Penaeus monodon high-density culture optimization method based on multi-source culture environment analysis

Abstract

The invention relates to the technical field of prawn culture, in particular to a high-density culture optimization method for penaeus monodon based on multi-source culture environment analysis. A scale evaluation system of different preset cultivation environment condition combinations is built based on high-density penaeus monodon cultivation to evaluate the trend radiation diagram, whether the current cultivation density scale of the penaeus monodon is reasonable or not is analyzed through thermal polymerization after evaluation, and a multi-source cultivation environment analysis result is obtained; if one or more high-density abnormal breeding areas exist in a multi-source breeding environment analysis result, a simplex table model is constructed with the maximum bearing force as a constraint and the expected breeding density coefficient as a target function, and the optimal solution of the simplex table model is solved to determine the optimal breeding density scale of the high-density abnormal breeding areas. According to the method, the multi-source culture environment data can be fused to reasonably and scientifically analyze the high-density culture of the penaeus monodon and make an accurate and optimized culture strategy, so that the yield of the penaeus monodon per unit area is increased, and the low survival rate of the high-density culture is improved.

Description

TECHNICAL FIELD

[0001] The present application relates to the technical field of prawn culture, and in particular to a high-density culture optimization method for Marsupenaeus japonicus based on multi-source culture environment analysis. BACKGROUND

[0002] As an important economic shrimp, Marsupenaeus japonicus is widely cultured in the world due to its fast growth rate, large individual size, and excellent meat quality. With the rapid development of aquaculture industry, high-density culture has become a major means to increase unit area yield. However, in high-density culture environment, Marsupenaeus japonicus is more susceptible to diseases, hypoxia, and water quality deterioration, leading to decreased survival rate, uneven growth, and increased physiological stress. In traditional culture mode, the rationality of high-density culture of Marsupenaeus japonicus is often optimized based on periodic manual detection or experience, which cannot achieve real-time response and precise intervention to sudden environmental changes, resulting in decision lag and increased risk. Furthermore, existing optimization methods cannot comprehensively analyze the rationality of high-density culture of Marsupenaeus japonicus based on the multi-source culture environment generated in the culture area, which may lead to large optimization deviation in subsequent high-density abnormal culture of Marsupenaeus japonicus, thereby increasing the mortality rate and individual weight of Marsupenaeus japonicus in high-density culture environment, and further deteriorating the surrounding culture environment, which is not conducive to the unit area yield of Marsupenaeus japonicus. SUMMARY

[0003] The present application overcomes the shortcomings of the prior art and provides a high-density culture optimization method for Marsupenaeus japonicus based on multi-source culture environment analysis.

[0004] To achieve the above purpose, the technical solution adopted by the present application is as follows:

[0005] The present application provides a high-density culture optimization method for Marsupenaeus japonicus based on multi-source culture environment analysis, comprising the following steps:

[0006] S102: Obtain multi-source culture environment data of Marsupenaeus japonicus in a target culture area, and map the multi-source culture environment data from a high-dimensional space to a low-dimensional embedding space that changes with culture density to obtain a trend radiation map of the multi-source culture environment;

[0007] S104: Evaluate the trend radiation map based on a scale evaluation system of different preset culture environment condition combinations for high-density Marsupenaeus japonicus culture, and analyze whether the current culture density scale of Marsupenaeus japonicus is reasonable after heat aggregation, to obtain a multi-source culture environment analysis result;

[0008] S106: If there is one or more high-density abnormal aquaculture area in the multi-source cultivation environment analysis result, a simplex table model is constructed with the maximum carrying capacity as the constraint and the expected aquaculture density coefficient as the objective function, and the optimal solution of the simplex table model is solved to determine the optimal aquaculture density scale of the high-density abnormal aquaculture area;

[0009] S108: Hash misplacement calculates the to-be-balanced scale of the current aquaculture density scale relative to the optimal aquaculture density scale, and eliminates the to-be-balanced scale as a target constraint to explore the optimal aquaculture optimization parameters of different associated aquaculture planning factors in the experience aquaculture case, and optimizes the high-density aquaculture of Marsupenaeus japonicus according to the optimal aquaculture optimization parameters.

[0010] More specifically, the S102 step specifically includes the following steps:

[0011] Obtain the target aquaculture area of Marsupenaeus japonicus, and obtain the multi-source cultivation environment data of Marsupenaeus japonicus in the target aquaculture area through Internet of Things technology.

[0012] Extract the high-dimensional data space of the target aquaculture area, calculate the Manhattan distance between each local data point and the adjacent local data point in the high-dimensional data space, and take the adjacent local data point with a Manhattan distance less than a preset Manhattan distance as K nearest neighbors. A local scale probability distribution is constructed for each local data point based on the K nearest neighbors, and a local neighborhood graph is generated for each multi-source cultivation environment data.

[0013] Obtain the local area perception strategy of Internet of Things technology, and establish a joint fuzzy network for perception of different cultivation environment data sources in the target aquaculture area based on the local area perception strategy. The local neighborhood graph of each multi-source cultivation environment data is fused through the joint fuzzy network topology, and a high-dimensional fuzzy joint graph of the approximate spatial distribution of the multi-source cultivation environment data in the target aquaculture area is output.

[0014] Based on big data, obtain the trend criterion of different cultivation environments of Marsupenaeus japonicus in high-density aquaculture, and simultaneously obtain the current aquaculture density data of Marsupenaeus japonicus in the target aquaculture area. Introduce a logistic regression algorithm to fit the compromise processing tendency score of each multi-source cultivation environment data that changes with aquaculture density, with the current aquaculture density data as the processing variable and the trend criterion as the covariate.

[0015] Based on the compromise processing tendency score, a low-dimensional connection weight of the local data point is determined, and a low-dimensional embedding space of the target aquaculture area is constructed according to the low-dimensional connection weight. The high-dimensional fuzzy joint graph is mapped to the low-dimensional embedding space only under the premise that the high-dimensional connection weight and the low-dimensional connection weight are consistent with each other, and a trend radiation graph of the multi-source cultivation environment is obtained.

