Round-particle biological organic potassium chloride fertilizer, preparation method and formula optimization method

Abstract

The invention relates to the technical field of fertilizer preparation, and discloses a round-particle biological organic potassium chloride fertilizer, a preparation method and a formula optimization method.The round-particle biological organic potassium chloride fertilizer comprises the steps that firstly, a component category set of raw materials to be processed and a corresponding initial proportion set are obtained, and a processing risk value is calculated according to real-time process parameters; screening from the component category set according to the processing risk value to obtain a target component category and a corresponding target component proportion, determining a product defect coefficient based on the fertilizer product image and the fertilizer product data, and optimizing the target component proportion based on the product defect coefficient and a natural heuristic algorithm. By combining real-time process data, fertilizer product images and data analysis, the limitation of dependence on prior knowledge and empirical laws traditionally is effectively broken through, component categories and component proportions are dynamically adjusted, product defect coefficients are calculated, optimization is performed by means of a natural heuristic algorithm, interaction among components is fully considered in the scheme, and the method is suitable for large-scale popularization and application. The continuous optimization of the fertilizer production process is ensured.

Description

Technical Field

[0001] The present invention relates to the technical field of fertilizer preparation, and more specifically, to round granular bio-organic potassium chloride fertilizers, preparation methods, and formula optimization methods. Background Art

[0002] Round granular bio-organic potassium chloride fertilizer is a highly efficient fertilizer that combines bio-organic components with potassium chloride. Its round granular structure not only facilitates application and storage but also effectively reduces dust and caking phenomena. This fertilizer is rich in organic substances such as humic acid, which can improve soil structure, increase soil fertility, provide sufficient potassium elements for plants, thereby enhancing stress resistance and promoting root growth. When preparing organic potassium chloride fertilizers, potassium sources (such as potassium feldspar and potassium sulfate) and organic substances (such as organic acids or organic waste) are usually used as raw materials. Although there have been studies on optimizing the preparation process of organic fertilizers in the prior art, there are still some deficiencies.

[0003] For example, the patent application with publication number CN116730764A provides a fertilizer for improving soil environment and its preparation method. This patent optimizes the ratio between biochar, high-nitrogen phosphorus potassium diatomite, and serpentine powder to improve the use effect of the fertilizer. Similarly, the patent application with publication number CN113603537A provides a fertilizer formula for promoting flower bud differentiation and its preparation method. By optimizing the ratio of components such as amino acids, superphosphate, and magnesium sulfate, higher-quality chemical fertilizer crystals are prepared.

[0004] Although the above prior art has optimized the component formulations in the fertilizer preparation process, these techniques generally rely on prior knowledge or empirical rules and do not fully consider the interactions and influences between components in the actual fertilizer preparation process. This method may lead to limitations in the optimization results and fail to achieve the best component formulation, thus affecting the processing quality of the fertilizer.

[0005] In view of this, the present invention proposes round granular bio-organic potassium chloride fertilizers, preparation methods, and formula optimization methods to solve the above problems. Summary of the Invention

[0006] To overcome the above-mentioned defects of the prior art, the present invention provides round granular bio-organic potassium chloride fertilizers, preparation methods, and formula optimization methods.

[0007] To achieve the above object, the present invention provides the following technical solutions:

[0008] In the first aspect, a formula optimization method for round granular bio-organic potassium chloride fertilizer, comprising:

[0009] Obtain the set of component categories of the raw material to be processed and the corresponding set of initial ratios. The set of component categories includes W initial component categories, and the set of initial ratios includes Q initial component ratios. The initial component categories and the initial component ratios correspond one by one. W and Q are integers greater than 1.

[0010] Process the raw material to be processed according to the initial component categories, initial component ratios, and preset standard processing parameters to obtain a fertilizer product. During the processing, obtain real-time process parameters, calculate the processing risk value based on the real-time process parameters, and screen out the target component categories and the corresponding target component ratios from the set of component categories according to the processing risk value.

[0011] Obtain the fertilizer product image and fertilizer product data corresponding to the target component categories, determine the product defect coefficient based on the fertilizer product image and fertilizer product data, and optimize the target component ratios based on the product defect coefficient and the nature-inspired algorithm to generate the optimal component ratios.

[0012] Furthermore, the real-time process parameters include the volatile organic concentration value, the reaction residue value, and the absolute value of the viscosity difference. The processing risk value is calculated based on the volatile organic concentration value, the reaction residue value, and the absolute value of the viscosity difference.

[0013] Furthermore, the method for screening out the target component categories includes:

[0014] Traverse the set of component categories, obtain the processing risk values corresponding to each initial component category in the set of component categories, sort the M processing risk values in ascending order, take the processing risk value at the first place as the target processing risk value, and take the initial component category corresponding to the target processing risk value as the target component category. W = M.

