps S140 and S150;

[0071] Figure 6 For this application Figure 1 Flowchart of processing of color defect results in steps S140 and S150;

[0072] Figure 7 For this application Figure 1 A flowchart of processing steps S140 and S150 for texture defect results;

[0073] Figure 8 This is a flow chart of another embodiment of the intelligent control method for shrimp feed production provided in this application. DETAILED DESCRIPTION

[0074] The embodiments of the present application are described in detail below, and examples of the embodiments are shown in the accompanying drawings, wherein the same or similar reference numerals throughout represent the same or similar elements or elements having the same or similar functions. The embodiments described below with reference to the accompanying drawings are exemplary and are only used to explain the present application, and cannot be understood as limiting the present application.

[0075] In the description of the present application, it should be understood that descriptions involving orientation, such as up, down, front, back, left, right, etc., indicating orientations or positional relationships, are based on the orientations or positional relationships shown in the accompanying drawings, and are only for the convenience of describing the present application and simplifying the description, and do not indicate or imply that the device or element referred to must have a specific orientation, be constructed and operated in a specific orientation, and therefore should not be understood as a limitation on the present application.

[0076] In the description of this application, if there is a description of first or second, it is only for the purpose of distinguishing technical features, and cannot be understood as indicating or implying relative importance or implicitly indicating the number of the indicated technical features or implicitly indicating the order of the indicated technical features.

[0077] In the description of this application, unless otherwise clearly defined, terms such as setting, installing, connecting, etc. should be understood in a broad sense, and technicians in the relevant technical field can reasonably determine the specific meanings of the above terms in this application based on the specific content of the technical solution.

[0078] In the field of shrimp feed production, with the vigorous development of aquaculture, the requirements for shrimp feed quality and production efficiency are increasing. The traditional shrimp feed production process mainly relies on manual experience for quality control and production link regulation, which has many disadvantages.

[0079] Manual testing of shrimp feed quality is often subjective, inefficient, and prone to misjudgment. Due to visual fatigue and individual differences in the human eye, it is difficult to maintain accurate and consistent quality assessment standards during long-term, large-scale production processes. For example, in judging the particle integrity, color uniformity, and the presence of impurities in shrimp feed, the accuracy and stability of manual testing are difficult to guarantee. This results in some shrimp feed of poor quality being mixed with qualified products and flowing into the market, affecting shrimp farming results and increasing farmers' farming risks; on the other hand, some shrimp feed whose quality could have been improved by fine-tuning production parameters is misjudged as unqualified, resulting in a waste of production resources.

[0080] At the same time, the parameters of the shrimp feed production process are manually adjusted with obvious lag. When problems with the shrimp feed quality are found, it takes a long time to check the processing links that may have problems, such as the raw material mixing ratio, granulation temperature, drying time, etc., and then adjust the relevant parameters based on experience. This trial-and-error adjustment method is not only inefficient, but also difficult to quickly and accurately determine the optimal parameter combination, making it difficult for the production process to be continuously and stably in the best state, which in turn affects the overall production efficiency of the shrimp feed and the consistency of product quality.

[0081] Based on this, the present application also provides an intelligent control method for shrimp feed production, an electronic device and a storage medium to solve the above-mentioned technical problems. The technical solutions provided by the present application are described in detail one by one below.

[0082] The shrimp feed production process generally includes processes such as material separation, filtering, crushing, batching, mixing, tempering, granulation, drying, cooling and packaging. Some of the processes may be repeated multiple times according to the components of the shrimp feed. For example, if the addition time of a certain raw material needs to be strictly controlled, the processes of material separation, filtering, crushing, batching and mixing may need to be repeated multiple times. The specific production process of the shrimp feed is not limited in this application.

[0083] First, refer to Figure 1 The present application provides a method for intelligently controlling shrimp feed production, which may include but is not limited to the following steps:

[0084] Step S110: acquiring initial images at multiple angles at the finished product discharge end of the shrimp feed production line;

[0085] Step S120: performing preprocessing and feature extraction processing on the multiple initial images to obtain multi-dimensional features;

[0086] Step S130: inputting the multi-dimensional features into a pre-trained shrimp feed quality assessment model to obtain a first quality assessment result; wherein the shrimp feed quality assessment model is obtained by training with a plurality of historical shrimp feed samples;

[0087] Step S140: determining the first processing link that may affect the shrimp feed processing quality according to the first quality assessment result, and obtaining initial parameters of the first processing link;

[0088] Step S150: adjusting the initial parameters again according to the first quality assessment result and the preset target quality standard to obtain the target parameters of the first processing link;

[0089] Step S160: re-collecting images, analyzing and judging the shrimp feed after adjusting the target parameters to obtain a second quality assessment result, and adjusting parameters according to the second quality assessment result until the second quality assessment result reaches the target quality standard.