[0016] More specifically, the S104 step specifically includes the following steps:

[0017] dividing the target aquaculture area into a plurality of sub-aquaculture areas based on the current breeding density, and obtaining a trend radiation index of each sub-aquaculture area about the multi-source cultivation environment reflecting the rationality of the breeding density through a trend radiation diagram;

[0018] obtaining a scale evaluation system of different preset cultivation environment condition combinations for high-density M. rosenbergii aquaculture, constructing an aquaculture density scale evaluation model according to the scale evaluation system, and performing targeted evaluation of the trend radiation index in different degrees through the aquaculture density scale evaluation model to generate a series of aquaculture density scale evaluation values of the trend radiation index caused by M. rosenbergii in each sub-aquaculture area;

[0019] obtaining variety information and phenotype specification parameters of M. rosenbergii in each sub-aquaculture area, retrieving a thermal color gamut reference table about different unit density cultivation of the current M. rosenbergii variety based on the variety information and phenotype specification parameters in a big data network, assigning a scale space rendering scale of each sub-aquaculture area based on the thermal color gamut reference table, and constructing a thermal space adjacency matrix of the target aquaculture area;

[0020] weighting and aggregating a certain aquaculture density scale evaluation value in the thermal space adjacency matrix to the aquaculture density scale evaluation values of the mutual neighbors according to the scale space rendering scale, and repeating the weighting and aggregation step until all the aquaculture density scale evaluation values are rendered, to obtain an aquaculture density scale thermal map of each sub-aquaculture area;

[0021] extracting an RGB chroma value corresponding to the maximum aquaculture density scale of each sub-aquaculture area from the aquaculture density scale thermal map, defining it as a limit RGB chroma value, and obtaining the current aquaculture density scale in each sub-aquaculture area and querying the corresponding RGB chroma value, defining it as a present RGB chroma value;

[0022] if the present RGB chroma value has exceeded the limit RGB chroma value, the sub-aquaculture area is marked as a high-density abnormal aquaculture area, and a multi-source cultivation environment analysis result is obtained.

[0023] More specifically, the S106 step specifically includes the following steps:

[0024] If the multi-source cultivation environment analysis result shows that there is at least one and more high-density abnormal aquaculture area, a water environment model is introduced to calculate the maximum carrying capacity of the water environment in the high-density abnormal aquaculture area.

[0025] obtaining the directional breeding demand of M. rosenbergii, extracting the expected aquaculture density scale of M. rosenbergii through the directional breeding demand, and obtaining a density scale-density coefficient-unit density scalar mapping table based on a big data network;

[0026] The desired aquaculture density scale is queried using the density scale-density coefficient-unit density scalar mapping table, and several potential density coefficients and the corresponding potential unit density scalars are obtained when the desired aquaculture density scale is reached.

[0027] Using the desired aquaculture density coefficient as the objective function, several potential density coefficients as coefficients, and each potential unit density scalar as a constant variable, a simplex tableau model based on the penalty constraint function of maximum carrying capacity is constructed, which relies on the objective function row, coefficient matrix, and constant terms. The bottom row of the objective function row in the simplex tableau model is then examined to see one or more density coefficients recorded therein.

[0028] If any density coefficient is greater than 0, the simplex method is introduced to select the unit density scalar of the constant variable corresponding to the maximum density coefficient from the fastest growth direction of the penalty constraint function to perform the minimum ratio objective function line test and solve, so as to approximate the maximum carrying capacity to calculate the optimal solution to achieve the desired breeding density scale, and output the input and output variables.

[0029] Repeat the above steps of verifying the density coefficients and calculating the introduced and removed variables to continuously update the simplex tableau model until there is no density coefficient greater than 0, thus obtaining the terminated simplex tableau model.

[0030] By terminating the simplex tableau model, a series of optimal aquaculture density control solutions are determined. These solutions are then used to optimize the current aquaculture density scale, thus obtaining the optimal aquaculture density scale for the high-density abnormal aquaculture area.

[0031] More specifically, step S108 includes the following steps:

[0032] A hash misalignment algorithm is introduced. Based on the species information and phenotypic specifications of the tiger prawn, the scale misalignment hash function of the current farming density relative to the optimal farming density is quantitatively calculated in the hash misalignment algorithm. The scale misalignment hash function is used to determine the scale to be balanced for high-density abnormal farming.

[0033] The preset aquaculture plan currently being implemented for high-density cultured tiger prawns is obtained. Based on big data, a tiger prawn aquaculture knowledge graph is obtained. The tiger prawn aquaculture knowledge graph is used to identify the correlation between the scale to be balanced and each aquaculture plan factor in the preset aquaculture plan, and the correlation degree of each aquaculture plan factor is obtained.

[0034] If the correlation is greater than the preset correlation, the aquaculture planning factor is labeled as a related aquaculture planning factor; if the correlation is less than the preset correlation, the aquaculture planning factor is labeled as an unrelated aquaculture planning factor. The set of related aquaculture planning factors and the set of unrelated aquaculture planning factors that achieve the optimal aquaculture density scale for the scale to be balanced are obtained, and the factor ratio between the set of related aquaculture planning factors and the set of unrelated aquaculture planning factors is extracted.

[0035] Based on big data, several empirical cases of high-density tiger prawn farming were obtained. Through these empirical cases, the parameter range of farming cases to achieve the scale to be balanced was extracted. According to the factor proportion, the parameter range of farming cases was divided into exploration mode and tracking mode for optimization parameters.

[0036] The parameters of aquaculture cases with the constraint criterion of canceling the scale to be balanced were explored locally and globally using exploratory mode and tracking mode respectively, resulting in the first-class and second-class exploratory patterns of aquaculture case parameters;

[0037] The global optimal solution for the range of aquaculture case parameters is updated based on the first and second exploration patterns of the aquaculture case parameters. The optimal aquaculture optimization parameters for eliminating the unbalanced scale of different related aquaculture planning factors are obtained. The high-density aquaculture of tiger prawns is optimized based on the optimal aquaculture optimization parameters.

[0038] More specifically, the process of using exploration mode and tracking mode to conduct local and global explorations of the parameters of aquaculture cases with the constraint of canceling the scale to be balanced, respectively, to obtain the first-class and second-class exploration patterns of the aquaculture case parameters, specifically includes the following steps:

[0039] For the exploration mode, the perturbation limit of the parameter range of the breeding case, different change dimensions, and the number of dimensions of each change dimension are preset. The unbalanced scale is removed as the constraint criterion. Based on the number of dimensions, the constraint criterion of each change dimension is explored by using each breeding case parameter in the parameter range of the breeding case to perform local slight perturbation exploration until the perturbation limit is reached, and multiple first new exploration sites are obtained.

[0040] For the tracking mode, a safe speed is preset for the parameter range of the breeding case. According to the constraint criteria, the predetermined position and predetermined speed of each breeding case parameter in different changed dimensions are set within the parameter range of the breeding case. Based on the predetermined position and predetermined speed, the constraint criteria are explored globally from the current global position in each changed dimension, and the predetermined speed is always kept within the safe speed to obtain multiple second new exploration sites.