[0015] Furthermore, the method for determining the product defect coefficient includes:

[0016] Calculate the absolute value of the gray difference between each point in the fertilizer product image and the corresponding point in the standard product image, and form a gray difference sequence. Calculate the corresponding gray standard deviation according to the gray difference sequence, calculate the similarity between the fertilizer product data and the standard product data, and calculate and generate the product defect coefficient based on the gray standard deviation and the similarity.

[0017] Furthermore, the method for calculating and generating the product defect coefficient based on the gray standard deviation and the similarity includes:

[0018]

[0019] In the formula, FQC is the product defect coefficient, GSD is the gray standard deviation, SIM is the similarity, and F1 and F2 are both corresponding weight factors.

[0020] Further, the method for optimizing the target group allocation ratio includes:

[0021] S301: Randomly select a target group allocation ratio as the initial dynamic ratio EP initial , preset the initial temperature T0 of simulated annealing and the cooling rate α, where 1 > α > 0;

[0022] S302: Take the product defect coefficient f(EP initial ) under the initial dynamic ratio EP initial as the objective function value;

[0023] S303: Add a random perturbation to the current dynamic ratio EP current to generate a new dynamic ratio EP new , and the starting point of the current dynamic ratio EP current is the initial dynamic ratio EP initial ;

[0024] S304: Calculate the new product defect coefficient f(EP new ) corresponding to the new dynamic ratio EP new , compare the new product defect coefficient f(EP new ) with the current product defect coefficient f(EP current ), and the current product defect coefficient f(EP current ) corresponds to the current dynamic ratio EP current ;

[0025] S305: Determine whether to update the dynamic ratio, let T = α × T, and return to S303, where the starting point of T is T0;

[0026] S306: Repeat the above S303 - S305 until T < T min , stop the iteration, output the optimal dynamic region EP best , and take the optimal dynamic region EP best as the standard dynamic region, T is the temperature parameter, and T min is the preset minimum temperature value.

[0027] Further, the method for generating the new dynamic ratio EP new includes:

[0028] EP new = EP current + Δ EP ;

[0029] where Δ EP is the random perturbation;

[0030] The method for determining whether to update the dynamic ratio includes:

[0031] If f(EP new ) < f(EP current ), then let EP current = EP new ;

[0032] If f(EP new ) ≥ f(EP current ), then randomly generate a probability threshold r between 0 and 1, and determine whether the probability K is greater than the probability threshold r. If so, then let EP current = EP new . If not, then keep EP current unchanged. The probability K is:

[0033]

[0034] where exp(·) is the exponential function with base e, and e is the natural constant.

[0035] Furthermore, the method for obtaining the initial ratio set includes:

[0036] Input the group classification set into a pre-constructed ratio output model to obtain the corresponding initial ratio set;

[0037] The construction method of the ratio output model includes:

[0038] Obtain a sample data set, where the sample data set includes a historical group classification set and a historical initial ratio set;

[0039] Divide the sample data set into a sample training set and a sample test set, and construct a regression network;

[0040] Use the historical group classification set in the sample training set as the input data of the regression network, and use the historical initial ratio set in the sample training set as the output data of the regression network to train the regression network to obtain an initial regression network for predicting the real-time initial ratio set;

[0041] Use the sample test set to test the initial regression network, and output the initial regression network that satisfies being less than the preset error value as the ratio output model.

[0042] In the second aspect, a method for preparing round particle bio-organic potassium chloride fertilizer is implemented according to the above-mentioned round particle bio-organic potassium chloride fertilizer formula optimization method.

[0043] In the third aspect, round particle bio-organic potassium chloride fertilizer is prepared according to the above-mentioned method for preparing round particle bio-organic potassium chloride fertilizer.

[0044] Compared with the prior art, the beneficial effects of the present invention are:

[0045] The present invention first obtains the set of component categories of the raw materials to be processed and the corresponding initial ratio set, then calculates the processing risk value according to the real-time process parameters, screens the target component category and the corresponding target component ratio from the set of component categories according to the processing risk value, obtains the fertilizer product image and fertilizer product data corresponding to the target component category, determines the product defect coefficient based on the fertilizer product image and fertilizer product data, and optimizes the target component ratio based on the product defect coefficient and the nature-inspired algorithm. By combining the analysis of real-time process data, fertilizer product images and data, the present invention effectively breaks through the limitations of traditional reliance on prior knowledge and empirical rules. By dynamically adjusting the component categories and component ratios, calculating the product defect coefficient, and optimizing with the help of the nature-inspired algorithm, the solution fully considers the interaction between components, ensures the continuous optimization of the fertilizer production process, and thus improves the processing quality of fertilizers and the consistency of products. Brief Description of the Drawings

