Organic microelement water-soluble fertilizer and preparation method thereof

Abstract

The invention relates to the field of preparation of water-soluble fertilizers, in particular to an organic microelement water-soluble fertilizer and a preparation method thereof. The method comprises the following steps: fermenting wine brewing vinasse to obtain vinasse fermentation liquor; evaporating and concentrating the vinasse fermentation liquor to obtain concentrated liquor; the concentrated solution is added into a reaction kettle to be stirred, nitrogen, phosphorus pentoxide, potassium oxide and a suspending agent are sequentially added in the stirring process to obtain a mixed solution, and the stirring time is 2 hours; grinding the mixed solution by using a shear type colloid mill to obtain a liquid fertilizer; and conveying the liquid fertilizer to a filling line for filling to obtain the finished water-soluble fertilizer. Further, the collected solution state image can be subjected to image analysis by using a computer vision technology based on deep learning, and whether each component in the mixed solution is uniformly distributed or not is judged based on color distribution characteristics and texture distribution characteristics of the solution state, so that the automatic and intelligent control of the grinding process is realized.

Description

Technical Field

[0001] The present application relates to the field of water-soluble fertilizer preparation, and in particular to an organic trace element water-soluble fertilizer and a preparation method thereof. Background Art

[0002] With the rapid development of modern agriculture, fertilizer is one of the key factors for crop growth, and its type and quality directly affect the yield and quality of crops. Although traditional fertilizers such as nitrogen fertilizer, phosphorus fertilizer and potassium fertilizer can meet the basic nutritional needs of crops, they are obviously insufficient in trace element supplementation and nutrient utilization. In recent years, with the deepening of the understanding of crop nutritional needs, water-soluble fertilizers rich in organic trace elements have gradually received widespread attention in the agricultural field.

[0003] As a byproduct rich in organic matter, wine lees contain a large amount of organic matter, trace elements and other nutrients, and are an ideal raw material for preparing organic trace element water-soluble fertilizers. However, in the preparation process of organic trace element water-soluble fertilizers, traditional preparation methods usually rely on the operator's experience or preset parameters to control the reaction conditions and the stirring and grinding process. They are too dependent on the operator's professional level, have certain subjectivity and inaccuracy, and have a low level of automation. Large-scale water-soluble fertilizer production requires a lot of human resources. Therefore, an optimized preparation method of organic trace element water-soluble fertilizers is expected to achieve precise control and optimization of the preparation process. Summary of the invention

[0004] This application is made in view of the above problems. One object of this application is to provide an organic trace element water-soluble fertilizer and a preparation method thereof.

[0005] The embodiment of the present application provides a method for preparing an organic trace element water-soluble fertilizer, which comprises:

[0006] fermenting wine lees to obtain wine lees fermentation liquid;

[0007] The wine lees fermentation liquid is evaporated and concentrated to obtain a concentrated liquid;

[0008] Add the concentrated solution into a reaction kettle and stir, and add nitrogen, phosphorus pentoxide, potassium oxide and a suspending agent in sequence during the stirring process to obtain a mixed solution, wherein the stirring time is 2 hours;

[0009] Grinding the mixed solution using a shearing colloid mill to obtain liquid fertilizer;

[0010] The liquid fertilizer is transported to a filling line for filling to obtain a finished water-soluble fertilizer.

[0011] For example, according to the method for preparing the organic trace element water-soluble fertilizer of the embodiment of the present application, the mixed liquid is ground using a shearing colloid mill to obtain liquid fertilizer, comprising:

[0012] The solution state of the mixed solution is detected, and the grinding is stopped when N, P2O5, and K2O in the mixed solution are uniformly suspended in the solution.

[0013] For example, according to the method for preparing the organic trace element water-soluble fertilizer of the embodiment of the present application, the solution state of the mixed solution is detected, and the grinding is stopped when N, P2O5, and K2O in the mixed solution are uniformly suspended in the solution, comprising:

[0014] Acquiring a current solution state image collected by an underwater camera deployed in the mixed liquid;

[0015] Extracting solution color distribution features of the current solution state image to obtain a global solution color distribution feature vector;

[0016] Extracting solution texture distribution features of the current solution state image to obtain a global solution texture distribution feature vector;

[0017] Inputting the global solution color distribution feature vector and the global solution texture distribution feature vector into a balanced threshold feature vector adaptive fusion module to obtain a solution color-texture global distribution fusion feature vector;

[0018] Whether to stop stirring is determined based on the solution color-texture global distribution fusion feature vector.

[0019] For example, according to the method for preparing organic trace element water-soluble fertilizer according to an embodiment of the present application, extracting the solution color distribution characteristics of the current solution state image to obtain a global solution color distribution feature vector includes:

[0020] Calculating a color histogram of the current solution state image to obtain a solution state color histogram;

[0021] The solution state color histogram is input into a solution color feature extractor based on the first ViT model to obtain the global solution color distribution feature vector.

[0022] For example, according to the method for preparing organic trace element water-soluble fertilizer according to an embodiment of the present application, extracting the solution texture distribution characteristics of the current solution state image to obtain a global solution texture distribution feature vector includes:

[0023] Calculating a histogram of directional gradients of the current solution state image to obtain a histogram of directional gradients of the solution state;

[0024] The solution state directional gradient histogram is input into a solution texture feature extractor based on a second ViT model to obtain the global solution texture distribution feature vector.