[0041] The fitness of each first and second nascent exploration site is evaluated by constraint criteria. Only nascent exploration sites with fitness greater than the preset fitness are extracted and marked as the optimal nascent exploration sites. The optimal nascent exploration sites replace the current exploration position of the constraint criteria to generate the first and second exploration patterns of aquaculture case parameters.

[0042] The second aspect of the present invention provides a high-density culture optimization system for Penaeus monodon based on multi-source culture environment analysis. The high-density culture optimization system for Penaeus monodon includes a memory and a processor. The memory stores a program for a high-density culture optimization method for Penaeus monodon based on multi-source culture environment analysis. When the program for the high-density culture optimization method for Penaeus monodon is executed by the processor, the steps of the high-density culture optimization method for Penaeus monodon described in any one of the present invention are implemented.

[0043] This invention addresses the technical deficiencies in the prior art, and its beneficial technical effects are as follows:

[0044] Multi-source breeding environment data of tiger prawns in the target breeding area are acquired. The multi-source breeding environment data is fused and mapped from a high-dimensional space to a low-dimensional embedding space in which the multi-source breeding environment changes due to breeding density, resulting in a trend radiation map of the multi-source breeding environment. The trend radiation map is evaluated based on a scale evaluation system for creating different preset breeding environment conditions for high-density tiger prawn breeding. After evaluation, a thermal aggregation analysis is performed to determine whether the current breeding density of tiger prawns is reasonable, resulting in a multi-source breeding environment analysis result. If the multi-source breeding environment analysis result shows one or more abnormal high-density breeding areas, a simplex tableau model is constructed with the maximum carrying capacity as a constraint and the expected breeding density coefficient as the objective function. The optimal solution of the simplex tableau model is solved to determine the optimal breeding density of the abnormal high-density breeding areas. The unbalanced scale of the current breeding density scale relative to the optimal breeding density scale is calculated using hash misalignment. With the elimination of the unbalanced scale scale as the objective constraint, the optimal breeding optimization parameters for different associated breeding planning factors are explored in empirical breeding cases. The high-density breeding of tiger prawns is optimized based on the optimal breeding optimization parameters. This invention integrates multi-source cultivation environment data to conduct rational and scientific analysis of high-density Penaeus monodon farming and make precise optimizations to modern farming strategies, thereby optimizing the farming conditions of Penaeus monodon, increasing the yield of Penaeus monodon per unit area, significantly reducing mortality and disease risk, and improving the sustainable utilization rate of energy, water and other resources. Attached Figure Description

[0045] To more clearly illustrate the technical solutions in the embodiments of the present invention or the prior art, the drawings used in the description of the embodiments or the prior art will be briefly introduced below. Obviously, the drawings described below are only some embodiments of the present invention. For those skilled in the art, other embodiments can be obtained from these drawings without creative effort.

[0046] Figure 1 A flowchart of the first method for optimizing high-density culture of Penaeus monodon based on multi-source culture environment analysis is shown.

[0047] Figure 2 The second method flowchart of the optimization method for high-density culture of Penaeus monodon based on multi-source culture environment analysis is shown;

[0048] Figure 3 A system framework diagram of the high-density culture optimization system for Penaeus monodon based on multi-source culture environment analysis is shown. Detailed Implementation

[0049] To better understand the above-mentioned objectives, features, and advantages of the present invention, the present invention will be further described in detail below with reference to the accompanying drawings and specific embodiments. It should be noted that, unless otherwise specified, the embodiments and features described in these embodiments can be combined with each other.

[0050] Many specific details are set forth in the following description in order to provide a full understanding of the invention. However, the invention may also be practiced in other ways different from those described herein, and therefore the scope of protection of the invention is not limited to the specific embodiments disclosed below.

[0051] The first aspect of this invention provides an optimization method for high-density culture of Penaeus monodon based on multi-source culture environment analysis, such as... Figure 1 As shown, it includes the following steps:

[0052] S102: Obtain multi-source breeding environment data of tiger prawns in the target breeding area, fuse and map the multi-source breeding environment data from the high-dimensional space to the low-dimensional embedding space in which the multi-source breeding environment changes due to breeding density, and obtain the trend radiation map of the multi-source breeding environment.

[0053] S104: The trend radiation diagram is evaluated based on the scale evaluation system of creating different preset breeding environment conditions for high-density tiger prawn farming. After evaluation, the current breeding density of tiger prawns is analyzed to determine whether it is reasonable, and the results of multi-source breeding environment analysis are obtained.

[0054] S106: If the multi-source breeding environment analysis results show one or more high-density abnormal breeding areas, then a simplex tableau model is constructed with the maximum carrying capacity as a constraint and the expected breeding density coefficient as the objective function. The optimal solution of the simplex tableau model is solved to determine the optimal breeding density scale of the high-density abnormal breeding area.

[0055] S108: Hash misalignment calculation of the unbalanced size of the current breeding density relative to the optimal breeding density, with the goal of eliminating the unbalanced size, explores the optimal breeding optimization parameters for different associated breeding planning factors in empirical breeding cases, and optimizes the high-density breeding of tiger prawns based on the optimal breeding optimization parameters.

[0056] More specifically, step S102, as follows: Figure 2 As shown, the specific steps include:

[0057] S202: Obtain the target farming area of ​​tiger prawns and acquire multi-source breeding environment data of tiger prawns in different local high-density farming locations within the target farming area through Internet of Things technology;

[0058] S204: Extract the high-dimensional data space of the target aquaculture area, calculate the Manhattan distance between each local data point and its neighboring local data points in the high-dimensional data space, take the neighboring local data points whose Manhattan distance is less than the preset Manhattan distance as the K nearest neighbors, construct a local scale probability distribution for each local data point based on the K nearest neighbors, and generate a local neighborhood map for each multi-source aquaculture environment data.

[0059] S206: Obtain the local perception strategy of IoT technology, establish a joint fuzzy network for perceiving different breeding environment data sources in the target breeding area based on the local perception strategy, fuse the local neighborhood graph of each multi-source breeding environment data through the topology of the joint fuzzy network, and output a high-dimensional fuzzy joint graph of the approximate spatial distribution of multi-source breeding environment data in the target breeding area.