[0046] Figure 1 It is a schematic flow chart of the method for optimizing the formula of round granular bio-organic potassium chloride fertilizer in the present invention;

[0047] Figure 2 It is a schematic diagram of the method for screening the target component category in the present invention;

[0048] Figure 3 It is a schematic diagram of the method for determining the product defect coefficient in the present invention;

[0049] Figure 4 It is a schematic diagram of an electronic device in the present invention. Detailed Embodiments

[0050] Next, the technical solutions in the embodiments of the present invention will be clearly and completely described in conjunction with the accompanying drawings in the embodiments of the present invention. Obviously, the described embodiments are only a part of the embodiments of the present invention, rather than all the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by those of ordinary skill in the art without creative efforts shall fall within the protection scope of the present invention.

[0051] Embodiment 1

[0052] Please refer to Figure 1 As shown, this embodiment discloses and provides a method for optimizing the formula of round granular bio-organic potassium chloride fertilizer, including:

[0053] S10: Obtain the set of component categories of the raw materials to be processed and the corresponding initial ratio set. The set of component categories includes W initial component categories, and the initial ratio set includes Q initial component ratios. The initial component categories and the initial component ratios correspond one by one. W and Q are integers greater than 1;

[0054] In this embodiment, the raw material to be processed refers to the fertilizer used for producing organic potassium chloride, and the initial component categories at least include one potassium source category and one organic substance category. Exemplarily, the set of component categories includes 3 initial component categories. The first initial component category includes potassium chloride and organic acid, the second initial component category includes potassium sulfate and humus, and the third initial component category includes potassium feldspar and plant residues. It can be understood that the set of component categories is predetermined before preparation, and the initial ratio set is determined according to the set of component categories.

[0055] The method for obtaining the initial ratio set includes:

[0056] Input the set of component categories into a pre-constructed ratio output model to obtain the corresponding initial ratio set.

[0057] The construction method of the ratio output model includes:

[0058] Obtain a sample data set, where the sample data set includes a historical set of component categories and a historical initial ratio set;

[0059] Divide the sample data set into a sample training set and a sample test set, and construct a regression network;

[0060] Use the historical set of component categories in the sample training set as the input data of the regression network, and use the historical initial ratio set in the sample training set as the output data of the regression network to train the regression network to obtain an initial regression network for predicting the real-time initial ratio set;

[0061] Use the sample test set to test the initial regression network, and output the initial regression network that meets the requirement of being less than the preset error value as the ratio output model. The initial regression network is a deep neural network model.

[0062] It should be noted that the above historical initial ratio set is set according to prior knowledge or expert experience and is used as the training data of the ratio output model. For example, if the first initial component category includes potassium chloride and organic acid, and the second initial component category includes potassium sulfate and humus, then the first initial ratio corresponding to the first initial component category is 70:30, and the second initial ratio corresponding to the second initial component category is 60:40. It should be noted that the above content is only an exemplary illustration, and this embodiment does not limit this.

[0063] S20: Process the raw material to be processed according to the initial component categories, the initial component ratios, and the preset standard processing parameters to obtain a fertilizer product. During the processing, obtain real-time process parameters, calculate the processing risk value according to the real-time process parameters, and screen the target component category and the corresponding target component ratio from the set of component categories according to the processing risk value;

[0064] Among them, the standard processing parameters include but are not limited to standard processing temperature, standard processing humidity, standard processing time, and standard stirring speed, etc. The fertilizer product refers to organic potassium chloride fertilizer. It should be noted that in the process of treating the raw materials to be processed to obtain the fertilizer product, there are multiple treatment steps, such as stirring and mixing steps, reaction steps, drying steps, and granulation steps. Each treatment step is carried out according to the standard processing parameters. Exemplarily, in the stirring and mixing step, the raw materials to be processed need to be stirred according to the standard stirring speed. Similarly, in the reaction step, it needs to be controlled according to the standard processing temperature and standard processing time. Since each treatment step is prior art, the details of each treatment step are not elaborated in detail in this embodiment.

[0065] It should be noted that the above standard processing parameters are also preset by those skilled in the art. Those skilled in the art can determine the standard processing parameters through literature research, experimental data analysis, or computer simulation, etc. During the processing, different initial group categories can be divided into different groups, for example, divided into W groups. Each group only differs in the initial group category and its corresponding initial group ratio, and the rest of the processing process remains the same.