[0025] For example, according to the method for preparing organic trace element water-soluble fertilizer in an embodiment of the present application, the global solution color distribution feature vector and the global solution texture distribution feature vector are input into a balanced threshold feature vector adaptive fusion module to obtain a solution color-texture global distribution fusion feature vector, including:

[0026] The global solution color distribution feature vector and the global solution texture distribution feature vector are fused through a multi-dimensional fusion module to obtain a first solution color-texture feature distribution fusion representation vector, a second solution color-texture feature distribution fusion representation vector, and a third solution color-texture feature distribution fusion representation vector;

[0027] Respectively calculating the equalization threshold values ​​of the first solution color-texture feature distribution fusion representation vector, the second solution color-texture feature distribution fusion representation vector, and the third solution color-texture feature distribution fusion representation vector to obtain a first equalization threshold value, a second equalization threshold value, and a third equalization threshold value;

[0028] Determine a first weight parameter and a second weight parameter based on the first equalization threshold value, the second equalization threshold value, and the third equalization threshold value, wherein the first weight parameter is an average of the first equalization threshold value, the second equalization threshold value, and the third equalization threshold value, and the second weight parameter is a difference between one and the first weight parameter;

[0029] The global solution color distribution feature vector is weighted position by position with the first weight parameter to obtain a weighted global solution color distribution feature vector, and the global solution texture distribution feature vector is weighted position by position with the second weight parameter to obtain a weighted global solution texture distribution feature vector;

[0030] The weighted global solution color distribution feature vector and the weighted global solution texture distribution feature vector are added together to obtain the solution color-texture global distribution fusion feature vector.

[0031] For example, according to the method for preparing an organic trace element water-soluble fertilizer in an embodiment of the present application, the global solution color distribution feature vector and the global solution texture distribution feature vector are subjected to a multi-dimensional fusion module to obtain a first solution color-texture feature distribution fusion representation vector, a second solution color-texture feature distribution fusion representation vector, and a third solution color-texture feature distribution fusion representation vector, including:

[0032] Cascading the global solution color distribution feature vector and the global solution texture distribution feature vector to obtain the first solution color-texture feature distribution fusion representation vector;

[0033] Adding the global solution color distribution feature vector and the global solution texture distribution feature vector by position to obtain the second solution color-texture feature distribution fusion representation vector;

[0034] The global solution color distribution feature vector and the global solution texture distribution feature vector are multiplied by position points to obtain the third solution color-texture feature distribution fusion representation vector.

[0035] For example, according to the method for preparing an organic trace element water-soluble fertilizer according to an embodiment of the present application, the first solution color-texture feature distribution fusion representation vector, the second solution color-texture feature distribution fusion representation vector, and the third solution color-texture feature distribution fusion representation vector are respectively calculated to obtain a first equilibrium threshold value, a second equilibrium threshold value, and a third equilibrium threshold value, including:

[0036] Multiplying the first solution color-texture feature distribution fusion representation vector by a first predetermined transformation vector to obtain a first threshold scoring coefficient;

[0037] The first threshold scoring coefficient and the first bias parameter are added together and then passed through a sigmoid activation function to obtain the first equalization threshold value.

[0038] For example, according to the method for preparing organic trace element water-soluble fertilizer in an embodiment of the present application, determining whether to stop stirring based on the solution color-texture global distribution fusion feature vector includes:

[0039] The solution color-texture global distribution fusion feature vector is input into a classifier-based stirring controller to obtain a control instruction, wherein the control instruction is used to indicate whether to stop stirring.

[0040] An embodiment of the present application further provides an organic trace element water-soluble fertilizer, wherein the organic trace element water-soluble fertilizer is prepared by any of the above-described methods for preparing the organic trace element water-soluble fertilizer.

[0041] According to the organic trace element water-soluble fertilizer and its preparation method of the embodiment of the present application, it utilizes computer vision technology based on deep learning to perform image analysis on the collected solution state images, and judges whether the components in the mixed solution are evenly distributed based on the color distribution characteristics and texture distribution characteristics of the solution state, thereby realizing automated and intelligent control of the grinding process. BRIEF DESCRIPTION OF THE DRAWINGS

[0042] In order to more clearly illustrate the technical solutions of the embodiments of the present application, the drawings of the embodiments of the present application are briefly introduced below. Obviously, the drawings described below only relate to some embodiments of the present application, and are not intended to limit the present application.

[0043] Figure 1 The application framework schematic diagram of the method for preparing the organic trace element water-soluble fertilizer in the embodiment of the present application is shown;

[0044] Figure 2 A flow chart of a method for preparing an organic trace element water-soluble fertilizer in an embodiment of the present application is shown;

[0045] Figure 3 A flowchart of sub-step S540 of the method for preparing organic trace element water-soluble fertilizer in an embodiment of the present application is shown;

[0046] Figure 4 A flow chart of sub-step S544 of the method for preparing organic trace element water-soluble fertilizer in an embodiment of the present application is shown;

[0047] Figure 5 A schematic diagram of the structure of the preparation system of the organic trace element water-soluble fertilizer in the embodiment of the present application is shown; and

[0048] Figure 6 A schematic diagram of a storage medium according to an embodiment of the present application is shown. DETAILED DESCRIPTION

[0049] The technical solutions in the embodiments of the present application will be described clearly and completely below in conjunction with the accompanying drawings. Obviously, the described embodiments are only some embodiments of the present application, not all embodiments. Based on the embodiments of the present application, all other embodiments obtained by ordinary technicians in this field without creative work also fall within the scope of protection of the present application.