[0090] In step S110 to step S160, at the finished product discharge end of the shrimp feed production line, an initial image is obtained by using image acquisition equipment at multiple angles, which can fully reflect the appearance information of the finished shrimp feed, such as particle shape, color, surface texture, etc. The multiple initial images collected are preprocessed, and feature extraction is performed to extract multi-dimensional features. The extracted multi-dimensional features are input into a pre-trained shrimp feed quality assessment model, which is obtained based on a large number of historical shrimp feed samples and can accurately output the first quality assessment result according to the input features. According to the first quality assessment result, combined with the production process knowledge and the internal logic of the model, the first processing link that may affect the processing quality of the shrimp feed is determined, and the initial parameters are adjusted again according to the first quality assessment result and the preset target quality standard to obtain the target parameters of the first processing link. The shrimp feed after adjusting the target parameters is re-imaged, analyzed and judged to obtain the second quality assessment result. Repeat this process, that is, adjust the parameters again according to the second quality assessment result, until the second quality assessment result reaches the target quality standard. After each parameter adjustment, the adjustment effect can be quickly evaluated through image acquisition and analysis to form a closed-loop intelligent control system. The use of image acquisition and analysis technology and a well-trained quality assessment model avoids the subjectivity and instability of human eye detection. Through multi-dimensional feature extraction and the learning ability of the model, the quality of shrimp feed can be judged more accurately. Once the quality assessment results deviate, the processing links that may have problems can be quickly determined based on the model and production process knowledge, and the relevant parameters can be quickly adjusted using intelligent algorithms, solving the problem of insufficient accuracy and stability of manual detection, reducing misjudgments, ensuring that only qualified shrimp feed enters the market, ensuring the effect of shrimp farming, reducing the risks of farmers, and improving the production quality of products. At the same time, compared with manual trial-and-error investigation and adjustment, it greatly shortens the time and improves production efficiency.

[0091] It is understandable that, referring to Figure 2 The multi-dimensional features include geometric features, color features and texture features. In step S120, the following steps may be included but are not limited to:

[0092] Step S210: removing high-frequency noise of the initial image and suppressing low-frequency interference of the initial image to obtain a first image;

[0093] Step S220: grayscale the first image to obtain a target image;

[0094] Step S230: extracting edge information of the target image, and obtaining geometric features according to the edge information;

[0095] Step S240: performing color conversion and statistical processing on the initial image to obtain color features;

[0096] Step S250: obtaining texture features according to target images at multiple different angles.

[0097] In step S210, a suitable filtering algorithm, such as Gaussian filtering, is used to remove high-frequency noise in the initial image. Such noise usually appears as small spots or pepper-like interference in the image, which will affect the accuracy of subsequent feature extraction. At the same time, methods such as mean filtering are used to suppress low-frequency interference, which may cause the overall image to be blurred or large areas of abnormal grayscale changes to appear. This makes the image smoother and clearer while retaining the key information of the shrimp feed, which is conducive to subsequent processing.

[0098] In step S220, the first image is grayed to obtain a target image. Specifically, the gray value can be calculated based on the weighted average of the red, green and blue channels in the color image. The gray scale processing can simplify the image data and reduce the amount of calculation. In many cases, the gray scale image is sufficient to reflect the important features of the shrimp feed, such as the shape and texture, and is convenient for subsequent edge extraction and texture analysis operations.

[0099] In step S230, edge detection is performed on the target image using an edge detection operator. Based on the detected edge information, the geometric features of the shrimp feed, such as the perimeter, area, aspect ratio, circularity, etc. of the particles, can be calculated. For example, the perimeter can be approximately calculated by the number of edge pixels, the area can be obtained by counting the number of pixels surrounded by the edge, the aspect ratio is determined by calculating the ratio of the length to the width of the circumscribed rectangle of the particle, and the circularity is calculated according to the relationship formula between the perimeter and the area. These geometric features can intuitively reflect the shape characteristics of the shrimp feed particles.

[0100] In step S240, the RGB color space of the target image is converted to other spaces that are more suitable for color feature analysis, such as the HSV color space. In the HSV space, the hue (H) can reflect the basic color categories of the shrimp feed (such as yellow, brown, etc.), the saturation (S) indicates the vividness of the color, and the value (V) reflects the brightness of the color. The color features can be obtained by performing statistical processing on each color channel of the converted image, such as calculating the histogram distribution of the hue channel. For example, the number of pixels in different hue intervals is counted to determine the uniformity of the shrimp feed color and whether there are abnormal color areas.

[0101] In step S250, texture features are obtained according to target images at multiple different angles. First, for a single target image, a gray level co-occurrence matrix (GLCM) method can be used. By calculating the joint probability distribution of pixel grayscale pairs at different distances and directions in the image, texture feature parameters such as energy, contrast, correlation, and entropy are extracted from the GLCM. For example, energy reflects the uniformity of the image grayscale distribution and the coarseness of the texture, contrast reflects the degree of difference in the pixel grayscale values ​​in the image, correlation represents the linear relationship between pixels, and entropy reflects the randomness or complexity of the image. Then, by combining the texture feature parameters of target images at multiple different angles, the surface texture characteristics of the shrimp feed can be more comprehensively described, because the shrimp feed may present different texture performances at different angles, and the multi-angle texture features can more accurately judge the consistency of the texture and processing technology of the shrimp feed.

[0102] The extraction of multi-dimensional features (geometry, color, texture) can comprehensively evaluate the quality of shrimp feed from different aspects. Geometric features reflect the shape, color features reflect the color changes of raw materials and processing, and texture features reflect the texture. This comprehensive evaluation capability helps to more accurately locate the root cause of shrimp feed quality problems, so as to take targeted measures to improve them.