[0060] S208: Based on big data, trend criteria for different breeding environments during high-density farming of Penaeus monodon are obtained. The current breeding density data of Penaeus monodon in the target breeding area is obtained simultaneously. Logistic regression algorithm is introduced to fit the compromise processing tendency score of each multi-source breeding environment data that changes due to breeding density with the current breeding density data as the processing variable and the trend criteria as the covariate.

[0061] S210: Based on the compromise processing tendency score, determine the low-dimensional connection weights of local data points, construct the low-dimensional embedding space of the target breeding area according to the low-dimensional connection weights, and map the high-dimensional fuzzy joint graph to the low-dimensional embedding space while maintaining the consistency between the high-dimensional connection weights and the low-dimensional connection weights, to obtain the trend radiation map of the multi-source breeding environment.

[0062] It should be noted that while high-density farming of tiger prawns can significantly increase yield per unit area, it also faces several negative impacts on the farming environment. For example, high-density farming can lead to an accumulation of excrement and uneaten feed, which can increase the levels of toxic substances. In addition, pathogens (such as white spot disease virus and Vibrio luminifera) can spread rapidly in high-density environments, which can easily lead to a sharp increase in the mortality rate of tiger prawns. Therefore, the trend of multi-source breeding environment indirectly reflects the rationality of high-density farming of Penaeus monodon. Thus, this method analyzes the current high-density farming status of Penaeus monodon from the perspective of multi-source breeding environment. This method first integrates the multi-source breeding environment data monitored in the target farming area into a global fuzzy graph of adjacent retention pattern according to the local adjacency of spatial location, that is, a high-dimensional fuzzy joint graph, thereby forming a topological approximation structure in a high-dimensional space. This high-dimensional fuzzy joint graph can be regarded as the fuzzy skeleton of the rationality of the multi-source breeding environment data for the evaluation of high-density farming of Penaeus monodon in the high-dimensional space of the target farming area. It captures the spatial trend topological structure of the entire multi-source breeding environment dataset, thereby constructing an approximate manifold structure of multi-source breeding environment data in high-dimensional space, realizing the fusion transformation of multi-source breeding environment from local data capture to global data modeling. Subsequently, based on the trend criteria of different cultivation environments during high-density farming of Penaeus monodon, variable analysis was conducted on the current farming density data. This reflects the compromise tendency of diverse cultivation environments with Penaeus monodon farming density, thereby revealing the degree of impact of changes in current farming density on different multi-source cultivation environments. This provides a low-dimensional embedding space, a low-dimensional trend expression underlying logic space for the approximate high-dimensional trend structure of multi-source cultivation environment data in space. This enables the monitoring of multi-source cultivation environment data to be further visualized and mapped into a trend radiation map that can reflect the rationality of high-density farming of Penaeus monodon from the perspective of multi-source cultivation environments, providing a highly credible analytical basis for subsequent high-density farming.

[0063] It should be noted that a local-scale probability distribution is constructed for each local data point based on the K nearest neighbors whose Manhattan distance is less than the preset Manhattan distance. This reflects the local similarity of the multi-source breeding environment. The connection weights between data structure points are the joint of these local-scale probability distributions, thus associating each pair of points with a connection strength. This makes the structural topology of the trend changes in the multi-source breeding environment caused by high-density aquaculture more coherent and realistic, improving the continuity, authenticity, and completeness of the data structure in the high-dimensional space. By maintaining the consistency between the high-dimensional and low-dimensional connection weights, the high-dimensional fuzzy joint graph is mapped to the low-dimensional embedding space. This allows data points with similar (high connection strength) multi-source breeding environments in the high-dimensional space to be closer in the low-dimensional space, while unrelated points in the high-dimensional space are farther away. This significantly reduces dimensionality while preserving proximity relationships, thus forming a "radial cluster" trend image in visualization. For example, under a certain degree of high-density aquaculture, microbial activity may show a dramatic increase, making the carbonate system unstable. This improves the trend expression ability and accuracy of the multi-source breeding environment in reflecting the high-density aquaculture status of Penaeus monodon.

[0064] More specifically, step S104 includes the following steps:

[0065] Based on the current breeding density, the target breeding area is divided into several sub-breeding areas. The trend radiation index of each sub-breeding area reflecting the rationality of the breeding density in the multi-source breeding environment is obtained through the trend radiation map.

[0066] A scale evaluation system for creating different preset breeding environment conditions for high-density Penaeus monodon farming is obtained. A farming density scale assessment model is constructed according to the scale evaluation system. The trend radiation index is evaluated to different degrees through the farming density scale assessment model, and a series of farming density scale assessment values ​​are generated for the trend radiation index caused by Penaeus monodon in each sub-farming area.

[0067] Obtain the species information and phenotypic specifications of tiger prawns in each sub-culture area. Based on the species information and phenotypic specifications, retrieve the thermal color gamut reference table for different unit density culture of the current tiger prawn species in the big data network. Based on the thermal color gamut reference table, allocate the scale space rendering scale of each sub-culture area and construct the thermal space adjacency matrix of the target culture area.

[0068] In the thermal adjacency matrix, a certain aquaculture density scale assessment value is weighted and aggregated with the aquaculture density scale assessment values ​​of its neighbors according to the scale spatial rendering scale. The above weighted aggregation steps are repeated until all aquaculture density scale assessment values ​​are rendered, and the aquaculture density scale heatmap of each sub-aquaculture area is obtained.

[0069] The RGB chromaticity values ​​corresponding to the maximum breeding density scale of each sub-breeding area are extracted by the breeding density scale heat map and defined as the limit RGB chromaticity value. The current breeding density scale of each sub-breeding area is obtained and the corresponding RGB chromaticity value is queried and defined as the current RGB chromaticity value.

[0070] If the current RGB chromaticity value has exceeded the limit RGB chromaticity value, then the sub-culture area is marked as a high-density abnormal culture area, and the multi-source culture environment analysis results are obtained.