[0050] The terms used in this specification are those common terms currently widely used in the art in consideration of the functions of the present application, but these terms may vary according to the intention of a person of ordinary skill in the art, precedents, or new technologies in the art. In addition, specific terms may be selected by the applicant, and in this case, their detailed meanings will be described in the detailed description of the present application. Therefore, the terms used in the specification should not be understood as simple names, but rather as a general description based on the meaning of the terms and the present application.

[0051] Although the present application makes various references to certain modules in the system according to the embodiments of the present application, any number of different modules can be used and run on the user terminal and / or server. The modules are only illustrative, and different aspects of the system and method can use different modules.

[0052] Flowcharts are used in the present application to illustrate the operations performed by the system according to the embodiments of the present application. It should be understood that the preceding or following operations are not necessarily performed accurately in order. On the contrary, various steps may be processed in reverse order or simultaneously as required. At the same time, other operations may also be added to these processes, or a certain step or several steps of operations may be removed from these processes.

[0053] Figure 1 A schematic diagram of the application architecture of the method for preparing organic trace element water-soluble fertilizer in an embodiment of the present application is shown, including a server 100 and a terminal device 200.

[0054] The terminal device 200 and the server 100 can be connected via the Internet to achieve mutual communication. Optionally, the above-mentioned Internet uses standard communication technology and / or protocol. The Internet is usually the Internet, but it can also be any network, including but not limited to any combination of a local area network (LAN), a metropolitan area network (MAN), a wide area network (WAN), a mobile, wired or wireless network, a dedicated network or a virtual private network. In some embodiments, the data exchanged through the network is represented by technologies and / or formats including Hyper Text Markup Language (HTML), Extensible Markup Language (XML), etc. In addition, conventional encryption technologies such as Secure Socket Layer (SSL), Transport Layer Security (TLS), Virtual Private Network (VPN), Internet Protocol Security (IPsec) can also be used to encrypt all or some links. In other embodiments, customized and / or dedicated data communication technologies can also be used to replace or supplement the above-mentioned data communication technologies.

[0055] The server 100 can provide various network services for the terminal device 200, wherein the server 100 can be a single server, a server cluster consisting of several servers, or a cloud computing center. Specifically, the server 100 may include a processor 110 (Center Processing Unit, CPU), a memory 120, an input device 130, and an output device 140, etc. The input device 130 may include a keyboard, a mouse, a touch screen, etc., and the output device 140 may include a display device, such as a liquid crystal display (Liquid Crystal Display, LCD), a cathode ray tube (Cathode Ray Tube, CRT), etc.

[0056] The memory 120 may include a read-only memory (ROM) and a random access memory (RAM), and provides the processor 110 with program instructions and data stored in the memory 120. In an embodiment of the present application, the memory 120 may be used to store a program of the method for preparing the organic trace element water-soluble fertilizer in an embodiment of the present application.

[0057] The processor 110 calls the program instructions stored in the memory 120, and the processor 110 is used to execute the steps of the preparation method of any organic trace element water-soluble fertilizer in the embodiments of the present application according to the obtained program instructions.

[0058] In addition, the application architecture diagram in the embodiment of the present application is intended to more clearly illustrate the technical solution in the embodiment of the present application, and does not constitute a limitation on the technical solution provided in the embodiment of the present application. Of course, for other application architectures and business applications, the technical solution provided in the embodiment of the present application is also applicable to similar problems.

[0059] The following is a non-limiting description of the preparation method of the organic trace element water-soluble fertilizer provided according to at least one embodiment of the present application through several examples or embodiments. As described below, different features in these specific examples or embodiments can be combined with each other without conflicting with each other to obtain new examples or embodiments, and these new examples or embodiments also belong to the scope of protection of this application.

[0060] The present application provides a method for preparing an organic trace element water-soluble fertilizer, such as Figure 2As shown, it specifically includes the following steps: S510, fermenting wine lees to obtain wine lees fermentation liquid; S520, evaporating and concentrating the wine lees fermentation liquid to obtain a concentrated solution; S530, adding the concentrated solution into a reactor for stirring, and adding nitrogen, phosphorus pentoxide, potassium oxide and a suspending agent in sequence during the stirring process to obtain a mixed solution, wherein the stirring time is 2 hours; S540, grinding the mixed solution with a shearing colloid mill to obtain a liquid fertilizer; S550, conveying the liquid fertilizer to a filling line for filling to obtain a finished water-soluble fertilizer.