[0066] The above real-time process parameters include but are not limited to volatile organic concentration value, reaction residue value, and absolute value of viscosity difference. The volatile organic concentration value refers to the total concentration value of volatile organic compounds (VOCs) monitored in real time. The total concentration value refers to the overall concentration of all volatile organic compounds in unit volume of air during the reaction process, that is, the sum of the concentrations of various VOCs compounds. In the preparation process of organic potassium chloride fertilizer, volatile organic compounds (VOCs) are usually generated in the reaction step. Volatile organic compounds (VOCs) include formaldehyde, acetic acid vapor, and ethylene, etc. The reaction residue value refers to the weight of the by-products generated after the reaction. In the preparation process of organic potassium chloride fertilizer, the by-products are also generated in the reaction step. The by-products can be inorganic salts such as calcium sulfate and sodium sulfate, and precipitates such as calcium carbonate. It can be understood that after the reaction step is completed, the by-products are obtained through filtration, and then the corresponding by-product weight is obtained according to the corresponding weight sensor.

[0067] The absolute value of viscosity difference refers to the absolute value of the difference between the maximum real-time viscosity and the viscosity standard value. During the stirring and mixing step, the real-time viscosity value is obtained through a viscosity sensor, and the difference between the maximum real-time viscosity and the viscosity standard value is calculated. The larger the absolute value of viscosity difference, the greater the deviation in the mixing process. The larger the above volatile organic concentration value, the more intense the decomposition or side reaction of the organic matter during the reaction process, resulting in incomplete reaction phenomenon and greater impact on the environment. The larger the reaction residue value, the less fully the reaction has proceeded.

[0068] The method for calculating the processing risk value based on real-time process parameters includes:

[0069]

[0070] In the formula, PRV is the processing risk value, VDA is the absolute value of viscosity difference, VOC is the volatile organic concentration value, REV is the reaction residue value, max(·) is the maximum value function, min(·) is the minimum value function, cot -1 (·) is the arccotangent function, tan -1 (·) is the arctangent function, cosh[·] is the hyperbolic cosine function, ln{·} is the natural logarithm function with base e, and e is the natural constant.

[0071] In this embodiment, the absolute value of viscosity difference and the volatile organic concentration value are taken as examples. The larger the absolute value of viscosity difference, the greater the deviation in the mixing process; the larger the volatile organic concentration value, the greater the deviation in the reaction process. From the above, it can be seen that when the processing risk value is larger, it indicates that there are greater deviations and potential quality problems in the processing process.

[0072] Referring to Figure 2 , the method for screening the target group classification includes:

[0073] Traverse the set of group classifications, obtain the processing risk value corresponding to each initial group classification in the set of group classifications, sort the M processing risk values in ascending order, take the processing risk value at the first place as the target processing risk value, and take the initial group classification corresponding to the target processing risk value as the target group classification, where W = M.

[0074] Since each initial group classification corresponds to a processing risk value, and the larger the processing risk value, the greater the deviation and potential quality problems in the processing process. Therefore, in this embodiment, the smallest processing risk value is selected from the M processing risk values as the target processing risk value, and the initial group classification corresponding to the target processing risk value is taken as the target group classification. Similarly, the corresponding initial group allocation ratio is taken as the target group allocation ratio. In this way, the optimal initial group allocation ratio is selected from the set of group classifications, effectively reducing the deviation in the processing process, reducing potential quality problems, ensuring that the quality of the finally produced fertilizer product is more stable and reliable. By selecting the group classification with the smallest processing risk value and the corresponding initial group allocation ratio, the formula can be optimized during the production process, reducing problems such as incomplete reaction, generation of volatile organic compounds (VOCs), and viscosity deviation in the mixing process. This method can ensure that the execution of each processing step is more stable, ultimately improving the performance and consistency of the fertilizer product, and reducing waste and environmental impact in production.

[0075] The initial group classification and ratio obtained in this embodiment are based on prior knowledge or model calculations. However, in the actual production process, it may be affected by factors such as the environment, raw material quality, and equipment operating status. By introducing real-time monitoring and risk value calculation, this application can further optimize the group classification according to the real-time situation in production. This sequential combination can reduce deviations in actual production, such as incomplete reactions, uneven mixing, excessive volatile organic compounds, etc., thereby improving production stability and product quality.

[0076] S30: Obtain the fertilizer product image and fertilizer product data corresponding to the target group classification, determine the product defect coefficient based on the fertilizer product image and fertilizer product data, and optimize the target group ratio based on the product defect coefficient and the nature-inspired algorithm to generate the optimal group ratio.