[0071] It should be noted that while trend radiation maps can be used to observe the rationality of high-density farming of Penaeus monodon under different cultivation environments, traditional multi-source cultivation environment analysis methods often rely on human experience, resulting in significant judgment bias. To address this, this method first extracts relevant indicators reflecting the rationality of farming density—the trend radiation index—from the constructed trend radiation map. Notably, this method divides the target farming area into several sub-farming areas based on the distribution differences of the current farming density, enabling more detailed anomaly detection in high-density Penaeus monodon farming and improving the efficiency of global collaborative analysis. As illustrated by the examples above, high-density farming of Penaeus monodon can lead to trend changes in the rearing environment. Therefore, this method constructs a farming density scale assessment model by referencing a scale evaluation system for creating different combinations of preset rearing environment conditions in high-density Penaeus monodon farming. This assessment model is then used to further evaluate the model, determining the farming density scale assessment value for each sub-farming area, representing the trend radiation index of Penaeus monodon causing the surrounding rearing environment. This series of farming density scale assessment values ​​is a preliminary quantitative density inference model based on known rearing environment data for the farming and growth of Penaeus monodon in different regions, thus enabling the formulation of a maximum farming density summary for each sub-farming area that conforms to the trend of the surrounding rearing environment. Subsequently, according to the scale rendering scale of the heat map color gamut reference table for different unit density farming of the current Penaeus monodon species, the series of farming density scale assessment values ​​are fitted to the corresponding sub-farming areas, thereby creating a heat map-style farming density chart for each sub-farming area, i.e., a farming density scale heat map. If the current RGB chromaticity value corresponding to the current stocking density in a sub-culture area exceeds the limit RGB chromaticity value corresponding to the maximum stocking density as interpreted in the heatmap of that stocking density, it indicates that the current stocking density in that sub-culture area exceeds the high-density stocking threshold of the surrounding culture environment. This can easily lead to a series of negative anomalies in the Penaeus monodon stocking in that area, such as the accumulation of ammonia nitrogen and nitrite, insufficient dissolved oxygen, drastic pH fluctuations, and increased stress response. Therefore, it is designated as a high-density abnormal stocking area. This method can define a specific maximum critical reasonable high-density stocking limit for Penaeus monodon in different regions based on the trend radiation feedback of multi-source culture environments. Using this limit, the reasonableness of the current stocking density can be determined, thereby quickly locating high-density abnormal areas in the real-time Penaeus monodon stocking process and achieving the source tracing effect of high-density stocking anomalies based on multi-source culture environment analysis.

[0072] More specifically, step S106 includes the following steps:

[0073] If the multi-source aquaculture environment analysis results show that there is at least one or more high-density abnormal aquaculture areas, then a water environment model is introduced to calculate the high-density abnormal aquaculture areas and obtain the maximum carrying capacity of the water environment within the high-density abnormal aquaculture areas.

[0074] Obtain the targeted aquaculture demand of tiger prawns, extract the expected aquaculture density scale of tiger prawns from the targeted aquaculture demand, and obtain a density scale-density coefficient-unit density scalar mapping table based on big data network;

[0075] The desired aquaculture density scale is queried using the density scale-density coefficient-unit density scalar mapping table, and several potential density coefficients and the corresponding potential unit density scalars are obtained when the desired aquaculture density scale is reached.

[0076] Using the desired aquaculture density coefficient as the objective function, several potential density coefficients as coefficients, and each potential unit density scalar as a constant variable, a simplex tableau model based on the penalty constraint function of maximum carrying capacity is constructed, which relies on the objective function row, coefficient matrix, and constant terms. The bottom row of the objective function row in the simplex tableau model is then examined to see one or more density coefficients recorded therein.

[0077] If any density coefficient is greater than 0, the simplex method is introduced to select the unit density scalar of the constant variable corresponding to the maximum density coefficient from the fastest growth direction of the penalty constraint function to perform the minimum ratio objective function line test and solve, so as to approximate the maximum carrying capacity to calculate the optimal solution to achieve the desired breeding density scale, and output the input and output variables.

[0078] Repeat the above steps of verifying the density coefficients and calculating the introduced and removed variables to continuously update the simplex tableau model until there is no density coefficient greater than 0, thus obtaining the terminated simplex tableau model.

[0079] By terminating the simplex tableau model, a series of optimal aquaculture density control solutions are determined. These solutions are then used to optimize the current aquaculture density scale, thus obtaining the optimal aquaculture density scale for the high-density abnormal aquaculture area.

[0080] It should be noted that when one or more abnormally high-density farming areas exist within the target farming area, it indicates an unreasonable localized high-density distribution of Penaeus monodon within that area. This makes the aquatic environment in that area unable to support its normal and rational farming growth. Therefore, the Penaeus monodon density in the abnormally high-density farming areas should be maintained within a suitable range. Optimizing this suitable range requires considering both the maximum carrying capacity of the aquatic environment and maximizing the achievement of desired requirements. This ensures that farming needs are met while minimizing the exceedance of the aquatic environment's carrying capacity for high-density farming. To address this, this method obtains a density scale-density coefficient-unit density scalar mapping table. This table describes the proposed density coefficients, unit density quantities, distribution, and range (forming unit density scalars) under different density scales. Using this mapping table, the density coefficients and unit density scalars that meet the desired farming density scale can be extracted using the density scale as an index, thus providing reliable scale specifications and parameter options for subsequent optimization of the optimal farming density scale. Subsequently, a simplex tableau model is constructed using the retrieved density coefficients and unit density scalars. This model combines the objective function row, coefficient matrix, and constant terms. The method also incorporates the simplex method, an iterative algorithm for solving linear programming problems. The simplex method accurately determines the maximum value of the desired stocking density coefficient (objective function) for tiger prawn farming under the constraint of maximum carrying capacity (linear constraint). This simplex tableau model provides a standard data input format for the simplex method. The penalty constraint function based on maximum carrying capacity restricts the growth direction of the objective function, ensuring that the optimal value is obtained under the penalty constraint. This guarantees that density adjustments simultaneously meet both farming needs and the carrying capacity of the aquatic environment, improving the adaptability and appropriateness of abnormal stocking density optimization.

[0081] It should be noted that if any density coefficient is greater than 0, it indicates that there is still room for improvement in the objective function through non-basic variables. This means that the density coefficient and its corresponding unit density scalar, as a feasible solution, are not the optimal solution. Further, the simplex method should be used to select the unit density scalar of the constant variable corresponding to the largest density coefficient from the direction of fastest growth of the penalty constraint function to perform a minimum ratio objective function test. This identifies the variable that improves the current objective function the fastest (i.e., the one that contributes the most to the growth of the objective function), ensuring that each iteration moves towards the optimal value of the objective function. The minimum ratio method effectively maintains the feasibility of the optimized solution, ensuring that each solution is still feasible. This method can optimize the current unreasonable farming density in high-density aquaculture areas, ensuring that it meets the expected farming density requirements without exceeding the carrying capacity of the aquatic environment. This improves the rationality and accuracy of regional high-density farming of tiger prawns and reduces the frequency of anomalies (water quality deterioration, high disease incidence, high stress, decreased feed conversion rate, or fierce competition for food) in high-density tiger prawn farming.