[0061] Specifically, in the preparation process of water-soluble fertilizer, in order to ensure that various components (such as nitrogen, phosphorus pentoxide, potassium oxide, etc.) are evenly distributed in the liquid fertilizer and avoid uneven plant growth caused by local aggregation of components, the mixed solution needs to be ground to improve the particle size distribution and viscosity of the solution, increase the dissolution rate and solubility of the fertilizer in water, and ensure the fluidity of the liquid fertilizer so that it is easy to fill. However, during the grinding process, if the grinding time is too short, the particle size distribution of the solution may be uneven; and if the grinding time is too long, the interaction between the particles is enhanced, which may cause the particles to re-aggregate and form larger precipitates, thereby affecting the solubility of the fertilizer and the nutrient absorption of the plant. Therefore, in order to ensure the quality of the fertilizer, it is necessary to accurately control the grinding time according to the state of the solution, and stop grinding when N, P2O5, and K2O in the mixed solution are evenly suspended in the solution to achieve the best dissolution effect and nutrient release. Accordingly, in step S540, the mixed solution is ground using a shearing colloid mill to obtain liquid fertilizer, including: detecting the solution state of the mixed solution, and stopping the grinding when N, P2O5, and K2O in the mixed solution are uniformly suspended in the solution.

[0062] However, the traditional grinding time control method mainly relies on the operator's manual experience or preset time threshold, has certain subjectivity and inaccuracy, and has a low level of automation. In response to this technical problem, the technical concept of the present application is to use an underwater camera to collect the state image of the solution in real time, and use computer vision technology based on deep learning to perform image analysis on the collected solution state image, and judge whether the components in the mixed solution are evenly distributed based on the color distribution characteristics and texture distribution characteristics of the solution state, so as to accurately control the grinding time. In this way, the automation and intelligent control of the grinding process can be achieved, avoiding unnecessary energy waste and extended production time, thereby improving the preparation efficiency and product quality of organic trace element water-soluble fertilizers.

[0063] Based on this, Figure 3As shown, the solution state of the mixed solution is detected, and grinding is stopped when N, P2O5, and K2O in the mixed solution are uniformly suspended in the solution, including: S541, obtaining a current solution state image collected by an underwater camera deployed in the mixed solution; S542, extracting solution color distribution characteristics of the current solution state image to obtain a global solution color distribution feature vector; S543, extracting solution texture distribution characteristics of the current solution state image to obtain a global solution texture distribution feature vector; S544, inputting the global solution color distribution feature vector and the global solution texture distribution feature vector into an equalization threshold feature vector adaptive fusion module to obtain a solution color-texture global distribution fusion feature vector; S545, determining whether to stop stirring based on the solution color-texture global distribution fusion feature vector.

[0064] Specifically, in the technical solution of the present application, first, the current solution state image collected by the underwater camera deployed in the mixed liquid is obtained. It should be understood that by deploying the underwater camera in the mixed liquid, the particle distribution, suspension state and possible precipitation or stratification phenomenon in the mixed liquid can be observed in real time and intuitively, and the subtle changes of the solution during the grinding process can be captured, thereby providing important image basis for the accurate analysis of the solution state, which helps to achieve accurate control of the grinding process according to the solution state.

[0065] Next, considering that the distribution of N, P2O5, K2O and other components in the solution is usually reflected in the color distribution of the solution. Therefore, in order to obtain the color distribution information of the solution, in the technical solution of the present application, the color histogram of the current solution state image is further calculated, and the distribution frequency of different colors in the image is counted to obtain the solution state color histogram, so as to quantify and characterize the color distribution of the solution, thereby reflecting the uniformity of each component in the solution.

[0066] Then, since the solution state color histogram only provides statistical information about the solution color distribution, in order to further consider the spatial relationship of the solution color distribution and learn deeper global distribution characteristics of the solution color from the solution state color histogram, in the technical solution of the present application, a solution color feature extractor based on the first ViT model is used to process the solution state color histogram. Specifically, the ViT model (Visual Transformer) is a deep learning model based on the Transformer architecture, which is good at capturing global structural information in images. Here, by dividing the solution state color histogram into a sequence of image blocks and inputting it into the ViT model, the ViT model performs a global correlation analysis based on the self-attention mechanism to learn the global pattern of the solution color distribution and obtain the global solution color distribution feature vector, so as to more comprehensively describe the distribution uniformity of each component in the solution.

[0067] Accordingly, in step S542, the solution color distribution characteristics of the current solution state image are extracted to obtain a global solution color distribution feature vector, including: calculating the color histogram of the current solution state image to obtain a solution state color histogram; inputting the solution state color histogram into a solution color feature extractor based on the first ViT model to obtain the global solution color distribution feature vector.

[0068] At the same time, considering that the component particles in the solution not only differ in color, their visual texture features also have an important characterization effect on the uniformity of particle distribution in the solution. Therefore, in the technical solution of the present application, the directional gradient histogram of the current solution state image is further calculated to obtain the solution state directional gradient histogram. Among them, the directional gradient histogram (HOG) is an effective texture feature description method that can capture the texture structure and local shape information in the image. By calculating the directional gradient histogram of the current solution state image, the edge information in the image, such as the outline, shape and possible particle aggregation information of the suspended particles, can be captured, thereby assisting in judging whether the distribution of particles in the solution is uniform. Then, a solution texture feature extractor based on the second ViT model is used to perform feature mining on the solution state directional gradient histogram. Similarly, the powerful global feature learning ability of the ViT model is used to perform global feature modeling on the directional gradient histogram to capture the complex pattern of solution texture distribution, obtain the global solution texture distribution feature vector, and further enhance the ability to judge the uniformity of particles in the solution.

[0069] Correspondingly, in step S543, the solution texture distribution characteristics of the current solution state image are extracted to obtain a global solution texture distribution feature vector, including: calculating the directional gradient histogram of the current solution state image to obtain a solution state directional gradient histogram; inputting the solution state directional gradient histogram into a solution texture feature extractor based on a second ViT model to obtain the global solution texture distribution feature vector.