[0077] In this embodiment, the fertilizer product image can be the surface crystal image of round granular organic potassium chloride fertilizer. As can be seen from the above, the process of obtaining the fertilizer product from the raw materials to be processed includes multiple processing steps, such as the stirring and mixing step, the reaction step, the drying step, and the granulation step. In the reaction step, the initial formation of crystals occurs in a solution or wet state. In this stage, the organic matter reacts with the inorganic potassium source to form potassium salts with a crystal structure. As the reaction progresses, the crystals gradually grow and expand in the fertilizer particles. In the subsequent drying step and granulation step, the removal of moisture and the control of temperature promote the crystal structure to become more stable and continue to grow. The temperature and humidity changes during the drying process will affect the crystallization rate and size of the crystals. The increase in temperature usually accelerates the evaporation of moisture and provides more space for the growth of crystals. The applied pressure and movement further enhance the crystal arrangement by controlling the shape and density of the particles. When the particles are subjected to external forces, the crystals begin to be evenly distributed along the surface and inside of the particles, thus forming the final crystallized structure.

[0078] Refer to Figure 3 , and the method for determining the product defect coefficient includes:

[0079] Calculate the absolute value of the gray difference between each point in the fertilizer product image and the corresponding point in the standard product image, and form a gray difference sequence. Calculate the corresponding gray standard deviation according to the gray difference sequence, calculate the similarity between the fertilizer product data and the standard product data, and calculate and generate the product defect coefficient based on the gray standard deviation and the similarity.

[0080] It should be noted that the standard product image usually refers to the surface crystal image of round granular organic potassium chloride fertilizer with quality meeting the standard. These images are used as references for comparison with the images of the fertilizer products to be processed. Each point in the fertilizer product image refers to each pixel point. It can be understood that since the specifications of the fertilizer product image are the same as those of the standard product image, each pixel point in the fertilizer product image can find a corresponding point in the standard product image. The gray difference sequence refers to the sequence formed by arranging in order the gray value differences between each pixel point in the fertilizer product image and the corresponding pixel point in the standard product image. Calculating the standard deviation is a prior art, and this embodiment will not elaborate on it too much.

[0081] Among them, the fertilizer product data is obtained through real-time detection of the fertilizer product. The data includes but is not limited to particle strength value, moisture content value, organic matter content value, and potassium content value. The standard product data refers to the physical and chemical characteristic data of round granular organic potassium chloride fertilizer with quality meeting the standard, which is used as a reference or benchmark for comparison with the data of the fertilizer product to be processed. The standard product data also includes standard particle strength value, standard moisture content value, standard organic matter content value, standard potassium content value, etc.

[0082] It should be added that calculating the similarity between the fertilizer product data and the standard product data can be achieved by comparing the fertilizer product data and the standard product data item by item and calculating the Euclidean distance or cosine similarity. This can quantify the difference degree between the two, so as to obtain the similarity between the fertilizer product data and the standard product data. This similarity provides a scientific basis for evaluating the quality of the fertilizer product.

[0083] The method for calculating the product defect coefficient based on the gray standard deviation and similarity includes:

[0084]

[0085] In the formula, FQC is the product defect coefficient, GSD is the gray standard deviation, SIM is the similarity, and F1 and F2 are both corresponding weight factors.

[0086] From the above content, it can be seen that the larger the gray standard deviation, the greater the gray value difference between the fertilizer product image and the standard product image, which means that the surface crystal structure of the fertilizer product is quite different from that of the standard product. The greater the similarity, the higher the similarity between the fertilizer product data and the standard product data, which means that the physical and chemical characteristics (such as particle strength, moisture content, potassium content, etc.) of the fertilizer product are closer to those of the standard product, and the product quality is better and meets the standard requirements. Therefore, in this embodiment, the product defect coefficient is negatively correlated with the product quality.

[0087] The nature-inspired algorithm can be a simulated annealing algorithm. The methods for optimizing the target group allocation ratio include:

[0088] S301: Randomly select a target group allocation ratio as the initial dynamic ratio EP initial , preset the initial temperature T0 and the cooling rate α of the simulated annealing, where 1 > α > 0;

[0089] It can be understood that in this embodiment, first, an initial group classification is selected from the W initial group classifications as the target group classification. However, the target group classification can correspond to several target group allocation ratios. Exemplarily, if the target group classification is potassium chloride and organic acid, the target group allocation ratio can be 70:30 or 80:35. Although the target group allocation ratio is obtained through the ratio output model, these group allocation ratios are only initial estimates based on historical data and experience and are not necessarily the optimal combination. Therefore, further optimization through the simulated annealing algorithm can explore among different target group allocation ratios to find the group allocation ratio with the optimal performance in the actual production process.

[0090] S302: Take the product defect coefficient f(EP initial ) under the initial dynamic ratio EP initial as the objective function value;

[0091] In this embodiment, a benchmark optimization objective is defined through the above steps, and all subsequent solutions will be compared with this benchmark to determine whether to move in a more optimal direction.