[0103] Reference Figure 3 It can be understood that the historical shrimp feed samples include sample images and actual defect type labels corresponding to each sample image. The shrimp feed quality assessment model can be obtained through the following steps:

[0104] Step S310: obtaining geometric sample features, color sample features and texture sample features according to the sample image;

[0105] Step S320: respectively inputting the geometric sample features, the color sample features and the texture sample features into the initial blanking quality assessment model for classification processing to obtain a predicted defect type label;

[0106] Step S330: Obtaining a loss value according to the actual defect type label and the predicted defect type label;

[0107] Step S340: adjusting the parameters of the shrimp feed quality assessment model according to the loss value until the loss value reaches a preset loss threshold.

[0108] For each sample image in the historical shrimp feed sample, it is processed according to the feature extraction method mentioned above. First, the high-frequency noise of the sample image is removed and the low-frequency interference is suppressed to obtain the first sample image. Then the first sample image is grayed to obtain the target sample image. Then the geometric sample features, color sample features and texture sample features of the target sample image are extracted. The extracted geometric sample features, color sample features and texture sample features are respectively input into the initial shrimp feed quality assessment model. This model can be a classification model based on deep learning. The model classifies each sample image according to the different dimensional features of the input and predicts the corresponding predicted defect type label. For example, the predicted label may include different types of defect categories such as particle shape defects, uneven color, rough texture, etc. According to the actual defect type label and the predicted defect type label corresponding to each sample image, the loss value is calculated, and the parameters of the shrimp feed quality assessment model are adjusted using an optimization algorithm based on the calculated loss value. Using historical shrimp feed sample data to train the quality assessment model has transformed the quality assessment process from traditional manual experience judgment to data-driven scientific assessment. The model's prediction results are based on learning and statistical laws from a large number of samples, avoiding the subjectivity and uncertainty of manual evaluation and improving the reliability of quality assessment.

[0109] Reference Figure 4 It is understandable that step S130 may include but is not limited to the following steps:

[0110] Step S410: inputting geometric features, color features and texture features into a pre-trained quality assessment model;

[0111] Step S420: extracting the target perimeter, target area and target circularity according to the geometric features;

[0112] Step S430: obtaining a geometric defect result according to the target perimeter and a preset perimeter threshold range, the target area and a preset area threshold range, the target circularity and a preset circularity threshold range;

[0113] Step S440: obtaining a first color intensity, a second color intensity, and a third color intensity according to the color characteristics;

[0114] Step S450: obtaining a color defect result according to the first color intensity and the preset first color range, the second color intensity and the preset second color range, and the third color intensity and the preset third color range;

[0115] Step S460: determining the target smoothness of the shrimp feed surface according to the texture characteristics;

[0116] Step S470: obtaining a texture defect result according to the target smoothness and a preset smoothness threshold range;

[0117] Step S480: Generate a first quality assessment result according to the geometric defect result, the color defect result and the texture defect result.

[0118] In step S420 to step S430, the target perimeter, target area and target circularity are extracted according to the input geometric features. Then, the target perimeter is compared with the preset perimeter threshold range. For example, if the preset shrimp pellet perimeter normal range is within [C min ,C max ], when the target circumference is less than C min or greater than C max When the target area is less than the preset area threshold range, the target circularity is compared with the preset circularity threshold range.

[0119] In step S440 to step S450, based on the input color features, the first color intensity, the second color intensity and the third color intensity are obtained in the quality assessment model. These color intensities can be the intensity of the main color components in a specific color space (such as hue, saturation and brightness in the HSV color space) or after statistical processing. Then, the first color intensity is compared with the preset first color range. For example, when judging whether the color of shrimp feed is normal, if the first color intensity corresponds to the main color of the shrimp feed, and the preset normal color range is [H min ,H max ], when the first color intensity is less than H min or greater than H max , there may be color anomalies. Similarly, the second color intensity is compared with the preset second color range, and the third color intensity is compared with the preset third color range. A color defect result is obtained based on the comparison result. If all color intensities are within their respective preset ranges, the color defect result may be "no color defect"; otherwise, the color defect result will indicate which color intensity is out of range and causes the color problem, such as "too high saturation" or "insufficient brightness".

[0120] In step S460 to step S470, the target smoothness of the shrimp feed surface is determined in the quality assessment model using the input texture features. The target smoothness can be indirectly measured by parameters in the texture features (such as the energy in the gray level co-occurrence matrix, etc.). A higher energy may indicate a more uniform texture and a relatively smooth surface; a lower energy may indicate a complex texture and an uneven surface. The target smoothness is compared with a preset smoothness threshold range. For example, the preset normal range of smoothness is [S min ,S max ], if the target smoothness is less than S min, indicating that the surface of the shrimp feed may be too rough; if it is greater than S max , the surface may be too smooth and does not meet the texture requirements of normal shrimp feed. According to the comparison results, the texture defect results are obtained, such as "rough surface" or "too smooth surface".