[0070] After the above processing, the global solution color distribution feature vector and the global solution texture distribution feature vector respectively provide different information dimensions about the uniformity of the distribution of suspended particles in the solution. By fusing the two, the uniformity of the distribution of suspended particles in the solution can be more comprehensively characterized. At the same time, considering that as the grinding process proceeds, the contributions of the color characteristics and texture characteristics of the solution to the judgment of the uniformity of the solution are not completely consistent. Therefore, in the technical solution of the present application, an adaptive fusion module of a balanced threshold feature vector is introduced to perform adaptive balanced fusion processing on the global solution color distribution feature vector and the global solution texture distribution feature vector to obtain a solution color-texture global distribution fusion feature vector. Specifically, the adaptive fusion module of the balanced threshold feature vector adopts an adaptive weight allocation strategy to dynamically adjust the fusion weights of the two based on the correlation information between the global color characteristics and the global texture characteristics of the solution to optimize the feature fusion effect and generate a more comprehensive and accurate description of the solution state, thereby improving the accuracy of judging the timing of grinding completion.

[0071] Accordingly, in step S544, if Figure 4As shown, the global solution color distribution feature vector and the global solution texture distribution feature vector are input into the equalization threshold feature vector adaptive fusion module to obtain the solution color-texture global distribution fusion feature vector, including: S5441, the global solution color distribution feature vector and the global solution texture distribution feature vector are passed through a multi-dimensional fusion module to obtain a first solution color-texture feature distribution fusion representation vector, a second solution color-texture feature distribution fusion representation vector and a third solution color-texture feature distribution fusion representation vector; S5442, the equalization threshold values ​​of the first solution color-texture feature distribution fusion representation vector, the second solution color-texture feature distribution fusion representation vector and the third solution color-texture feature distribution fusion representation vector are calculated respectively to obtain a first equalization threshold value, a second equalization threshold value and a third equalization threshold value; S544 43. Determine a first weight parameter and a second weight parameter based on the first equalization threshold value, the second equalization threshold value and the third equalization threshold value, wherein the first weight parameter is the average of the first equalization threshold value, the second equalization threshold value and the third equalization threshold value, and the second weight parameter is the difference between one and the first weight parameter; S5444. Perform position-by-position weighting on the global solution color distribution feature vector with the first weight parameter to obtain a weighted global solution color distribution feature vector, and perform position-by-position weighting on the global solution texture distribution feature vector with the second weight parameter to obtain a weighted global solution texture distribution feature vector; S5445. Add corresponding elements of the weighted global solution color distribution feature vector and the weighted global solution texture distribution feature vector to obtain the solution color-texture global distribution fusion feature vector.

[0072] Among them, in step S5441, the global solution color distribution feature vector and the global solution texture distribution feature vector are passed through a multi-dimensional fusion module to obtain a first solution color-texture feature distribution fusion representation vector, a second solution color-texture feature distribution fusion representation vector and a third solution color-texture feature distribution fusion representation vector, including: cascading the global solution color distribution feature vector and the global solution texture distribution feature vector to obtain the first solution color-texture feature distribution fusion representation vector; adding the global solution color distribution feature vector and the global solution texture distribution feature vector by position to obtain the second solution color-texture feature distribution fusion representation vector; multiplying the global solution color distribution feature vector and the global solution texture distribution feature vector by position point to obtain the third solution color-texture feature distribution fusion representation vector.

[0073] Among them, in step S5442, the equalization threshold values ​​of the first solution color-texture feature distribution fusion representation vector, the second solution color-texture feature distribution fusion representation vector and the third solution color-texture feature distribution fusion representation vector are calculated respectively to obtain the first equalization threshold value, the second equalization threshold value and the third equalization threshold value, including: using a first predetermined transformation vector to multiply the first solution color-texture feature distribution fusion representation vector to obtain a first threshold scoring coefficient; adding the first threshold scoring coefficient and the first bias parameter and then passing through a sigmoid activation function to obtain the first equalization threshold value.

[0074] In a specific example, the global solution color distribution feature vector and the global solution texture distribution feature vector are input into the equalization threshold feature vector adaptive fusion module to obtain the solution color-texture global distribution fusion feature vector, including: processing the global solution color distribution feature vector and the global solution texture distribution feature vector with the following adaptive fusion formula to obtain the solution color-texture global distribution fusion feature vector, wherein the adaptive fusion formula is:

[0075]

[0076] Wherein, v1 is the global solution color distribution feature vector, v2 is the global solution texture distribution feature vector, W1, W2 and W3 are respectively the first predetermined transformation vector, the second predetermined transformation vector and the third predetermined transformation vector, b1, b2 and b3 are respectively the first bias parameter, the second bias parameter and the third bias parameter, sigmoid represents the activation function, concat(·,·) represents the cascade processing, represents positional addition, ⊙ represents positional point multiplication, t1, t2 and t3 are the first, second and third equalization thresholds respectively, and t1, t2, t3∈[0,1], v c is the color-texture global distribution fusion feature vector of the solution.