[0082] More specifically, step S108 includes the following steps:

[0083] A hash misalignment algorithm is introduced. Based on the species information and phenotypic specifications of the tiger prawn, the scale misalignment hash function of the current farming density relative to the optimal farming density is quantitatively calculated in the hash misalignment algorithm. The scale misalignment hash function is used to determine the scale to be balanced for high-density abnormal farming.

[0084] The preset aquaculture plan currently being implemented for high-density cultured tiger prawns is obtained. Based on big data, a tiger prawn aquaculture knowledge graph is obtained. The tiger prawn aquaculture knowledge graph is used to identify the correlation between the scale to be balanced and each aquaculture plan factor in the preset aquaculture plan, and the correlation degree of each aquaculture plan factor is obtained.

[0085] If the correlation is greater than the preset correlation, the aquaculture planning factor is labeled as a related aquaculture planning factor; if the correlation is less than the preset correlation, the aquaculture planning factor is labeled as an unrelated aquaculture planning factor. The set of related aquaculture planning factors and the set of unrelated aquaculture planning factors that achieve the optimal aquaculture density scale for the scale to be balanced are obtained, and the factor ratio between the set of related aquaculture planning factors and the set of unrelated aquaculture planning factors is extracted.

[0086] Based on big data, several empirical cases of high-density tiger prawn farming were obtained. Through these empirical cases, the parameter range of farming cases to achieve the scale to be balanced was extracted. According to the factor proportion, the parameter range of farming cases was divided into exploration mode and tracking mode for optimization parameters.

[0087] The parameters of aquaculture cases with the constraint criterion of canceling the balance scale were explored locally and globally using exploratory mode and tracking mode respectively, resulting in the first-class and second-class exploratory patterns of aquaculture case parameters;

[0088] The global optimal solution for the range of aquaculture case parameters is updated based on the first and second exploration patterns of the aquaculture case parameters. The optimal aquaculture optimization parameters for eliminating the unbalanced scale of different associated aquaculture planning factors are obtained. The high-density aquaculture of tiger prawns is optimized based on the optimal aquaculture optimization parameters.

[0089] It should be noted that high-density culture of tiger prawns is typically regulated according to a pre-set culture plan. Adjusting parameters at different levels optimizes the high-density culture behavior of tiger prawns, thereby further improving the cultivation environment. However, existing culture optimization techniques usually require manual calculation and analysis to adjust parameters, which significantly increases the output of human and material resources. Furthermore, high-density culture optimization relying on manual experience can produce significant errors, easily backfiring and worsening the surrounding cultivation environment, reducing the survival rate, individual quality, and yield per unit area of ​​tiger prawns in high-density culture. To address this, this method calculates the scale misalignment between the determined optimal culture density and the current actual culture density, i.e., the unbalanced scale quantity. This unbalanced scale quantity is a necessary optimization prerequisite; only by eliminating this unbalanced scale quantity can the current culture density within the region reach the optimal scale. Adjusting the density of tiger prawns typically requires targeted aquaculture planning elements. For example, adjusting and optimizing temperature and salinity can alleviate the aggregation of tiger prawns, thereby reducing abnormally high densities. Therefore, this method further utilizes a tiger prawn aquaculture knowledge graph to identify the scale to be balanced and each aquaculture planning factor in the preset aquaculture plan. Aquaculture planning factors generally include multiple aspects such as water quality management, temperature control, salinity control, bottom sediment management, feed management, nutrition optimization, feeding and digestion promotion, immune enhancement, and microecological management. This allows us to obtain the relevant aquaculture planning factors that correspond to and match the scale to be balanced, as well as the non-relevant aquaculture planning factors with low matching degree. This clarifies the necessary factor components for eliminating the scale to be balanced, ensuring the targeting and directionality of high-density aquaculture optimization.

[0090] It should be noted that the factor ratio between the associated and unassociated aquaculture planning factor sets defines the mode ratio, which can divide the parameter range of the aquaculture case to be explored into exploration mode and tracking mode for optimizing parameters. This dynamically balances local optimization (fine search) and global search (escaping local optima), ensuring the diversity of the starting point for optimal parameters, preventing getting trapped in local optima, and improving the balance between subsequent local and global search behaviors. Since the ultimate goal of the exploration is to remove the unbalanced scale, this method uses the removal of the unbalanced scale as a constraint criterion. It uses each aquaculture case parameter in the parameter range to perform local exploration in exploration mode and global exploration in tracking mode to search for aquaculture case parameters and obtain the exploration endpoint network, namely the first-class and second-class exploration patterns of aquaculture case parameters. These first-class and second-class exploration patterns of aquaculture case parameters reveal the locked range of the global optimal solution, making the location of the best aquaculture optimization parameters more accurate. This improves the reliability and accuracy of high-density aquaculture optimization for tiger prawns, reduces optimization bias, reduces mortality and low survival rates of tiger prawns in high-density aquaculture environments, and comprehensively increases yield and production capacity per unit area.

[0091] More specifically, the process of using exploration mode and tracking mode to conduct local and global explorations of the parameters of aquaculture cases with the constraint of canceling the scale to be balanced, respectively, to obtain the first-class and second-class exploration patterns of the aquaculture case parameters, specifically includes the following steps:

[0092] For the exploration mode, the perturbation limit of the parameter range of the breeding case, different change dimensions, and the number of dimensions of each change dimension are preset. The unbalanced scale is removed as the constraint criterion. Based on the number of dimensions, the constraint criterion of each change dimension is explored by using each breeding case parameter in the parameter range of the breeding case to perform local slight perturbation exploration until the perturbation limit is reached, and multiple first new exploration sites are obtained.

[0093] For the tracking mode, a safe speed is preset for the parameter range of the breeding case. According to the constraint criteria, the predetermined position and predetermined speed of each breeding case parameter in different changed dimensions are set within the parameter range of the breeding case. Based on the predetermined position and predetermined speed, the constraint criteria are explored globally from the current global position in each changed dimension, and the predetermined speed is always kept within the safe speed to obtain multiple second new exploration sites.

[0094] The fitness of each first and second nascent exploration site is evaluated by constraint criteria. Only nascent exploration sites with fitness greater than the preset fitness are extracted and marked as the optimal nascent exploration sites. The optimal nascent exploration sites replace the current exploration position of the constraint criteria to generate the first and second exploration patterns of aquaculture case parameters.