[0121] It is understandable that for the geometric defect results, refer to Figure 5 In step S140, the following steps may be included but not limited to:

[0122] Step S510: when there is a geometric defect result in the first quality assessment result, determining that the first processing link that may affect the shrimp feed processing quality is a crushing link or a granulation link;

[0123] Step S520: respectively obtaining the initial parameters of the crushing process and the initial parameters of the granulation process;

[0124] Correspondingly, in step S150, the following steps may be included but not limited to:

[0125] Step S530: generating a first adjustment strategy according to the geometric defect result and a preset target quality standard;

[0126] Step S540: According to the first adjustment strategy, the initial parameters of the crushing and the initial parameters of the granulation are adjusted to obtain the crushing target parameters of the crushing stage and the granulation target parameters of the granulation stage.

[0127] In step S510 to step S520, when there is a geometric defect result in the first quality assessment result, it indicates that there is a problem with the shape of the shrimp feed. Since the shape of the shrimp feed is mainly formed in the crushing and granulation stages, the first processing stage that may affect the processing quality of the shrimp feed is determined to be the crushing stage or the granulation stage. For example, if the size of the shrimp feed particles is uneven, it may be caused by the inconsistent degree of crushing of the raw materials in the crushing stage; if the shape of the shrimp feed particles is irregular, it is likely caused by factors such as the mold and pressure in the granulation stage.

[0128] For a certain crushing step, obtain its initial crushing parameters. These parameters may include the speed of the crusher, the mesh size, the feed speed, etc. For example, the speed of the crusher will affect the degree of crushing of the raw material, the mesh size determines the size range of the raw material particles after crushing, and the feed speed is related to the efficiency and quality of the crushing. At the same time, obtain the initial granulation parameters of the granulation step. The initial granulation parameters may include the mold size of the granulator, the pressure of the roller, the speed of the cutter, etc. For example, the mold size directly determines the basic shape and size of the shrimp feed particles, the pressure of the roller affects the compactness and shape integrity of the particles, and the cutter speed affects the length and other geometric characteristics of the particles.

[0129] In step S530 to step S540, a first adjustment strategy is generated according to the geometric defect result and the preset target quality standard. If the geometric defect result is that the shrimp feed particles are too large, and the target quality standard requires the particles to be of moderate size, then the first adjustment strategy may include reducing the screen aperture of the crushing link, increasing the cutter speed of the granulation link, or adjusting the mold size of the granulator to reduce the particle size. If the shape of the shrimp feed particles is irregular, the first adjustment strategy may be to adjust the roller pressure of the granulation link to make it more uniform, or to check whether the granulator mold is damaged and perform corresponding repairs or replacements. According to the first adjustment strategy, the crushing initial parameters and the granulation initial parameters are adjusted to obtain the crushing target parameters of the crushing link and the granulation target parameters of the granulation link. By associating the geometric defect results with the processing links, the crushing link or granulation link that may affect the shape quality of the shrimp feed can be accurately located. Shrimp feed of uniform size and regular shape allows shrimp to swallow better and reduces feed waste, thereby improving shrimp growth efficiency and breeding benefits. By continuously adjusting the parameters of the crushing and granulation links, the production process can more accurately meet the target quality standards, thereby improving the controllability and stability of the production process.

[0130] It is understandable that for color defect results, refer to Figure 6 In step S140, the following steps may be included but not limited to:

[0131] Step S610: when there is a color defect result in the first quality assessment result, determining that the first processing link that may affect the shrimp material processing quality is the ingredient link, the conditioning link or the drying link;

[0132] Step S620: respectively obtaining initial parameters of batching in the batching process, initial parameters of conditioning in the conditioning process, and initial parameters of drying in the drying process.

[0133] Correspondingly, in step S150, the following steps may be included but not limited to:

[0134] Step S630: generating a second adjustment strategy according to the color defect result and a preset target quality standard;

[0135] Step S640: According to the second adjustment strategy, the batching initial parameters, the conditioning initial parameters and the drying initial parameters are adjusted to obtain the batching target parameters of the batching link, the conditioning target parameters of the conditioning link and the drying target parameters of the drying link.

[0136] In step S610 to step S620, when there is a color defect result in the first quality assessment result, it is necessary to consider which links in the shrimp feed processing process may affect the color. First, determine that the first processing link that may affect the processing quality of the shrimp feed is the batching link, the conditioning link or the drying link. In the batching link, the type and proportion of the raw materials will affect the final color of the shrimp feed. For example, different proportions of raw materials of different colors will cause the color of the shrimp feed to change. The conditioning link mainly involves operations such as humidification, heating and adding additives to the shrimp feed, which may cause the color of the shrimp feed to change. For example, excessive heating may darken the color of the shrimp feed. In the drying link, if factors such as temperature and time are not properly controlled, it will also cause problems with the color of the shrimp feed. For example, too high a drying temperature may cause the shrimp feed to turn yellow.

[0137] For the batching process, obtain the initial parameters of the batching. These parameters include the amount of various raw materials added, the color characteristics of the raw materials, etc. For the conditioning process, obtain the initial conditioning parameters, mainly the conditioning temperature, conditioning time, the type and amount of additives, etc. At the same time, for the drying process, obtain the initial drying parameters, such as drying temperature, drying time, ventilation volume, etc. For example, if the drying temperature is too high or the drying time is too long, the color of the shrimp material may change, and the ventilation volume will also affect the uniformity of drying, thereby affecting the consistency of color.