[0077] Then, the solution color-texture global distribution fusion feature vector is input into a classifier-based stirring controller to obtain a control instruction, and the control instruction is used to indicate whether to stop grinding. In the technical solution of the present application, the stirring controller uses a trained classifier model to perform feature discrimination on the solution color-texture global distribution fusion feature vector, and judges whether the current solution state meets the condition for stopping stirring, that is, whether N, P2O5, and K2O in the mixed solution are evenly suspended in the solution, based on the feature information in the solution color-texture global distribution fusion feature vector, thereby outputting a control instruction on whether to stop grinding, so as to ensure that the grinding process is terminated under the best dissolution effect and nutrient release conditions.

[0078] Accordingly, in step S545, based on the solution color-texture global distribution fusion feature vector, determining whether to stop stirring includes: inputting the solution color-texture global distribution fusion feature vector into a classifier-based stirring controller to obtain a control instruction, wherein the control instruction is used to indicate whether to stop stirring.

[0079] It should be understood that the role of the classifier is to use the given categories and known training data to learn classification rules and classifiers, and then classify (or predict) unknown data. Logistic regression and SVM are often used to solve binary classification problems. For multi-class classification problems, logistic regression or SVM can also be used, but multiple binary classifications are required to form a multi-classification. However, this is prone to errors and is not efficient. Commonly used multi-classification methods include the Softmax classification function.

[0080] In the technical solution of the present application, the global solution color distribution feature vector represents the global color distribution feature of the current solution state image, and the global solution texture distribution feature vector represents the global texture distribution feature of the current solution state image. When the global solution color distribution feature vector and the global solution texture distribution feature vector are input into the equalization threshold feature vector adaptive fusion module, adaptive feature fusion between the two is performed based on the correlation information between the global color feature and the global texture feature of the current solution state. However, considering the characteristic modal difference between the global solution color distribution feature vector and the global solution texture distribution feature vector, the solution color-texture global distribution fusion feature vector obtained by position-by-position weighting will have the phenomenon of local feature fuzziness, that is, the solution color-texture global distribution fusion feature vector has local distribution differences, which will lead to its harmony offset relative to the predetermined category when passing through the classifier, affecting the classification iteration effect.

[0081] In a preferred embodiment, inputting the solution color-texture global distribution fusion feature vector into a classifier-based stirring controller to obtain a control instruction comprises the following steps: optimizing the solution color-texture global distribution fusion feature vector to obtain an optimized solution color-texture global distribution fusion feature vector; passing the optimized solution color-texture global distribution fusion feature vector through the classifier-based stirring controller to obtain the control instruction.

[0082] Among them, the solution color-texture global distribution fusion feature vector is optimized to obtain the optimized solution color-texture global distribution fusion feature vector, including: calculating the autocorrelation matrix of the solution color-texture global distribution fusion feature vector and its own transposition, and calculating the inner product of the i-th and j-th row vectors of the autocorrelation matrix as the matrix value of the (i, j)-th position of the weight matrix to obtain the weight matrix; after matrix multiplication of the weight matrix and the autocorrelation matrix, further matrix-vector multiplication with the solution color-texture global distribution fusion feature vector is performed to obtain a correction vector; and the correction vector is dot-multiplied with the solution color-texture global distribution fusion feature vector to obtain the optimized solution color-texture global distribution fusion feature vector.

[0083] In a specific example, optimizing the solution color-texture global distribution fusion feature vector to obtain an optimized solution color-texture global distribution fusion feature vector includes: optimizing the solution color-texture global distribution fusion feature vector using the following optimization formula to obtain the optimized solution color-texture global distribution fusion feature vector; wherein the optimization formula is:

[0084]

[0085] Wherein, V represents the solution color-texture global distribution fusion feature vector, (·) T represents the transpose of a vector, represents matrix multiplication operation, ⊙ represents dot product, M represents the autocorrelation matrix of the solution color-texture global distribution fusion feature vector, V i is the i-th row vector of the autocorrelation matrix, V j is the jth row vector of the autocorrelation matrix, m i,j represents the inner product of the i-th and j-th row vectors of the autocorrelation matrix, L is the length of the solution color-texture global distribution fusion feature vector, M' is the weight matrix, and V' represents the optimized solution color-texture global distribution fusion feature vector.

[0086] Accordingly, in the technical solution of the present application, based on the self-correlation dimension of the solution color-texture global distribution fusion feature vector as the object to be modulated, the correlation expansion is performed based on the spatial sub-dimensional complexity of the high-dimensional feature space of the feature distribution, thereby introducing the decomposable correlation dimension offset of the solution color-texture global distribution fusion feature vector into the heterogeneous correlation embedding space, thereby enhancing its correlation self-consistency through the joint fine-tuning of the decomposable dimension set represented by the heterogeneous correlation embedding space, so as to improve the predetermined category harmony of the feature set of the solution color-texture global distribution fusion feature vector in the associated target classification domain, thereby improving the classification training speed and the accuracy of the classification results when the solution color-texture global distribution fusion feature vector is classified by a classifier-based stirring controller.

[0087] Furthermore, the present application also provides an organic trace element water-soluble fertilizer, which is prepared by any of the above-mentioned methods for preparing the organic trace element water-soluble fertilizer.