[0095] It should be noted that by utilizing each breeding case parameter within the breeding case parameter range to perform local exploration in exploration mode and global exploration in tracking mode, each breeding case parameter can be regarded as an individual cat. These individual cats are assigned to exploration and tracking modes to capture prey (optimal breeding parameters). The local slight perturbation exploration mode for each breeding case parameter simulates the behavior of a cat observing its surroundings while stationary. It allows the cat to try multiple slightly perturbed new locations near its current position, selecting the optimal one. This achieves local tentative optimization of the solution, enhancing search accuracy and uncovering more potential optimal solutions in promising areas. This effectively avoids missing nearby optimal solutions due to global jumps. The global exploration tracking mode for breeding case parameters simulates the behavior of a cat chasing prey after discovering a target. The cat approaches the individual with the best breeding parameters in the current group, updating its position and speed, thereby guiding the group towards potentially optimal areas. This global search effectively maintains search efficiency and promotes the improvement of the overall quality of optimal solutions. Changing dimensions controls how many dimensions are selected and modified in each local perturbation, enabling directional perturbation of the search solution space while maintaining strong local search capabilities. Ultimately, the first and second exploration patterns discovered through local and global explorations constitute the precise value range of the parameters for the aquaculture case.

[0096] The second aspect of this invention provides an optimization system for high-density culture of Penaeus monodon based on multi-source culture environment analysis, such as... Figure 3 As shown, the high-density aquaculture optimization system for tiger prawns includes a memory 31 and a processor 32. The memory 31 stores a program for a high-density aquaculture optimization method for tiger prawns based on multi-source breeding environment analysis. When the program for the high-density aquaculture optimization method for tiger prawns is executed by the processor 32, any of the steps of the high-density aquaculture optimization method for tiger prawns described above are implemented.

[0097] The above are merely specific embodiments of the present invention, but the scope of protection of the present invention is not limited thereto. Any variations or substitutions that can be easily conceived by those skilled in the art within the technical scope disclosed in the present invention should be included within the scope of protection of the present invention. Therefore, the scope of protection of the present invention should be determined by the scope of the claims.

Claims

1. A method for optimizing high-density culture of Penaeus monodon based on multi-source culture environment analysis, characterized in that, Includes the following steps: S102: Obtain multi-source breeding environment data of tiger prawns in the target breeding area, fuse and map the multi-source breeding environment data from the high-dimensional space to the low-dimensional embedding space in which the multi-source breeding environment changes due to breeding density, and obtain the trend radiation map of the multi-source breeding environment. S104: The trend radiation diagram is evaluated based on the scale evaluation system of creating different preset breeding environment conditions for high-density tiger prawn farming. After evaluation, the current breeding density of tiger prawns is analyzed to determine whether it is reasonable, and the results of multi-source breeding environment analysis are obtained. S106: If the multi-source breeding environment analysis results show one or more high-density abnormal breeding areas, then a simplex tableau model is constructed with the maximum carrying capacity as a constraint and the expected breeding density coefficient as the objective function. The optimal solution of the simplex tableau model is solved to determine the optimal breeding density scale of the high-density abnormal breeding area. S108: Hash misalignment calculation of the unbalanced size of the current breeding density relative to the optimal breeding density, with the goal of eliminating the unbalanced size, explores the optimal breeding optimization parameters for different associated breeding planning factors in empirical breeding cases, and optimizes the high-density breeding of tiger prawns based on the optimal breeding optimization parameters.

2. The method for optimizing high-density culture of Penaeus monodon based on multi-source culture environment analysis according to claim 1, characterized in that, Step S102 specifically includes the following steps: The target farming area for tiger prawns was identified, and multi-source breeding environment data of tiger prawns in different local high-density farming locations within the target farming area were obtained through Internet of Things (IoT) technology. Extract the high-dimensional data space of the target aquaculture area, calculate the Manhattan distance between each local data point and its neighboring local data points in the high-dimensional data space, and take the neighboring local data points whose Manhattan distance is less than the preset Manhattan distance as the K nearest neighbors. Construct a local scale probability distribution for each local data point based on the K nearest neighbors to generate a local neighborhood map for each multi-source aquaculture environment data. The local perception strategy of IoT technology is obtained, and a joint fuzzy network for sensing different breeding environment data sources in the target breeding area is established based on the local perception strategy. The local neighborhood graph of each multi-source breeding environment data is fused through the topology of the joint fuzzy network, and a high-dimensional fuzzy joint graph of the approximate spatial distribution of multi-source breeding environment data in the target breeding area is output. Based on big data, trend criteria for different breeding environments during high-density farming of Penaeus monodon are obtained. Simultaneously, the current breeding density data of Penaeus monodon in the target breeding area is obtained. Logistic regression algorithm is introduced to fit the compromise processing tendency score of each multi-source breeding environment data as a result of changes in breeding density, with the current breeding density data as the processing variable and the trend criteria as the covariate. Based on the compromise processing tendency score, low-dimensional connection weights are determined for local data points. A low-dimensional embedding space for the target breeding area is constructed based on the low-dimensional connection weights. Under the premise of maintaining the consistency between the high-dimensional connection weights and the low-dimensional connection weights, the high-dimensional fuzzy joint graph is mapped to the low-dimensional embedding space to obtain the trend radiation map of the multi-source breeding environment.

3. The method for optimizing high-density culture of Penaeus monodon based on multi-source culture environment analysis according to claim 1, characterized in that, Step S104 specifically includes the following steps: Based on the current breeding density, the target breeding area is divided into several sub-breeding areas. The trend radiation index of each sub-breeding area reflecting the rationality of the breeding density in the multi-source breeding environment is obtained through the trend radiation map. A scale evaluation system for creating different preset breeding environment conditions for high-density Penaeus monodon farming is obtained. A farming density scale assessment model is constructed according to the scale evaluation system. The trend radiation index is evaluated to different degrees through the farming density scale assessment model, and a series of farming density scale assessment values ​​are generated for the trend radiation index caused by Penaeus monodon in each sub-farming area. Obtain the species information and phenotypic specifications of tiger prawns in each sub-culture area. Based on the species information and phenotypic specifications, retrieve the thermal color gamut reference table for different unit density culture of the current tiger prawn species in the big data network. Based on the thermal color gamut reference table, allocate the scale space rendering scale of each sub-culture area and construct the thermal space adjacency matrix of the target culture area. In the thermal adjacency matrix, a certain aquaculture density scale assessment value is weighted and aggregated with the aquaculture density scale assessment values ​​of its neighbors according to the scale spatial rendering scale. The above weighted aggregation steps are repeated until all aquaculture density scale assessment values ​​are rendered, and the aquaculture density scale heatmap of each sub-aquaculture area is obtained. The RGB chromaticity values ​​corresponding to the maximum breeding density scale of each sub-breeding area are extracted by the breeding density scale heat map and defined as the limit RGB chromaticity value. The current breeding density scale of each sub-breeding area is obtained and the corresponding RGB chromaticity value is queried and defined as the current RGB chromaticity value. If the current RGB chromaticity value has exceeded the limit RGB chromaticity value, then the sub-culture area is marked as a high-density abnormal culture area, and the multi-source culture environment analysis results are obtained.