[0138] In step S630 to step S640, a second adjustment strategy is generated according to the color defect result and the preset target quality standard. If the color defect result is that the shrimp material is too dark in color, and the target quality standard requires a lighter color, the second adjustment strategy may include reducing the amount of darker raw materials added in the batching process; reducing the tempering temperature in the tempering process to avoid excessive heating that causes the color to darken; and appropriately reducing the drying temperature and shortening the drying time in the drying process to prevent the color from further deepening. If the shrimp material is uneven in color, the second adjustment strategy may be to strengthen the uniformity of raw material mixing in the batching process to ensure that the colors of the raw materials are evenly distributed; to ensure the uniformity of the tempering time and temperature in the tempering process to make the color changes of the shrimp material more consistent; and to optimize the ventilation system in the drying process to make the drying process more uniform, thereby improving the color consistency. According to the second adjustment strategy, the initial parameters of the batching are adjusted to obtain the batching target parameters of the batching process, the tempering target parameters of the tempering process, and the drying target parameters of the drying process. Reasonable adjustments to the parameters of the ingredients, conditioning and drying stages can effectively improve the color of the shrimp feed, making it more in line with quality standards, ensuring that the product color remains stable between different batches and meeting the market's requirements for consistent shrimp feed quality.

[0139] It is understandable that for texture defect features, refer to Figure 7 In step S140, the following steps may be included but not limited to:

[0140] Step S710: when there is a texture defect result in the first quality assessment result, determining that the first processing link that may affect the shrimp feed processing quality is a crushing link, a filtering link or a conditioning link;

[0141] Step S720: respectively obtaining initial crushing parameters of the crushing stage, initial filtering parameters of the filtering stage, and initial conditioning parameters of the conditioning stage.

[0142] Correspondingly, in step S150, the following steps may be included but not limited to:

[0143] Step S730: generating a third adjustment strategy according to the texture defect result and the preset target quality standard;

[0144] Step S740: According to the third adjustment strategy, the crushing initial parameters, the filtering initial parameters and the tempering initial parameters are adjusted to obtain the crushing target parameters of the crushing link, the filtering target parameters of the filtering link, and the tempering target parameters of the tempering link.

[0145] In step S710 to step S720, when there are texture defect results in the first quality assessment results, consider which links in the shrimp material processing process may have an impact on the texture. Determine the first processing link that may affect the processing quality of the shrimp material as the crushing link, the filtering link or the conditioning link. In the crushing link, the degree of crushing of the raw materials will affect the texture of the shrimp material. For example, the crushed particles are too large or too small, and the shape of the crushed particles is irregular, which may cause the final texture of the shrimp material to not meet the requirements. The filtering link is mainly to screen out particles that do not meet the size requirements. If the filtering effect is not good, the shrimp material will contain too many particles that are too large or too small, affecting the uniformity of the texture. The conditioning link changes the texture of the shrimp material through operations such as humidification, heating and adding additives, thereby affecting the texture. For example, if the amount of water added during the conditioning process is not appropriate or the additives are unevenly distributed, problems with the texture of the shrimp material will occur.

[0146] For the crushing process, the initial crushing parameters are obtained, including the speed of the crusher, crushing time, feed speed, mesh aperture, etc. For the filtration process, the initial filtration parameters are obtained, such as the mesh specifications of the filter, filtration speed, vibration frequency, etc. For the conditioning process, the initial conditioning parameters are obtained, mainly including the conditioning temperature, conditioning time, water addition amount, type and amount of additives, etc.

[0147] In step S730 to step S740, a third adjustment strategy is generated according to the texture defect result and the preset target quality standard. If the texture defect result is that the surface of the shrimp material is too rough, and the target quality standard requires a relatively smooth surface, the third adjustment strategy may include reducing the speed of the crusher in the crushing stage to make the crushed particles finer; replacing a finer screen in the filtering stage to remove larger particles; increasing the amount of water added in the conditioning stage to make the shrimp material softer, thereby improving the texture. If the texture of the shrimp material is uneven, the third adjustment strategy may be to adjust the feed speed and crushing time in the crushing stage to ensure that the crushed particle size is more uniform; optimize the filtering speed and vibration frequency in the filtering stage to ensure the consistency of the filtering effect; and strengthen the mixing uniformity of the additives in the conditioning stage to make the texture and texture of the shrimp material more uniform. According to the third adjustment strategy, the crushing initial parameters, the filtering initial parameters, and the conditioning initial parameters are adjusted to obtain the crushing target parameters of the crushing stage, the filtering target parameters of the filtering stage, and the conditioning target parameters of the conditioning stage, which can effectively improve the texture of the shrimp material and make it more in line with the quality standards.

[0148] Reference Figure 8 It is understandable that the intelligent control method for shrimp feed production provided in this application may also include but is not limited to the following steps:

[0149] Step S810: when adjusting parameters, obtaining the current timestamp;

[0150] Step S820: Generate change information according to the current timestamp, the adjusted first processing step, the initial parameter value before adjustment, the target parameter value after adjustment, and the corresponding first quality assessment result;

[0151] Step S830: Save the change information into a preset log text.