[0092] S303: Add random perturbations to the current dynamic ratio EP current to generate a new dynamic ratio EP new , where the starting point of the current dynamic ratio EP current is the initial dynamic ratio EP initial ;

[0093] In this embodiment, to increase the diversity of solutions, random perturbations are used to make the algorithm explore in the solution space and avoid falling into local optimal solutions.

[0094] S304: Calculate the new product defect coefficient f(EP new ) corresponding to the new dynamic ratio EP new , compare the new product defect coefficient f(EP new ) with the current product defect coefficient f(EP current ), and the current product defect coefficient f(EP current ) corresponds to the current dynamic ratio EP current ;

[0095] Among them, by comparing the objective function values of the new and old solutions, it can be determined whether the new solution is better, and then whether to accept the new solution. The advantage of doing this is to ensure that each step moves in the direction of minimizing the objective function, making the optimization process gradually approach the optimal solution.

[0096] S305: Determine whether to update the dynamic ratio, let T = α × T, and return to S303, where the starting point of T is T0;

[0097] In this embodiment, through the process of gradual cooling, it is ensured that the algorithm gradually converges to the optimal solution while exploring the solution space, balancing the capabilities of global exploration and local development.

[0098] S306: Repeat the above S303 - S305 until T < T min , stop the iteration, and output the optimal dynamic region EP best , and use the optimal dynamic region EP best as the standard dynamic region, T is the temperature parameter, and T min is the preset minimum temperature value.

[0099] The method for generating the new dynamic ratio EP new includes:

[0100] EP new = EP current + Δ EP ;

[0101] Among them, Δ EP is a random perturbation.

[0102] The method for determining whether to update the dynamic ratio includes:

[0103] If f(EP new ) < f(EP current ), then let EP current = EP new ;

[0104] If f(EP new ) ≥ f(EP current ), then randomly generate a probability threshold r between 0 and 1, determine whether the probability K is greater than the probability threshold r. If so, then let EP current = EP new , if not, then keep EP current unchanged, and the probability K is:

[0105]

[0106] Among them, exp(·) is the exponential function with base e, and e is the natural constant.

[0107] In this embodiment, the optimal target group classification and target group allocation ratio are first screened out by calculating the processing risk value through real-time monitoring of process parameters, and on this basis, further optimization is carried out. By analyzing the fertilizer product images and data, the product defect coefficient is determined, and the natural inspiration algorithm is used to optimize the target group allocation ratio. In this way, it not only ensures that the preliminary screened group allocation ratio can minimize the deviation and quality problems in the production process to the greatest extent, but also further improves the accuracy of the group allocation ratio through more refined defect analysis, making the fertilizer product more in line with the standards.

[0108] In this embodiment, the set of group classifications of the raw materials to be processed and the corresponding initial ratio set are first obtained, and then the processing risk value is calculated according to the real-time process parameters. The target group classification and the corresponding target group allocation ratio are screened out from the set of group classifications according to the processing risk value. The fertilizer product images and fertilizer product data corresponding to the target group classification are obtained. The product defect coefficient is determined based on the fertilizer product images and fertilizer product data. The target group allocation ratio is optimized based on the product defect coefficient and the natural inspiration algorithm. By combining the analysis of real-time process data, fertilizer product images and data, this embodiment effectively breaks through the limitations of traditional reliance on prior knowledge and empirical rules. By dynamically adjusting the group classification and group allocation ratio, calculating the product defect coefficient, and optimizing with the help of the natural inspiration algorithm, the scheme fully considers the interaction between components, ensures the continuous optimization of the fertilizer production process, and thus improves the processing quality of fertilizers and the consistency of products.

[0109] Example 2

[0110] For the parts not described in detail in this embodiment, refer to the description content of Example 1. A method for preparing round granular bio-organic potassium chloride fertilizer is provided, which is implemented based on the round granular bio-organic potassium chloride fertilizer formula optimization method in Example 1.

[0111] Example 3

[0112] This embodiment provides round granular bio-organic potassium chloride fertilizer, which is prepared by using the method for preparing round granular bio-organic potassium chloride fertilizer in Example 2.

[0113] Example 4

[0114] As Figure 4 shown, this embodiment provides an electronic device, including a storage, a processor, and a computer program stored in the storage and executable on the processor. When the processor executes the computer program, it implements the round granular bio-organic potassium chloride fertilizer formula optimization method provided by the above-mentioned various methods.