4. The method for optimizing high-density culture of Penaeus monodon based on multi-source culture environment analysis according to claim 1, characterized in that, Step S106 specifically includes the following steps: If the multi-source aquaculture environment analysis results show that there is at least one or more high-density abnormal aquaculture areas, then a water environment model is introduced to calculate the high-density abnormal aquaculture areas and obtain the maximum carrying capacity of the water environment within the high-density abnormal aquaculture areas. Obtain the targeted aquaculture demand of tiger prawns, extract the expected aquaculture density scale of tiger prawns from the targeted aquaculture demand, and obtain a density scale-density coefficient-unit density scalar mapping table based on big data network; The desired aquaculture density scale is queried using the density scale-density coefficient-unit density scalar mapping table, and several potential density coefficients and the corresponding potential unit density scalars are obtained when the desired aquaculture density scale is reached. Using the desired aquaculture density coefficient as the objective function, several potential density coefficients as coefficients, and each potential unit density scalar as a constant variable, a simplex tableau model based on the penalty constraint function of maximum carrying capacity is constructed, which relies on the objective function row, coefficient matrix, and constant terms. The bottom row of the objective function row in the simplex tableau model is then examined to see one or more density coefficients recorded therein. If any density coefficient is greater than 0, the simplex method is introduced to select the unit density scalar of the constant variable corresponding to the maximum density coefficient from the fastest growth direction of the penalty constraint function to perform the minimum ratio objective function line test and solve, so as to approximate the maximum carrying capacity to calculate the optimal solution to achieve the desired breeding density scale, and output the input and output variables. Repeat the above steps of verifying the density coefficients and calculating the introduced and removed variables to continuously update the simplex tableau model until there is no density coefficient greater than 0, thus obtaining the terminated simplex tableau model. By terminating the simplex tableau model, a series of optimal aquaculture density control solutions are determined. These solutions are then used to optimize the current aquaculture density scale, thus obtaining the optimal aquaculture density scale for the high-density abnormal aquaculture area.

5. The method for optimizing high-density culture of Penaeus monodon based on multi-source culture environment analysis according to claim 1, characterized in that, Step S108 specifically includes the following steps: A hash misalignment algorithm is introduced. Based on the species information and phenotypic specifications of the tiger prawn, the scale misalignment hash function of the current farming density relative to the optimal farming density is quantitatively calculated in the hash misalignment algorithm. The scale misalignment hash function is used to determine the scale to be balanced for high-density abnormal farming. The preset aquaculture plan currently being implemented for high-density cultured tiger prawns is obtained. Based on big data, a tiger prawn aquaculture knowledge graph is obtained. The tiger prawn aquaculture knowledge graph is used to identify the correlation between the scale to be balanced and each aquaculture plan factor in the preset aquaculture plan, and the correlation degree of each aquaculture plan factor is obtained. If the correlation is greater than the preset correlation, the aquaculture planning factor is labeled as a related aquaculture planning factor; if the correlation is less than the preset correlation, the aquaculture planning factor is labeled as an unrelated aquaculture planning factor. The set of related aquaculture planning factors and the set of unrelated aquaculture planning factors that achieve the optimal aquaculture density scale for the scale to be balanced are obtained, and the factor ratio between the set of related aquaculture planning factors and the set of unrelated aquaculture planning factors is extracted. Based on big data, several empirical cases of high-density tiger prawn farming were obtained. Through these empirical cases, the parameter range of farming cases to achieve the scale to be balanced was extracted. According to the factor proportion, the parameter range of farming cases was divided into exploration mode and tracking mode for optimization parameters. The parameters of aquaculture cases with the constraint criterion of canceling the balance scale were explored locally and globally using exploratory mode and tracking mode respectively, resulting in the first-class and second-class exploratory patterns of aquaculture case parameters; The global optimal solution for the range of aquaculture case parameters is updated based on the first and second exploration patterns of the aquaculture case parameters. The optimal aquaculture optimization parameters for eliminating the unbalanced scale of different associated aquaculture planning factors are obtained. The high-density aquaculture of tiger prawns is optimized based on the optimal aquaculture optimization parameters.

6. The method for optimizing high-density culture of Penaeus monodon based on multi-source culture environment analysis according to claim 5, characterized in that, The process of exploring the parameters of aquaculture cases with the constraint of canceling the unbalanced scale using exploratory and tracking modes, respectively, to obtain the first-class and second-class exploration patterns of the aquaculture case parameters, specifically includes the following steps: For the exploration mode, the perturbation limit of the parameter range of the breeding case, different change dimensions, and the number of dimensions of each change dimension are preset. The unbalanced scale is removed as the constraint criterion. Based on the number of dimensions, the constraint criterion of each change dimension is explored by using each breeding case parameter in the parameter range of the breeding case to perform local slight perturbation exploration until the perturbation limit is reached, and multiple first new exploration sites are obtained. For the tracking mode, a safe speed is preset for the parameter range of the breeding case. According to the constraint criteria, the predetermined position and predetermined speed of each breeding case parameter in different changed dimensions are set within the parameter range of the breeding case. Based on the predetermined position and predetermined speed, the constraint criteria are explored globally from the current global position in each changed dimension, and the predetermined speed is always kept within the safe speed to obtain multiple second new exploration sites. The fitness of each first and second nascent exploration site is evaluated by constraint criteria. Only nascent exploration sites with fitness greater than the preset fitness are extracted and marked as the optimal nascent exploration sites. The optimal nascent exploration sites replace the current exploration position of the constraint criteria to generate the first and second exploration patterns of aquaculture case parameters.

7. A high-density culture optimization system for Penaeus monodon based on multi-source culture environment analysis, characterized in that, The high-density aquaculture optimization system for tiger prawns includes a memory and a processor. The memory stores a high-density aquaculture optimization method program for tiger prawns based on multi-source culture environment analysis. When the high-density aquaculture optimization method program for tiger prawns is executed by the processor, the steps of the high-density aquaculture optimization method for tiger prawns as described in any one of claims 1-6 are implemented.

Images