[0152] In step S810 to step S830, when the adjustment operation of the parameters related to shrimp feed production is started, the system will call the time acquisition module. This timestamp can accurately record the specific time when the parameter adjustment event occurs. Based on the current timestamp obtained, the key information involved in this adjustment is collected to generate change information to clarify which specific processing link it is, such as the crushing link. Record the initial parameter values ​​before the crushing link is adjusted, such as "crusher speed: 1500 rpm, screen aperture: 3 mm, feed speed: 5 kg / min", and record the target parameter values ​​after adjustment, "crusher speed: 1200 rpm, screen aperture: 2.5 mm, feed speed: 4 kg / min". At the same time, the corresponding first quality assessment result is also recorded, such as the quality assessment result: there are geometric defects and the shrimp feed particles are not round enough. Subsequently, through the file writing operation, the generated change information is written into the log text line by line according to the established encoding format to achieve persistent storage of the change information. The log text records in detail the timestamp of each parameter adjustment and the corresponding key information, so that production personnel can clearly trace past production adjustments at any time. For example, when a batch of shrimp feed is found to have quality problems, the parameter adjustment process of each processing link before and after the production of the batch can be traced back by checking the log, and what kind of parameter changes were made based on what quality assessment results, which helps to quickly locate the root cause of the problem.

[0153] In a second aspect, the present application also provides an electronic device, comprising: at least one memory, at least one processor and at least one program, the program is stored in the memory, and the processor executes one or more programs to implement the above-mentioned intelligent control method for shrimp feed production.

[0154] The electronic device can fully reflect the appearance information of the finished shrimp feed, such as particle shape, color, surface texture, etc., by using image acquisition devices at multiple angles to acquire initial images at the finished product discharge end of the shrimp feed production line. The multiple acquired initial images are preprocessed and feature extracted to extract multi-dimensional features. The extracted multi-dimensional features are input into a pre-trained shrimp feed quality assessment model, which is trained based on a large number of historical shrimp feed samples and can accurately output the first quality assessment result according to the input features. According to the first quality assessment result, combined with the production process knowledge and the internal logic of the model, the first processing link that may affect the shrimp feed processing quality is determined, and the initial parameters are adjusted again according to the first quality assessment result and the preset target quality standard to obtain the target parameters of the first processing link. The shrimp feed after adjusting the target parameters is re-imaged, analyzed and judged to obtain the second quality assessment result. Repeat this process, that is, adjust the parameters again according to the second quality assessment result until the second quality assessment result reaches the target quality standard. After each parameter adjustment, the adjustment effect can be quickly evaluated through image acquisition and analysis to form a closed-loop intelligent control system. The use of image acquisition and analysis technology and a well-trained quality assessment model avoids the subjectivity and instability of human eye detection. Through multi-dimensional feature extraction and the learning ability of the model, the quality of shrimp feed can be judged more accurately. Once the quality assessment results deviate, the processing link that may have problems can be quickly determined based on the model and production process knowledge, and the relevant parameters can be quickly adjusted using intelligent algorithms. This solves the problem of insufficient accuracy and stability of manual detection, reduces misjudgments, ensures that only qualified shrimp feed enters the market, guarantees shrimp farming results, reduces risks for farmers, and improves product production quality. At the same time, compared with manual trial-and-error troubleshooting and adjustments, it greatly shortens time and improves production efficiency.

[0155] The memory, as a non-transient computer-readable storage medium, can be used to store non-transient software programs, non-transient computer executable programs and signals, such as program instructions / signals corresponding to the processing module in the embodiment of the present application. The processor executes various functional applications and data processing by running the non-transient software programs, instructions and signals stored in the memory, that is, the intelligent control method for shrimp feed production of the above method embodiment is realized.

[0156] The memory may include a program storage area and a data storage area, wherein the program storage area may store an operating system, an application required for at least one function; the data storage area may store relevant data of the above-mentioned shrimp feed production intelligent control method, etc. In addition, the memory may include a high-speed random access memory, and may also include a non-transient memory, such as at least one disk storage device, a flash memory device or other non-transient solid-state storage device. In some embodiments, the memory may optionally include a memory remotely arranged relative to the processor, and these remote memories may be connected to the processing module via a network. Examples of the above-mentioned network include, but are not limited to, the Internet, an intranet, a local area network, a mobile communication network, and combinations thereof.

[0157] One or more signals are stored in the memory, and when executed by one or more processors, the intelligent control method for shrimp feed production in any of the above method embodiments is executed.

[0158] In a third aspect, an embodiment of the present application provides a computer-readable storage medium, which stores a computer program. The computer program is executed by one or more processors, enabling the one or more processors to execute the intelligent control method for shrimp feed production in the above method embodiment.

[0159] The device embodiments described above are merely illustrative, wherein the units described as separate components may or may not be physically separated, and the components shown as units may or may not be physical units, i.e., may be located in one place, or may be distributed over multiple network units. Some or all of the units may be selected according to actual needs to achieve the purpose of the solution of this embodiment.