[0115] Since the electronic device introduced in this embodiment is the electronic device used to implement the method for optimizing the formula of round-particle bio-organic potassium chloride fertilizer in the embodiments of the present application, based on the method for optimizing the formula of round-particle bio-organic potassium chloride fertilizer introduced in the embodiments of the present application, those skilled in the art can understand the specific implementation manners and various forms of variation of the electronic device in this embodiment. Therefore, the specific implementation of how this electronic device implements the method in the embodiments of the present application will not be described in detail here. As long as the electronic device used by those skilled in the art to implement the method for optimizing the formula of round-particle bio-organic potassium chloride fertilizer in the embodiments of the present application falls within the scope of protection of the present application.

[0116] The above formulas are all calculated by taking the numerical values after dimensionless. The formula is a formula obtained by collecting a large amount of data and performing software simulation to approximate the real situation. The selection of the preset parameters, weights, and thresholds in the formula is set by those skilled in the art according to the actual situation.

[0117] The above embodiments can be implemented in whole or in part by software, hardware, firmware, or any combination thereof. When implemented using software, the above embodiments can be implemented in whole or in part in the form of a computer program product. The computer program product includes one or more computer instructions or computer programs. When the computer instructions or computer programs are loaded or executed on a computer, the processes or functions described in the embodiments of the present invention are generated in whole or in part. The computer can be a general-purpose computer, a special-purpose computer, a computer network, or other programmable devices. The computer instructions can be stored in a computer-readable storage medium or transmitted from one computer-readable storage medium to another. For example, the computer instructions can be transmitted from one website, computer, server, or data center to another website, computer, server, or data center via a wired or wireless network. The computer-readable storage medium can be any available medium that can be accessed by a computer or a data storage device such as a server or data center that includes one or more collections of available media. The available medium can be a magnetic medium (e.g., floppy disk, hard disk, magnetic tape), an optical medium (e.g., DVD), or a semiconductor medium. The semiconductor medium can be a solid-state drive.

[0118] Those of ordinary skill in the art can realize that the units and algorithm steps of each example described in combination with the embodiments disclosed in the present invention can be implemented by electronic hardware or a combination of computer software and electronic hardware. Whether these functions are executed in hardware or software depends on the specific application and design constraints of the technical solution. Professional technicians can use different methods to implement the described functions for each specific application, but such implementation should not be considered to exceed the scope of the present invention.

[0119] Those skilled in the art can clearly understand that for the convenience and brevity of description, the specific working processes of the systems, devices, and units described above can refer to the corresponding processes in the foregoing method embodiments and will not be elaborated herein.

[0120] In several embodiments provided by the present invention, it should be understood that the disclosed systems, devices, and methods can be implemented in other ways. For example, the device embodiments described above are merely illustrative. For example, the division of the units is only one type, and there can be other division methods in actual implementation. For example, 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 displayed or discussed couplings or direct couplings or communication connections to each other can be through some interfaces, and the indirect couplings or communication connections of the devices or units can be in electrical, mechanical, or other forms.

[0121] The units described as separate components may or may not be physically separated, and the components displayed as units may or may not be physical units, that is, they can be located in one place or distributed to multiple network units. Some or all of the units can be selected according to actual needs to achieve the purpose of the solution of this embodiment.

[0122] In addition, in each embodiment of the present invention, the functional units can be integrated in a processing unit, or each unit can exist physically alone, or two or more units can be integrated in one unit.

[0123] As described above, the above is only the specific implementation manner of the present invention, but the protection scope of the present invention is not limited thereto. Any person skilled in the art can easily think of changes or substitutions within the technical scope disclosed by the present invention, and all should be covered by the protection scope of the present invention. Therefore, the protection scope of the present invention should be subject to the protection scope of the claims.

[0124] Finally: The above is only the preferred embodiment of the present invention and is not used to limit the present invention. Any modifications, equivalent substitutions, improvements, etc. made within the spirit and principle of the present invention shall be included in the protection scope of the present invention.

Claims

1. A method for optimizing the formula of round particle biological organic potassium chloride fertilizer, characterized in that, Including: Obtain the set of component categories of the raw material to be processed and the corresponding initial ratio set. The set of component categories includes W initial component categories, and the initial ratio set includes Q initial component ratios. The initial component categories and the initial component ratios correspond one by one. W and Q are integers greater than 1; According to the initial component categories, initial component ratios, and preset standard processing parameters, process the raw material to be processed to obtain a fertilizer product. During the processing, obtain real-time process parameters, calculate the processing risk value according to the real-time process parameters, and screen out the target component category and the corresponding target component ratio from the set of component categories according to the processing risk value; Obtain the fertilizer product image and fertilizer product data corresponding to the target component category, determine the product defect coefficient based on the fertilizer product image and fertilizer product data, and optimize the target component ratio based on the product defect coefficient and the nature-inspired algorithm to generate the optimal component ratio.