[0088] Based on the above embodiments, see Figure 5 The figure shows a schematic diagram of the structure of a system 800 for preparing an organic trace element water-soluble fertilizer in an embodiment of the present application. The system 800 for preparing an organic trace element water-soluble fertilizer comprises: a fermentation module 810 for fermenting wine lees to obtain a lees fermentation liquid; an evaporation and concentration module 820 for evaporating and concentrating the lees fermentation liquid to obtain a concentrated liquid; a stirring module 830 for adding the concentrated liquid into a reactor for stirring, and adding nitrogen, phosphorus pentoxide, potassium oxide and a suspending agent in sequence during the stirring process to obtain a mixed liquid, wherein the stirring time is 2 hours; a grinding module 840 for grinding the mixed liquid using a shearing colloid mill to obtain a liquid fertilizer; and a filling module 850 for conveying the liquid fertilizer to a filling line for filling to obtain a finished water-soluble fertilizer.

[0089] Here, those skilled in the art will appreciate that the specific functions and operations of the various modules in the above-mentioned system for preparing organic trace element water-soluble fertilizer 800 have been described in detail above. Figures 2 to 4 The preparation method of the organic trace element water-soluble fertilizer has been described in detail, and therefore, its repeated description will be omitted.

[0090] Based on the above embodiments, another exemplary embodiment of an electronic device is also provided in the embodiments of the present application. In some possible implementations, the electronic device in the embodiments of the present application may include a memory, a processor, and a computer program stored in the memory and run on the processor, wherein the steps of the method for preparing the organic trace element water-soluble fertilizer in the above embodiments can be implemented when the processor executes the program.

[0091] For example, in the case of electronic equipment Figure 1 Taking the server 100 in the example as an example, the processor in the electronic device is the processor 110 in the server 100, and the memory in the electronic device is the memory 120 in the server 100.

[0092] An embodiment of the present application also provides a computer-readable storage medium. Figure 6 1 shows a schematic diagram of a computer-readable storage medium 1000 according to an embodiment of the present application. Figure 6 As shown, the computer-readable storage medium 1000 stores computer-executable instructions 1001. When the computer-executable instructions 1001 are executed by the processor, the method for preparing the organic trace element water-soluble fertilizer according to the embodiment of the present application described with reference to the above figures can be executed. The computer-readable storage medium includes, but is not limited to, for example, volatile memory and / or non-volatile memory. The volatile memory may, for example, include a random access memory (RAM) and / or a cache memory (cache), etc. The non-volatile memory may, for example, include a read-only memory (ROM), a hard disk, a flash memory, etc.

[0093] The embodiments of the present application also provide a computer program product or a computer program, which includes computer executable instructions, and the computer executable instructions are stored in a computer-readable storage medium. The processor of the computer device reads the computer executable instructions from the computer-readable storage medium, and the processor executes the computer executable instructions, so that the computer device executes the method for preparing the organic trace element water-soluble fertilizer according to the embodiment of the present application.

[0094] Those skilled in the art will appreciate that the contents disclosed in this application may be subject to various variations and improvements. For example, the various devices or components described above may be implemented by hardware, or by software, firmware, or a combination of some or all of the three.

[0095] In addition, although the present application makes various references to certain units in the system according to embodiments of the present application, any number of different units can be used and run on the client and / or server. The units are only illustrative, and different aspects of the system and method can use different units.

[0096] Those skilled in the art will appreciate that all or part of the steps in the above method can be completed by instructing related hardware through a program, and the program can be stored in a computer-readable storage medium, such as a read-only memory, a disk or an optical disk. Optionally, all or part of the steps in the above embodiment can also be implemented using one or more integrated circuits. Accordingly, each module / unit in the above embodiment can be implemented in the form of hardware or in the form of software function modules. The present application is not limited to any particular form of combination of hardware and software.

[0097] Unless otherwise defined, all terms (including technical and scientific terms) used herein have the same meaning as commonly understood by those of ordinary skill in the art to which this application belongs. It should also be understood that terms such as those defined in common dictionaries should be interpreted as having a meaning consistent with their meaning in the context of the relevant technology, and should not be interpreted in an idealized or extremely formal sense, unless explicitly defined as such herein.

[0098] The above is an explanation of the present application and should not be considered as a limitation thereof. Although several exemplary embodiments of the present application have been described, it will be readily understood by those skilled in the art that many modifications may be made to the exemplary embodiments without departing from the novel teachings and advantages of the present application. Therefore, all of these modifications are intended to be included within the scope of the present application as defined by the claims. It should be understood that the above is an explanation of the present application and should not be considered to be limited to the specific embodiments disclosed, and modifications to the disclosed embodiments and other embodiments are intended to be included within the scope of the appended claims. The present application is defined by the claims and their equivalents.

Claims

1. A method for preparing an organic trace element water-soluble fertilizer, characterized in that: include: fermenting wine lees to obtain wine lees fermentation liquid; The wine lees fermentation liquid is evaporated and concentrated to obtain a concentrated liquid; Add the concentrated solution into a reaction kettle and stir, and add nitrogen, phosphorus pentoxide, potassium oxide and a suspending agent in sequence during the stirring process to obtain a mixed solution, wherein the stirring time is 2 hours; Grinding the mixed solution using a shearing colloid mill to obtain liquid fertilizer; The liquid fertilizer is transported to a filling line for filling to obtain a finished water-soluble fertilizer.