[0160] Through the description of the above embodiments, it will be appreciated by those skilled in the art that all or some of the steps and systems in the disclosed methods above can be implemented as software, firmware, hardware and appropriate combinations thereof. Some physical components or all physical components can be implemented as software executed by a processor, such as a central processing unit, a digital signal processor or a microprocessor, or implemented as hardware, or implemented as an integrated circuit, such as an application-specific integrated circuit. Such software can be distributed on a computer-readable medium, and the computer-readable medium can include a computer storage medium (or a non-transitory medium) and a communication medium (or a temporary medium). As known to those skilled in the art, the term computer storage medium includes volatile and non-volatile, removable and non-removable media implemented in any method or technology for storing information (such as a computer-readable signal, a data structure, a program module or other data). Computer storage media include, but are not limited to, RAM, ROM, EEPROM, flash memory or other memory technology, CD-ROM, digital versatile disk (DVD) or other optical disk storage, magnetic cassette, magnetic tape, disk storage or other magnetic storage device, or any other medium that can be used to store desired information and can be accessed by a computer. Furthermore, it is well known to those skilled in the art that communication media generally embodies computer-readable signals, data structures, program modules, or other data in a modulated data signal such as a carrier wave or other transport mechanism, and may include any information delivery media.

[0161] It should be understood that in the present application, "at least one (item)" means one or more, and "plurality" means two or more. "And / or" is used to describe the association relationship of associated objects, indicating that three relationships may exist. For example, "A and / or B" can mean: only A exists, only B exists, and A and B exist at the same time, where A and B can be singular or plural. The character " / " generally indicates that the objects associated before and after are in an "or" relationship. "At least one of the following" or similar expressions refers to any combination of these items, including any combination of single or plural items. For example, at least one of a, b or c can mean: a, b, c, "a and b", "a and c", "b and c", or "a and b and c", where a, b, c can be single or multiple.

[0162] In the several embodiments provided in the present application, it should be understood that the disclosed devices and methods can be implemented in other ways. For example, the device embodiments described above are only schematic. For example, the division of the above units is only a logical function division. There may be other division methods in actual implementation, such as multiple units or components can be combined or integrated into another system, or some features can be ignored or not executed. Another point is that the mutual coupling or direct coupling or communication connection shown or discussed can be through some interfaces, indirect coupling or communication connection of devices or units, which can be electrical, mechanical or other forms.

[0163] The units described above as separate components may or may not be physically separated, and the components shown as units may or may not be physical units, that is, they may be located in one place or distributed on multiple network units. Some or all of the units may be selected according to actual needs to achieve the purpose of the solution of this embodiment.

[0164] In addition, each functional unit in each embodiment of the present application may be integrated into one processing unit, or each unit may exist physically separately, or two or more units may be integrated into one unit. The above-mentioned integrated unit may be implemented in the form of hardware or in the form of software functional units.

[0165] If the integrated unit is implemented in the form of a software functional unit and sold or used as an independent product, it can be stored in a computer-readable storage medium. Based on this understanding, the technical solution of the present application is essentially or the part that contributes to the prior art or all or part of the technical solution can be embodied in the form of a software product, and the computer software product is stored in a storage medium, including multiple instructions to enable a computer device (which can be a personal computer, server, or network device, etc.) to execute all or part of the steps of the methods of various embodiments of the present application. The aforementioned storage medium includes: U disk, mobile hard disk, read-only memory (Read-Only Memory, referred to as ROM), random access memory (Random Access Memory, referred to as RAM), disk or optical disk and other media that can store programs.

[0166] The embodiments of the present application are described in detail above in conjunction with the accompanying drawings, but the present application is not limited to the above embodiments, and various changes can be made within the knowledge scope of ordinary technicians in the relevant technical field without departing from the purpose of the present application.

Claims

1. An intelligent control method for shrimp feed production, characterized in that: include: Acquire initial images at multiple angles at the finished product discharging end of the shrimp feed production line; Performing preprocessing and feature extraction processing on the plurality of initial images to obtain multi-dimensional features; Inputting the multi-dimensional features into a pre-trained shrimp feed quality assessment model to obtain a first quality assessment result; wherein the shrimp feed quality assessment model is obtained by training with a plurality of historical shrimp feed samples; According to the first quality assessment result, determining the first processing link that may affect the shrimp feed processing quality, and obtaining initial parameters of the first processing link; The initial parameters are adjusted again according to the first quality assessment result and the preset target quality standard to obtain the target parameters of the first processing link; The shrimp feed after the target parameters are adjusted is subjected to image acquisition, analysis and determination processing again to obtain a second quality assessment result, and parameters are adjusted according to the second quality assessment result until the second quality assessment result reaches the target quality standard.

2. The intelligent control method for shrimp feed production according to claim 1, characterized in that: The multi-dimensional features include geometric features, color features and texture features; The preprocessing and feature extraction processing of the plurality of initial images to obtain multi-dimensional features includes: Removing high-frequency noise from the initial image and suppressing low-frequency interference from the initial image to obtain a first image; Gray-scale the first image to obtain a target image; Extracting edge information of the target image and obtaining geometric features according to the edge information; Performing color conversion and statistical processing on the initial image to obtain color features; Texture features are obtained according to the target images at multiple different angles.