2. The method for optimizing the formula of round particle biological organic potassium chloride fertilizer according to claim 1, characterized in that, The real-time process parameters include the volatile organic concentration value, reaction residue value, and absolute viscosity difference. The processing risk value is calculated according to the volatile organic concentration value, reaction residue value, and absolute viscosity difference.

3. The method for optimizing the formula of the round particle biological organic potassium chloride fertilizer according to claim 2, characterized in that, The method for screening out the target component category includes: Traverse the set of component categories, obtain the processing risk value corresponding to each initial component category in the set of component categories, arrange the M processing risk values in ascending order, use the processing risk value at the first place as the target processing risk value, and use the initial component category corresponding to the target processing risk value as the target component category, where W = M.

4. The method for optimizing the formula of the round particle biological organic potassium chloride fertilizer according to claim 3, characterized in that The method for determining the product defect coefficient includes: Calculate the absolute value of the gray difference between each point in the fertilizer product image and the corresponding point in the standard product image, and form a gray difference sequence. Calculate the corresponding gray standard deviation according to the gray difference sequence, calculate the similarity between the fertilizer product data and the standard product data, and calculate and generate the product defect coefficient according to the gray standard deviation and the similarity.

5. The method for optimizing the formulation of the round particle biological organic potassium chloride fertilizer according to claim 4, characterized in that, The method for calculating and generating the product defect coefficient according to the gray standard deviation and the similarity includes: In the formula, FQC is the product defect coefficient, GSD is the gray standard deviation, SIM is the similarity, and F1 and F2 are both corresponding weight factors.

6. The method for optimizing the formula of the round particle biological organic potassium chloride fertilizer according to claim 5, wherein, The method for optimizing the target component ratio includes: S301: Randomly select a target group allocation ratio as the initial dynamic ratio EP initial , set the initial temperature T0 of the simulated annealing and the cooling rate α, where 1 > α > 0; S302: Take the product defect coefficient f(EP initial ) under the initial dynamic ratio EP initial as the objective function value; S303: Add random perturbations to the current dynamic ratio EP current to generate a new dynamic ratio EP new , where the starting point of the current dynamic ratio EP current is the initial dynamic ratio EP initial ; S304: Calculate the new dynamic ratio EP new The corresponding new product defect coefficient f(EP new ), compare the new product defect coefficient f(EP new ) with the current product defect coefficient f(EP current ), and the current product defect coefficient f(EP current ) corresponds to the current dynamic ratio EP current Correspondingly; S305: Judge whether to update the dynamic ratio, let T = α × T, and return to S303. The starting point of T is T0; S306: Repeat the above S303 - S305 until T < T min , stop the iteration, and output the optimal dynamic region EP best , take the optimal dynamic region EP best as the standard dynamic region, T is the temperature parameter, and T min is the preset minimum temperature value.

7. The method for optimizing the formula of round granular bio-organic potassium chloride fertilizer according to claim 6, characterized in that, The method for generating the new dynamic ratio EP new comprises: EP new = EP current + Δ EP ; where, Δ EP is a random perturbation; The method for judging whether to update the dynamic ratio includes: If f(EP new ) < f(EP current ), then let EP current = EP new ; If f(EP new ) ≥ f(EP current ), then a probability threshold r between 0 and 1 is randomly generated, and it is determined whether the probability K is greater than the probability threshold r. If so, then let EP current = EP new . If not, then keep EP current unchanged. The probability K is: where exp(·) is the exponential function with e as the base, and e is the natural constant.

8. The method for optimizing the formula of the round granular bio-organic potassium chloride fertilizer according to claim 1, characterized in that The method for obtaining the initial ratio set includes: Input the set of component categories into the pre-constructed ratio output model to obtain the corresponding initial ratio set; The construction method of the ratio output model includes: Obtain a sample data set, where the sample data set includes a historical set of component categories and a historical initial ratio set; Divide the sample data set into a sample training set and a sample test set, and construct a regression network; Use the historical set of component categories in the sample training set as the input data of the regression network, and use the historical initial ratio set in the sample training set as the output data of the regression network to train the regression network to obtain an initial regression network for predicting the real-time initial ratio set; The initial regression network is tested using a sample test set, and the initial regression network that satisfies being less than a preset error value is output as a ratio output model.

9. Preparation method of round particle biological organic potassium chloride fertilizer, characterized in that, Implement according to the method for optimizing the formula of the round particle biological organic potassium chloride fertilizer described in any one of claims 1-8.

10. Round granular bio-organic potassium chloride fertilizer, characterized in that, Prepared according to the method for preparing the round particle biological organic potassium chloride fertilizer described in claim 9.

Images