2. The method for preparing the organic trace element water-soluble fertilizer according to claim 1, characterized in that: The mixed liquid is ground using a shearing colloid mill to obtain liquid fertilizer, comprising: The solution state of the mixed solution is detected, and the grinding is stopped when N, P2O5, and K2O in the mixed solution are uniformly suspended in the solution.

3. The method for preparing the organic trace element water-soluble fertilizer according to claim 2, characterized in that: Detecting the solution state of the mixed solution, and stopping grinding when N, P2O5, and K2O in the mixed solution are uniformly suspended in the solution, comprising: Acquiring a current solution state image collected by an underwater camera deployed in the mixed liquid; Extracting solution color distribution features of the current solution state image to obtain a global solution color distribution feature vector; Extracting solution texture distribution features of the current solution state image to obtain a global solution texture distribution feature vector; Inputting the global solution color distribution feature vector and the global solution texture distribution feature vector into a balanced threshold feature vector adaptive fusion module to obtain a solution color-texture global distribution fusion feature vector; Whether to stop stirring is determined based on the solution color-texture global distribution fusion feature vector.

4. The method for preparing the organic trace element water-soluble fertilizer according to claim 3, characterized in that: Extracting the solution color distribution feature of the current solution state image to obtain a global solution color distribution feature vector includes: Calculating a color histogram of the current solution state image to obtain a solution state color histogram; The solution state color histogram is input into a solution color feature extractor based on the first ViT model to obtain the global solution color distribution feature vector.

5. The method for preparing the organic trace element water-soluble fertilizer according to claim 4, characterized in that: Extracting the solution texture distribution feature of the current solution state image to obtain a global solution texture distribution feature vector includes: Calculating a histogram of directional gradients of the current solution state image to obtain a histogram of directional gradients of the solution state; The solution state directional gradient histogram is input into a solution texture feature extractor based on a second ViT model to obtain the global solution texture distribution feature vector.

6. The method for preparing the organic trace element water-soluble fertilizer according to claim 5, characterized in that: Inputting the global solution color distribution feature vector and the global solution texture distribution feature vector into a balanced threshold feature vector adaptive fusion module to obtain a solution color-texture global distribution fusion feature vector, including: The global solution color distribution feature vector and the global solution texture distribution feature vector are fused through a multi-dimensional fusion module to obtain a first solution color-texture feature distribution fusion representation vector, a second solution color-texture feature distribution fusion representation vector, and a third solution color-texture feature distribution fusion representation vector; Respectively calculating the equalization threshold values ​​of the first solution color-texture feature distribution fusion representation vector, the second solution color-texture feature distribution fusion representation vector, and the third solution color-texture feature distribution fusion representation vector to obtain a first equalization threshold value, a second equalization threshold value, and a third equalization threshold value; Determine a first weight parameter and a second weight parameter based on the first equalization threshold value, the second equalization threshold value, and the third equalization threshold value, wherein the first weight parameter is an average of the first equalization threshold value, the second equalization threshold value, and the third equalization threshold value, and the second weight parameter is a difference between one and the first weight parameter; The global solution color distribution feature vector is weighted position by position with the first weight parameter to obtain a weighted global solution color distribution feature vector, and the global solution texture distribution feature vector is weighted position by position with the second weight parameter to obtain a weighted global solution texture distribution feature vector; The weighted global solution color distribution feature vector and the weighted global solution texture distribution feature vector are added together to obtain the solution color-texture global distribution fusion feature vector.

7. The method for preparing the organic trace element water-soluble fertilizer according to claim 6, characterized in that: The global solution color distribution feature vector and the global solution texture distribution feature vector are fused through a multi-dimensional fusion module to obtain a first solution color-texture feature distribution fusion representation vector, a second solution color-texture feature distribution fusion representation vector, and a third solution color-texture feature distribution fusion representation vector, including: Cascading the global solution color distribution feature vector and the global solution texture distribution feature vector to obtain the first solution color-texture feature distribution fusion representation vector; Adding the global solution color distribution feature vector and the global solution texture distribution feature vector by position to obtain the second solution color-texture feature distribution fusion representation vector; The global solution color distribution feature vector and the global solution texture distribution feature vector are multiplied by position points to obtain the third solution color-texture feature distribution fusion representation vector.

8. The method for preparing the organic trace element water-soluble fertilizer according to claim 7, characterized in that: Respectively calculating the equalization threshold values ​​of the first solution color-texture feature distribution fusion representation vector, the second solution color-texture feature distribution fusion representation vector, and the third solution color-texture feature distribution fusion representation vector to obtain a first equalization threshold value, a second equalization threshold value, and a third equalization threshold value, including: Multiplying the first solution color-texture feature distribution fusion representation vector by a first predetermined transformation vector to obtain a first threshold scoring coefficient; The first threshold scoring coefficient and the first bias parameter are added together and then passed through a sigmoid activation function to obtain the first equalization threshold value.

9. The method for preparing the organic trace element water-soluble fertilizer according to claim 8, characterized in that: Determining whether to stop stirring based on the solution color-texture global distribution fusion feature vector includes: The solution color-texture global distribution fusion feature vector is input into a classifier-based stirring controller to obtain a control instruction, wherein the control instruction is used to indicate whether to stop stirring.

10. An organic trace element water-soluble fertilizer, characterized in that: The organic trace element water-soluble fertilizer is prepared by the preparation method of the organic trace element water-soluble fertilizer according to any one of claims 1 to 9.

Images