3. The intelligent control method for shrimp feed production according to claim 2, characterized in that: The historical shrimp feed samples include sample images and actual defect type labels corresponding to each sample image; The shrimp feed quality assessment model is obtained by the following steps: According to the sample image, a geometric sample feature, a color sample feature and a texture sample feature are obtained; The geometric sample features, the color sample features and the texture sample features are respectively input into an initial blanking quality assessment model for classification processing to obtain a predicted defect type label; Obtaining a loss value according to the actual defect type label and the predicted defect type label; According to the loss value, the parameters of the shrimp feed quality assessment model are adjusted until the loss value reaches a preset loss threshold.

4. The intelligent control method for shrimp feed production according to claim 2, characterized in that: The multi-dimensional features are input into a pre-trained shrimp feed quality assessment model to obtain a first quality assessment result, including: Inputting the geometric features, the color features, and the texture features into a pre-trained quality assessment model; Extracting the target perimeter, target area and target circularity according to the geometric features; Obtaining a geometric defect result according to the target perimeter and a preset perimeter threshold range, the target area and a preset area threshold range, the target circularity and a preset circularity threshold range; According to the color characteristics, a first color intensity, a second color intensity and a third color intensity are obtained; Obtaining a color defect result according to the first color intensity and a preset first color range, the second color intensity and a preset second color range, and the third color intensity and a preset third color range; Determining a target smoothness of the shrimp feed surface according to the texture characteristics; Obtaining a texture defect result according to the target smoothness and a preset smoothness threshold range; A first quality assessment result is generated according to the geometric defect result, the color defect result and the texture defect result.

5. The intelligent control method for shrimp feed production according to claim 4, characterized in that: The first processing link that may affect the shrimp feed processing quality is determined according to the first quality assessment result, and the initial parameters of the first processing link are obtained, including: When there is a geometric defect result in the first quality assessment result, determining that the first processing link that may affect the shrimp feed processing quality is a crushing link or a granulation link; respectively obtaining the initial parameters of the pulverizing step and the initial parameters of the granulating step; Correspondingly, the initial parameters are adjusted again according to the first quality assessment result and the preset target quality standard to obtain the target parameters of the first processing link, including: Generating a first adjustment strategy according to the geometric defect result and a preset target quality standard; According to the first adjustment strategy, the crushing initial parameters and the granulation initial parameters are adjusted to obtain crushing target parameters of the crushing link and granulation target parameters of the granulation link.

6. The intelligent control method for shrimp feed production according to claim 4, characterized in that: The first processing link that may affect the shrimp feed processing quality is determined according to the first quality assessment result, and the initial parameters of the first processing link are obtained, including: When there is a color defect result in the first quality assessment result, determining that the first processing link that may affect the shrimp feed processing quality is the ingredient link, the conditioning link or the drying link; Respectively obtaining initial parameters of batching in the batching process, initial parameters of conditioning in the conditioning process, and initial parameters of drying in the drying process; Correspondingly, the initial parameters are adjusted again according to the first quality assessment result and the preset target quality standard to obtain the target parameters of the first processing link, including: generating a second adjustment strategy according to the color defect result and a preset target quality standard; According to the second adjustment strategy, the batching initial parameters, the conditioning initial parameters and the drying initial parameters are adjusted to obtain the batching target parameters of the batching link, the conditioning target parameters of the conditioning link and the drying target parameters of the drying link.

7. The intelligent control method for shrimp feed production according to claim 4, characterized in that: The first processing link that may affect the shrimp feed processing quality is determined according to the first quality assessment result, and the initial parameters of the first processing link are obtained, including: When there is a texture defect result in the first quality assessment result, determining that the first processing link that may affect the shrimp feed processing quality is a crushing link, a filtering link or a conditioning link; Respectively obtaining initial crushing parameters of the crushing step, initial filtering parameters of the filtering step, and initial conditioning parameters of the conditioning step; Correspondingly, the initial parameters are adjusted again according to the first quality assessment result and the preset target quality standard to obtain the target parameters of the first processing link, including: generating a third adjustment strategy according to the texture defect result and a preset target quality standard; According to the third adjustment strategy, the crushing initial parameters, the filtering initial parameters and the tempering initial parameters are adjusted to obtain the crushing target parameters of the crushing link, the filtering target parameters of the filtering link and the tempering target parameters of the tempering link.

8. The method for intelligently controlling shrimp feed production according to claim 1, characterized in that: Also includes: When adjusting parameters, get the current timestamp; Generate change information according to the current timestamp, the adjusted first processing link, the initial parameter value before adjustment, the target parameter value after adjustment, and the corresponding first quality assessment result; The change information is saved in a preset log text.

9. An electronic device, characterized in that: include: at least one memory; at least one processor; at least one program; The program is stored in the memory, and the processor executes at least one of the programs to implement the intelligent control method for shrimp feed production according to any one of claims 1 to 8.

10. A computer-readable storage medium, characterized in that: The computer-readable storage medium stores computer-executable signals, and the computer-executable signals are used to execute the intelligent control method for shrimp feed production according to any one of claims 1 to 8.

Images